Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A computer visual-input system for the automatic recognition of blood cells Cossalter, John George 1970

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1970_A7 C68.pdf [ 6.61MB ]
Metadata
JSON: 831-1.0102199.json
JSON-LD: 831-1.0102199-ld.json
RDF/XML (Pretty): 831-1.0102199-rdf.xml
RDF/JSON: 831-1.0102199-rdf.json
Turtle: 831-1.0102199-turtle.txt
N-Triples: 831-1.0102199-rdf-ntriples.txt
Original Record: 831-1.0102199-source.json
Full Text
831-1.0102199-fulltext.txt
Citation
831-1.0102199.ris

Full Text

A COMPUTER V I S U A L - I N P U T SYSTEM FOR THE AUTOMATIC  RECOGNITION OF BLOOD C E L L S  by  JOHN GEORGE COSSALTER  B . A . S c , The U n i v e r s i t y  of British  C o l u m b i a , 1968  A T H E S I S SUBMITTED I N P A R T I A L F U L F I L M E N T OF THE REQUIREMENTS  FOR THE DEGREE OF  MASTER OF A P P L I E D S C I E N C E  in  t h e Department o f  Electrical  We a c c e p t  this  Engineering  t h e s i s as conforming  required  Research  to the  standard  Supervisor  ,  Members o f t h e C o m m i t t e e  Head o f t h e D e p a r t m e n t  Members o f t h e D e p a r t m e n t of E l e c t r i c a l  Engineering  THE U N I V E R S I T Y OF B R I T I S H A u g u s t , 1970  COLUMBIA  In  presenting  this  an a d v a n c e d d e g r e e the L i b r a r y I  further  for  of  this  written  at  agree  freely  that permission  for  It  financial  of  Columbia,  British  by  for  gain  Columbia  shall  the  requirements  reference copying  of  I agree and this  that  not  copying  or  for that  study. thesis  t h e Head o f my D e p a r t m e n t  is understood  of  The U n i v e r s i t y o f B r i t i s h V a n c o u v e r 8, Canada  of  for extensive  permission.  Department  fulfilment  available  p u r p o s e s may be g r a n t e d  representatives. thesis  in p a r t i a l  the U n i v e r s i t y  s h a l l make i t  scholarly  by h i s  thesis  or  publication  be a l l o w e d w i t h o u t  my  ABSTRACT  A computer v i s u a l - i n p u t system was the c l a s s i f i c a t i o n of l e u k o c y t e s .  b u i l t f o r the purpose of  studying  I t c o n s i s t e d of an image d i s s e c t o r camera  i n t e r f a c e d d i r e c t l y to a D.E.C. PDP-9 computer; a d i s p l a y of the image was  also provided, using a monitoring The  field  scope.  d e s i g n and hardware arrangement of the system i s b r i e f l y  c r i b e d , w h i l e d e t a i l e d diagrams of the l o g i c  des-  networks are shown i n Appendix  II. Photomicrographs of n e u t r o p h i l s were used as a p a t t e r n s e t , i n a study of the computer c l a s s i f i c a t i o n of c e l l age of  f e a t u r e v e c t o r s was  noted  processed  Clustering  segmented c e l l s can be d i s t i n g u i s h e d .  of the n e u t r o p h i l n u c l e i was  the shape of a n u c l e u s  A square  performed and an a r e a o p e r a t o r  into a curvature function.  c u r v a t u r e f u n c t i o n , a measure of l o b u l a r i t y , •perimeter  lobularity.  i n a two-dimensional measurement space showing  t h a t metamyelocyte, banded and contour-trace  and  pre-  Peaks i n t h i s  as w e l l as the r a t i o of  the  to square r o o t of n u c l e a r area, a measure o f the i r r e g u l a r i t y i n the  n u c l e a r boundary, were used as o r i e n t a t i o n and The  a r e a o p e r a t o r was  from l e u k o c y t e images. filamentation,  features.  to be u n s u i t a b l e f o r e x t r a c t i n g c u r v a t u r e  I n cases o f extreme n u c l e a r c u r v a t u r e and  nuclear  the b a s i c f o r m u l a t i o n s of the o p e r a t o r were v i o l a t e d g i v i n g an  erroneous measure of The  found  size-independent  curvature.  g e n e r a l form o f the f r e q u e n c y  the image d i s s e c t o r camera was  derived.  the camera r e s o l u t i o n were found  The  spectrum of the v i d e o s i g n a l from s i g n a l bandwidth requirements  experimentally.  ii  and  TABLE OF CONTENTS Page L i s t of I l l u s t r a t i o n s  iv  Acknowledgement  vi  1.  2.  3.  Introduction  1  1.1  The D i f f e r e n t i a l Leukocyte Count  1  1.2  The Leukocyte Smear  3  1.3  The C l a s s i f i c a t i o n o f L e u k o c y t e s  4  The Computer V i s u a l - I n p u t  7  System  2.1  Methods o f P i c t u r e P r o c e s s i n g  7  2.2.  The D e s i g n o f the System  8  2.3  The D i s p l a y M o n i t o r  2.4  The Computer I n t e r f a c e  . . . . .  12 13  •  The System Software  15  3.1  T e s t Programs  15  3.2  The S i m u l a t i o n o f the A r e a Operator, and Ridge F u n c t i o n  3.3  Problems E n c o u n t e r e d i n Computing  . . . .  16 19  the Ridge F u n c t i o n  4.  A N e u t r o p h i l C l a s s i f i c a t i o n Study  23  5.  The System Parameters  29  6.  .  5.1  N o i s e C h a r a c t e r i s t i c s o f the Image D i s s e c t o r Tube  29  5.2  The S p a t i a l Frequency Spectrum  31  5.3  The  34  5.4  Bandwidth  Spatial Resolution  36  Requirements  Conclusions  • • •  38  APPENDIX I  40  APPENDIX I I  42  APPENDIX I I I  52  BIBLIOGRAPHY  53  iii  LIST OF ILLUSTRATIONS Figure  Page  1.1  A differential  1.2  Photomicrograph o f a n e u t r o p h i l  (2000 X ) 1.3  leukocyte  2  count e x a m i n a t i o n form surrounded by e r y t h r o c y t e s  4  :  The a r e a o p e r a t o r  p o s i t i o n e d a t the p o i n t  s  1  on the c u r v e  5  C 1.4  A t e s t p a t t e r n and i t s r i d g e f u n c t i o n  2.1  B l o c k diagram o f the v i s u a l - i n p u t system  10  2.2  A p i c t o r i a l diagram o f the t r a n s d u c e r  11  2.3  I n s t r u c t i o n s e x e c u t a b l e on the v i s u a l - i n p u t d e v i c e  3.1  The g r e y - l e v e l c r o s s - s e c t i o n o f a n e u t r o p h i l  16  3.2  The c o n t o u r - p a t h and r i d g e f u n c t i o n o f a n e u t r o p h i l n u c l e u s  19  3.3  C o n t o u r - t r a c e s showing the e f f e c t s o f i n c r e a s i n g the g r i d spacing  3.4  . . . . . . . .  6  •  . . . .  20  .  T y p i c a l e r r o r s made when o p e r a t i n g  on a c t u a l  neutrophil 22  nuclei 3.5  14  Ridge f u n c t i o n s d e r i v e d u s i n g  d i f f e r e n t area  operator 23  sizes 4.1  The g e n e r a l  4.2  Typical ridge functions  4.3  A measurement-space graph showing the c l u s t e r i n g o f  nuclear  shapes o f n e u t r o p h i l s  24 25  n e u t r o p h i l age groups  27  5.1  A model o f the d e g r a d a t i o n o f an image  34  5.2  A scan a c r o s s  36  the edge o f a r a z o r b l a d e  iv  Figure  Page  A2.1  Device c o n t r o l  43  A2.2  Mode c o n t r o l  44  A2.3  Data t r a n s f e r c o n t r o l  45  A2.4  A/D  46  A2.5  8 bit X  A2.6  X  (Y)  load pulse c o n t r o l  A2.7  X  (Y)  attenuator  A2.8  X  (Y) D/A  A2.9  X  (Y)  A3.1  Schematic o f the  converter (Y)  r i p p l e counter  47  . .  48 49  converter  . . .  50  data gates  51 logarithmic  v  amplifier  52  ACKNOWLEDGEMENT  I  would l i k e  for their financial  67-3350  grant  in  t o g r a t e f u l l y acknowledge the N a t i o n a l Research  support  through a s t u d e n t s h i p  expert  and the r e s e a r c h  t o my s u p e r v i s o r , Dr. J . S. MacDonald,  and guidance, and to Dr. R. ¥. Donaldson and Dr. R. H. Pearce  f o r t h e i r reading I  1968-69  1969-70.  Acknowledgement i s a l s o g i v e n f o r h i s support  in  and c o n s t r u c t i v e c r i t i c i s m o f the t h e s i s .  would a l s o l i k e  t o thank Mr. A. Leugner f o r h i s e f f i c i e n t and  c o n s t r u c t i o n o f the hardware, and Mr. B. J . Twaites and Mr. H. H. B l a c k  f o r i n v a l u a b l e a s s i s t a n c e i n o b t a i n i n g photographs and s l i d e s . due for  Council  Thanks a r e a l s o  to Heather DuBois f o r t y p i n g the m a n u s c r i p t and my f r i e n d s and c o l l e a g u e s proofreading.  vi  1  1.  1.1 The D i f f e r e n t i a l Leukocyte The important  differential  INTRODUCTION  Count  l e u k o c y t e (white b l o o d c e l l ) count i s a r o u t i n e and  d i a g n o s t i c t e s t used  i n c l i n i c a l medicine.  c o n s i s t s o f t a b u l a t i n g the r e l a t i v e f r e q u e n c y  The p r o c e s s , which  o f the d i f f e r e n t  types o f  l e u k o c y t e s , i s i n v a l u a b l e i n i n d i c a t i n g a l a r g e v a r i e t y o f human d i s o r d e r s , such as a c u t e i n f e c t i o n s ,  i n t o x i c a t i o n s o r drug r e a c t i o n s , m a l i g n a n c i e s and  s e v e r e hemorrhages. A sample o f p e r i p h e r a l b l o o d i s taken,  smeared on a g l a s s s l i d e and  a l l o w e d t o d r y . The s l i d e i s then commonly t r e a t e d w i t h Wright's c h e m i c a l l y dye the c e l l s a c c o r d i n g to t h e i r i o n i c a f f i n i t y , under a m i c r o s c o p e .  s t a i n , to  and examined  U s u a l l y , one hundred c e l l s a r e s c r u t i n i z e d and c l a s s e d  i n t o s t a n d a r d c a t e g o r i e s , each c a t e g o r y h a v i n g s i m i l a r m o r p h o l o g i c a l p r o p e r t i e s which d e f i n e i t s t o p o l o g y and b i o l o g i c a l s t r u c t u r e . shape, the presence used  of a n u c l e o l i ,  to c a l c u l a t e a percentage  form, used  i n a clinical The  s i z e , - t e x t u r e and c o l o r a r e t y p i c a l p r o p e r t i e s  list  of c e l l  types.  ( F i g u r e 1.1 shows a t y p i c a l  l a b o r a t o r y , to r e p o r t the " n o r m a l i t y " o f a b l o o d  f o l l o w i n g f a c t s suggest  to some degree,  N u c l e a r and c y t o p l a s m i c  that the d i f f e r e n t i a l  leukocyte  smear).  count,  s h o u l d be automated:  a) A s i d e from b e i n g a t e d i o u s t a s k , the p r o c e s s i s an expensive clinical  test,  c o n s i d e r i n g the c o s t o f the t e c h n i c i a n ' s time to  prepare and examine a b l o o d smear. b) A s i g n i f i c a n t clinical  p o r t i o n , about 10$, o f the smears p r o c e s s e d i n a  l a b o r a t o r y a r e abnormal and a r e brought  of a h e m a t o l o g i s t examination.  or experienced s p e c i a l i s t  to the a t t e n t i o n  f o r a more d e t a i l e d  2  UNIVERSITY O F TENNESSEE  HEMATOLOGY LABORATORY  MEMPHIS, TENNESSEE E X A M I N A T I O N OE P E R I P H E R A L IiLOOD  Nnme  Age  . .Date  M  F  W  C  Referred by  Red Blood Cells:  Thrombocytes: White Blood Cells: Number counted Per Cent  Norma  0-1 2-10 50-70 1-4 0-1  20-40 1-6  Stem Cell Myeloblast Progranulocyte N . Myelocyte N . Metamyelocyte N . Hand N . Segmented Eosinophil Basophil Lymphoblast Prolymphocyte Lymphocyte Monocyte Plasmocyte Atypical cell Disintegrated cells per 100 intact VVBC  Summary of Abnormalities:  Interpretation:  Examined by,  F i g u r e 1.1  A differential  l e u k o c y t e count e x a m i n a t i o n form.  ...  3  c) There w i l l clinical  be  a s t r o n g need i n the  screening  l a b o r a t o r y w i t h i n w h i c h an  system i s the d i f f e r e n t i a l d)  logical  blood  s u c h as  computer [l]-[7], fication  1.2  The  count i s not  (small cytoplasm .The  a remotely  Leukocyte  process  can  be  five  a)  Lymphocytes  b)  Monocytes  c)  Basophils  d)  Eosinophils  e)  Neutrophils  cells)  i n v o l v e d i n the expensive  automation  and  sequence  peripheral  complex software  algorithms  for cell  of  devices,  [ l ] w i t h a means f o r i m a g e i n p u t  to  a  classi-  [9].  i s composed o f  d i s t i n c t morphological  f r o m 5000 t o 9000 l e u k o c y t e s  f r o m 5 t o 50^u ( m i c r o n s )  i n size.  r a n g e f r o m 10  a f a c t o r o f 600:1  methods of  three basic c e l l s -  t h r o m b o c y t e s s u s p e n d e d i n a medium o f p l a s m a .  divided into  thesis,  morpho-  Smear  There are  this  hand  the b a s i s of t h e i r  (red blood  r e q u i r e s c o m p l e x and  Human p e r i p h e r a l b l o o d e r y t h r o c y t e s and  their  t h e means a r e a t  to simple  fragments which are  a computer and [ 5 ] , [8],  adaptable  erythrocytes  c o n t r o l l e d microscope  [ 2 ] , [4],  sub-  characteristics.  differential  clotting).  important  c o m p u t e r s and  q u a n t i t a t i v e l y , on  enumeration of leukocytes,  thrombocytes  automated  count.  i n pattern classification,  to c l a s s l e u k o c y t e s  as i s the  leukocyte  With the advent o f l a r g e - s c a l e d i g i t a l i n c r e a s e d use  The  future for a f u l l y  and  t o 15jU.  h a v e an  leukocytes  types:  per  cubic millimeter,  N e u t r o p h i l s , which are The  The  leukocytes,  erythrocytes  average s i z e of about  the  focal  ranging  point  of  outnumber l e u k o c y t e s Su.  by  4  A t y p i c a l leukocyte  as i t appears on a smear, shown i n F i g u r e  i s not s p h e r i c a l as found i n c i r c u l a t i n g blood, nucleus  of the c e l l may  touched, or o v e r l a p p e d  F i g u r e 1.2  1.3  The  the cytoplasm  is  The  probably  erythrocytes.  Photomicrograph of a n e u t r o p h i l surrounded by e r y t h r o c y t e s (2000x).  C l a s s i f i c a t i o n o f Leukocytes Looking  a t F i g u r e 1.2,  c l a s s i f y leukocytes in  but f l a t t e n e d i n t o a d i s c .  be f o l d e d upon i t s e l f and by,  1.2,  becomes apparent.  the image f i e l d .  dense n u c l e u s  cytoplasmic  and  the l e u k o c y t e  must be  Bourk [ l O ] has  s o l v e d t h i s problem  to  found optically adequately.  i s l o c a t e d , q u a n t i t a t i v e measurements must be made to  c l a s s i f y i t morphologically. o f the n u c l e u s  First,  T h i s i s not a g r e a t problem s i n c e l o c a t i n g the  l o c a t e s the c e l l .  Once the l e u k o c y t e  the problem o f f i n d i n g good a l g o r i t h m s  touching  boundary.  (The  F o l d i n g may  obscure the t r u e t o p o l o g i c a l n a t u r e  e r y t h r o c y t e s may software  complicate  the d e t e c t i o n of  problem o f the d e t e c t i o n of  the  overlapping  5  c e l l s has of  been s o l v e d by R i n t o l a and Hsu  [ l l ] ) . • Clearly,  l e u k o c y t e s must be made u s i n g f e a t u r e s which are  such as, g e o m e t r i c  classification  orientation-independent,  n u c l e a r shape.  T h i s t h e s i s i s p r i m a r i l y concerned n u c l e i of n e u t r o p h i l s .  The  w i t h the form o r shape of  n u c l e i are c o n t o u r - t r a c e d .  based on the t h e o r y f o r m u l a t e d contour  the  An  "area  the  operator",  by Connor [ l 2 ] , c o n v e r t s the shape of  the  into a "ridge function". Consider  the curve C, i n F i g u r e 1.3,  to be the boundary between a  l i g h t and a dark r e g i o n .  The  a r e a o p e r a t o r i s a d i s c of .area A^,  C.  A r i d g e f u n c t i o n V^(s),  moves i n one  h a v i n g i t s c e n t e r always on the  curve  a f u n c t i o n of a r c l e n g t h , i s formed as the  d i r e c t i o n around the curve C.  The  operator  curve C must be c o n t i n u o u s  with-  i  i n the o p e r a t o r d i s c and  a t each p o i n t s  a value  I  V r  1  |s  = A  , the a r e a o p e r a t o r i s d e f i n e d to have  - A d  H  ' |s  (l.l) v  I A  I  = the v a l u e  1  o f the r i d g e f u n c t i o n a t s .  = the a r e a o f the o p e r a t o r which l i e s i n  1  the l i g h t r e g i o n , e v a l u a t e d  i  at s .  From the d e f i n i t i o n o f V ( s ) , r  d  A  = — d ^ —  when C i s a s t r a i g h t l i n e .  (l.2)  A  O^V  r  d 7~<V  2  A  2  r  <A, d  when C i s concave w i t h  r e s p e c t to the (l«3)  dark r e g i o n .  when C i s convex w i t h r e s p e c t t o the (l»4)  dark r e g i o n . Intuitively,  i t i s seen t h a t V ( s ) w i l l g i v e some measure o f the shape o f t h e  curve.  I n f a c t , V ( s ) i s a n o n - l i n e a r f u n c t i o n o f the c u r v a t u r e , K ( s ) , o f C; r V ( s ) becomes p r o p o r t i o n a l t o K ( s ) i n the l i m i t o f s m a l l K. An important f a c t , r proven i n Connor's t h e s i s , w h i c h has become u s e f u l i n f e a t u r e e x t r a c t i o n , i s t h a t t h e extrema o f V ( s ) c o i n c i d e w i t h having  the extrema o f K ( s ) , both f u n c t i o n s  t h e same o r i g i n and d i r e c t i o n a l o n g C.  minimum c u r v a t u r e  Thus, p o i n t s o f maximum and  c a n be determined u s i n g t h e a r e a  operator.  F i g u r e 1.4 shows the r i d g e f u n c t i o n o f a two-lobed  two-lobed p a t t e r n  F i g u r e 1.4  figure.  ridge function  A t e s t p a t t e r n and i t s r i d g e f u n c t i o n .  7  This t h e s i s presents  an e x p l a n a t i o n  of p e r i p h e r a l hardware and  o p e r a t i o n of the computer v i s u a l - i n p u t system which w i l l study  o f p a t t e r n c l a s s i f i c a t i o n of An  the  be used f o r the  leukocytes.  i n v e s t i g a t i o n o f the f e a s i b i l i t y o f u s i n g the a r e a o p e r a t o r  preprocessing  technique  o f the system.  f o r f u t u r e work, was  used as a t e s t of the  A r e a l p a t t e r n s e t of l e u k o c y t e  c l a s s i f i c a t i o n study.  The  t r a n s p a r e n c i e s was  The  used i n a  s t u d y i n d i c a t e s t h a t metamyelocyte, banded  f e a t u r e and  shape of n u c l e i was  s h o u l d be  f o u n d to be  f u r t h e r explored.-  parameters i s made and  An  an important  a  performance  segmented n e u t r o p h i l s can be w e l l d e f i n e d i n measurement space, and classified.  as  and  thus  descriptive  e v a l u a t i o n of the system's p h y s i c a l  the advantages, of u s i n g a l a r g e r a p e r t u r e  image d i s s e c t o r  tube a r e o u t l i n e d i n Appendix I.  2.  2.1  THE  Methods of P i c t u r e  COMPUTER VISUAL-INPUT SYSTEM  Processing  P i c t u r e p r o c e s s i n g by a computer e n t a i l s d i g i t i z i n g a p i c t u r e frame i n t o an a r r a y of m by n g r i d p o i n t s . by some s o r t o f s c a n n i n g quantized,  instrument;  d i g i t a l l y coded and  Each of the mxn  points i s interrogated  the o p t i c a l d e n s i t y v a l u e  r e a d i n t o the computer.  The  p i c t u r e data r e q u i r e s t h a t the computer have random a c c e s s  p r o c e s s i n g of to any g r i d  s i n c e i t i s assumed t h a t t h e r e i s no a p r i o r i knowledge of the contained  this  point,  information  i n the p i c t u r e . The  f a s t e s t and most c o s t l y method of p r o c e s s i n g a p i c t u r e i s to  s y s t e m a t i c a l l y read  the o p t i c a l d e n s i t i e s of the mxn  h i g h - s p e e d memory of a computer, where Ledley  of a point i s  [2] u s e d a f l y i n g spot  p o i n t s d i r e c t l y i n t o the.  each p o i n t can be randomly  scanner to r e a d a r a s t e r of 700  by 500  accessed. points  through  a data channel  into  a n I B M 3 6 0 / 4 4 s y s t e m ' s c o r e memory.  m o n o c y t e c l a s s i f i c a t i o n was d o n e t o t a l l y w i t h i n o n l y b y t h e c o r e memory's c y c l e If  a memory i s s l o w e r  memory, t h e d a t a c a n be s y s t e m a t i c a l l y r e a d o n t o Although  an  data a c q u i s i t i o n from  such  t h a n t h a t d i s c u s s e d above, t h i s method h a s t h e advantage  t h e p i c t u r e may b e . p r o c e s s e d  at a later  instrument.  Mendelsohn e t a l [ 8 ] processed  spot scanner  was o p e r a t e d  directly  i n f o r m a t i o n onto magnetic tape. l e u k o c y t e s was t h e n d o n e f r o m  is  limited  time.  e x t e r n a l memory s u c h a s m a g n e t i c t a p e .  The  the computer a t speeds  t h e i n f o r m a t i o n c o n t a i n e d i n t h e mxn p o i n t s i s t o o l a r g e f o r t h e  computer's high-speed  that  H i s work on  through  date,  independent  of the scanning  a c t u a l l e u k o c y t e images. a microscope,  writing  The d i s c r i m i n a t i o n o f t h e f i v e  types of 800 computer.  s l o w e s t b u t p o t e n t i a l l y l e a s t c o s t l y method o f p i c t u r e o n - l i n e w i t h a computer.  processing  This type o f  e m p l o y s t h e a c t u a l p i c t u r e a s a memory, u s i n g t h e s c a n n i n g i n s t r u m e n t random-access t r a n s d u c e r . scanner  interfaced  Rosenberg's system  t o a PDP-9 c o m p u t e r .  PDP-9 w e r e i n t e r r o g a t e d o n e a t a t i m e ,  2.2 The D e s i g n  employed a f l y i n g  Picture grid  system  as a  spot  p o i n t s s p e c i f i e d by t h e  on-line.  o f the System  T h e r e a r e two f a c t o r s w h i c h a memory  [5]  flying  picture  the magnetic tape, u s i n g a Honeywell  to use a scanning instrument  A  make d i r e c t r e a d - i n o f p i c t u r e  data to  desirable: a ) F u r t h e r p r o c e s s i n g o f t h e p i c t u r e may be done a t a h i g h not h a v i n g t o communicate w i t h t h e slow apparatus,  peripheral picture  scanning  and  b) S u b s e q u e n t i n t e r r o g a t i o n s o f t h e same g r i d ducer  speed,  may g i v e d i f f e r e n t  optical  point through  density values.  a  trans-  Position  jitter  . 9  and  i l l u m i n a t i o n f l u x v a r i a t i o n s are inherent  to p i c t u r e  scanning  i n s t r u m e n t s and thus the same g r i d p o i n t may have a v a r i a b l e o p t i c a l density value.  Once the o p t i c a l d e n s i t y o f a p o i n t has  been w r i t t e n i n t o a memory, though, f u r t h e r i n t e r r o g a t i o n s o f that point w i l l  be the same.  A d i r e c t p i c t u r e - d a t a r e a d - i n i s d e s i r a b l e , but i n most cases a l a r g e h i g h - s p e e d memory u n i t i s t o o e x p e n s i v e f o r t h e average h o s p i t a l laboratory.  The system to be d e s c r i b e d  i n t h i s t h e s i s has a f a s t e r random  a c c e s s time t h a n e i t h e r a d i s c o r drum memory, and the o p e r a t o r  clinical  on-line processing  allows  t o observe, on a m o n i t o r i n g scope, i f t h e system i s f u n c t i o n i n g  properly. . The  u n i q u e p r o p e r t i e s o f the image d i s s e c t o r camera make i t a d a p t a b l e  f o r i n t e r f a c i n g with a small  computer such as t h e PDP-9, u s i n g  as a random-access " r e a d - o n l y " memory. image s t o r a g e  This  type o f v i d e o  Figure  a f l e x i b l e means o f p r o c e s s i n g  on a m o n i t o r i n g scope.  The system i s s e t up t o  the information  and s i m u l t a n e o u s l y  on t r a n s p a r e n c i e s ,  viewing the f i e l d  using  o f scan  The d i s p l a y makes i t p o s s i b l e t o m o n i t o r and d e t e c t  e r r o r s and i n i t i a l l y  The  the FDP-9 t o  diagram o f t h e computer v i s u a l - i n p u t system  c l a s s i f i c a t i o n studies.  the camera as the s c a n n i n g d e v i c e  can  the scan c a n be stopped  randomly.  2.1 shows a b l o c k  to be u s e d f o r l e u k o c y t e  processing  camera r e q u i r e s no  The d i r e c t i o n o f scan i s a l s o v a r i a b l e a l l o w i n g  i n t e r r o g a t e any g r i d p o i n t  provide  picture  and thus c a n be o p e r a t e d a t a v a r i a b l e scan r a t e w i t h o u t a  change i n t h e s i g n a l c u r r e n t a m p l i t u d e ; i n f a c t , completely.  the a c t u a l  a l i g n t h e scan f i e l d  of i n t e r e s t .  system hardware which i s b u i l t up m a i n l y from D.E.C. modules,  be d i v i d e d i n t o two main s e c t i o n s : a)  the i n t e r f a c i n g hardware which p r o v i d e s  c o n t r o l commands and d a t a  8 BIT  X RIPPLE  S  X f  X  LOAD Y PULSE  COUNTER COUNTER OUTPUT  DATA  GATES  DATA  LINES  PWR  CLR.  I?  I/O  TOGGLE)  X  LOAD  PULSE CONTROL  D/A X D/A OUTPUT.  ATTENUATOR  IOP.  &  MODE  SUBDEVICE SELECTION  x  CONVERTER  OUTPUT  DEVICE  5  3  CLOCK (X  LOAD X DATA  2-CLOCK  A V  LOAD LOAD CLEAR  DEVICE CONTROL  STOP.  POP-9  D/A  CONTROL  INPUT  CAMERA  DISPLAY  CONTROL  X  CONVERSION FLAG  DATA  READ  LINES  A/D  DISPLAY  DISPLAY  1  LOW-PASS  INPUT  FILTER  UNIT  18  LOW-PASS FILTER  READ  TRANSFER  LINES  CONTROL  J  LOG.  CAMERA  ™AMP.  10 DATA  LINES  VIDEO  SIGNAL  RO.  CAMERA  AA>  DISPLAY DEFLECTION  Y  CONVERSION COMPLETE  A/D  Y A/D DIGITAL ~J^D  OUTPUT  Y  DATA  GATES  1 Y COUNTER  FIGURE 2.1  TOGGLE  BLOCK DIAGRAM OF THE  8 BIT  Y  DEFLECTION  Y D/A OUTPUT ,  CONVERTER  IMAGE DISSECTOR  DATA  DATA  X  DEFLECTION  DEFLECTION  SYSTEM -TFM  ENABLE  A V  STRT.  X Y  START A/D CLEAR A/D  SKIPRQ.  CLOCK  ATTENUATOR  D/A  CONVERTER  S DATA  LINES  0  Y  LOAD DATA  Y LOAD PULSE CONTROL  Y COUNTER OUTPUT  Y RIPPLE  VISUAL-INPUT  COUNTER  SYSTEM.  O  11  t r a n s f e r between the camera and  the PDP-9,  and  b) the m o n i t o r i n g equipment which p r o v i d e s a t e l e v i s i o n - t y p e image of the f i e l d  of  scan.  Photomicrographs of l e u k o c y t e s to be examined are made i n t o 2" x transparent s l i d e s . by a d.c.  2"  Each s l i d e i s p l a c e d i n a s l i d e c o p i e r and i l l u m i n a t e d  lamp; an o p t i c a l image of the s l i d e i s f o c u s e d onto the p h o t o s e n s i t i v e  cathode of the  camera.  (See F i g u r e  2.2)  image d i s s e c t o r tube slide lamp  lens  photomultiplier  photocathode  A  F i g u r e 2.2  The  • aperture plane  A p i c t o r i a l diagram of the  photocathode emits  i n c i d e n t luminous f l u x . the anode.  magnetic focusing  video amplifier  transducer.  e l e c t r o n s w i t h a d e n s i t y p r o p o r t i o n a l to  An a c c e l e r a t i n g p o t e n t i a l a t t r a c t s these e l e c t r o n s to  F o c u s i n g i s accomplished,  u s i n g a l a r g e a x i a l magnetic f i e l d .  r e s u l t i s an e l e c t r o n i c image of the o p t i c a l image, f o c u s e d onto the plane.  the  S i n c e the m a g n i f i c a t i o n of the image i n the d r i f t  The  aperture  s e c t i o n of the image  d i s s e c t o r tube i s u n i t y , an a p e r t u r e plane v a r i a t i o n i n cathode c u r r e n t , I , i s g i v e n by the  equation  where S  = the photocathode  sensitivity.  F = the i n c i d e n t luminous f l u x . A s m a l l a p e r t u r e a t the anode plane  (a 1 m i l ( l O  inches)  diameter  c i r c u l a r a p e r t u r e i n t h i s c a s e ) a l l o w s a p o r t i o n of the image c u r r e n t to impinge on the f i r s t dynode of a p h o t o m u l t i p l i e r . a m p l i f i e d by a f a c t o r o f about 10^  and  fed i n t o a video a m p l i f i e r  the v i d e o s i g n a l v o l t a g e .  I t i s obvious,  literally  aperture.  d i s s e c t e d by the The  t h i s way,  then,  so t h a t any  any  the image and  p a r t of the image can be i n t e r r o g a t e d and  aperture are r e l a t i v e  c o n s i d e r a moving a p e r t u r e ,  2.3  The  Display  to one  another,  the  aperture  the r e s u l t i n g  signal  S i n c e the motion  i t i s convenient  to  Monitor  v a r i a b l e frame r a t e ,  to be p r o c e s s e d ,  t e l e v i s i o n - t y p e scan i s used.  image d i s s e c t o r camera i s low-pass f i l t e r e d  The  f i l t e r bandwidth of 20 kHz  a 256  by 256  spot,  v i d e o s i g n a l from  the  to i n c r e a s e the s i g n a l - t o - n o i s e  r a t i o and f e d to the i n t e n s i t y i n p u t of a T e k t r o n i x 602  display unit.  ( k i l o h e r t z ) the frame r a t e was  For a  a frame per  second.  l o n g - p e r s i s t e n c e phosphor of the scope p r o v i d e d f o r a good d i s p l a y ) . .  d e f l e c t i o n of the camera and bit  a horizontal  s c a n n i n g a s t a t i o n a r y image.  To p r o v i d e a d i s p l a y of the f i e l d  (The  and  p a r t of the image can be made to o v e r l i e  c u r r e n t i s p r o p o r t i o n a l to the t r a n s m i t t a n c e of the s l i d e . of  to produce  t h a t the e l e c t r o n i c image i s  whole e l e c t r o n i c image i s d e f l e c t e d by a v e r t i c a l  magnetic f i e l d in  T h i s c u r r e n t i s then  r i p p l e counter.  deflection control;  The  the scope are s i m u l t a n e o u s l y c o n t r o l l e d by a  16  f i r s t 8 b i t s of the c o u n t e r are u s e d as the h o r i z o n t a l  the l a t t e r 8 b i t s ,  c l o c k t o g g l e s the r i p p l e c o u n t e r ,  the v e r t i c a l c o n t r o l .  incrementing  A v a r i a b l e speed  the e x i s t i n g b i n a r y number;  the o v e r f l o w from the h o r i z o n t a l p a r t of the c o u n t e r i n c r e m e n t s part.  The  The r e s p e c t i v e b i n a r y numbers are g a t e d  the  vertical  to the 8 most s i g n i f i c a n t  bits  o f two 10 b i t D / A c o n v e r t e r s ,  which produce the d e f l e c t i o n v o l t a g e s .  The  frame r a t e o f the d i s p l a y c a n be c o n t r o l l e d by a d j u s t i n g the c l o c k r e p e t i t i o n rate.  (See F i g u r e  2.1 and t h e F i g u r e s  i n Appendix I I f o r d e t a i l s ) .  Only 256 i n c r e m e n t s o f d e f l e c t i o n a l o n g axes a r e r e q u i r e d unit.  to s t a y w i t h i n  the h o r i z o n t a l and v e r t i c a l  the p r a c t i c a l r e s o l u t i o n l i m i t  U s i n g o n l y 2"^ g r i d p o i n t s f o r d i s p l a y a l s o a l l o w s  which i s s u b j e c t i v e l y a d e s i r a b l e f e a t u r e f o r The  o f the d i s p l a y  a f a s t e r frame r a t e ,  monitoring.  a c t u a l s c a n a r e a on t h e photocathode i s a 0.8 i n c h square.  are 1024 by 1024 fundamental g r i d p o i n t s w i t h i n  t h a t a r e a and the d i s p l a y r a s t e r  merely uses every f o u r t h point i n every f o u r t h g r i d point  2.4 The Computer  There  line.  Interface  W i t h the i n t e r f a c e p r o v i d e d ,  the PDP-9 c o n t r o l s the mode o f o p e r a t i o n :  a) I n the d i s p l a y mode, the e x t e r n a l hardware i s p e r m i t t e d the f i e l d  to d i s p l a y  o f scan.  b) I n the computer-read mode, the PDP-9 d e f l e c t s the camera to the g r i d coordinates  i t s p e c i f i e s and r e a d s the o p t i c a l d e n s i t y o f  t h a t p o i n t on the t r a n s p a r e n c y image. A l t h o u g h the PDP-9 i / o bus c o n s i s t s o f 36 l i n e s ,  only  those  lines  r e q u i r e d f o r c o n t r o l o r d a t a t r a n s f e r were i n t e r f a c e d to the e x t e r n a l input device. all  The 18 b i d i r e c t i o n a l d a t a l i n e s were used f o r d a t a t r a n s f e r ;  18 l i n e s were w i r e d from the PDP-9 t o the d e v i c e , f o r f l e x i b i l i t y , a l t h o u g h  o n l y 10 a r e p r e s e n t l y i n a c t i v e use. sent via  t o the d e v i c e  device  (10-bit d e f l e c t i o n coordinate  and 6 - b i t o p t i c a l d e n s i t y codes a r e r e t u r n e d  the i / o bus d a t a The  two  visual-  codes a r e  t o the PDP-9  lines).  t h r e e IOP l i n e s s e r v e d  to transmit  and were p r o p e r l y g a t e d f o r e x e c u t i o n  SUBDEVICE SELECTION l i n e s ,  i n s t r u c t i o n s t o the e x t e r n a l  by the s i x DEVICE SELECTION and  i / o SKIP and READ REQUEST i n t e r r u p t e d the  14 computer f o r asynchronous r e a d i n g o f o p t i c a l d e n s i t y codes. The  t a b l e i n F i g u r e 2.3  o p e r a t i o n o f the s c a n n i n g  device.  shows the i n s t r u c t i o n codes which c o n t r o l the They form p a r t o f the e s s e n t i a l software •  f o r the system.  Code Word i n O c t a l  Operation  Performed  707001  s t o p the c l o c k and d i s a b l e the h o r i z o n t a l c o u n t e r g a t e s  707002  l o a d the  i / o data l i n e s i n t o  the h o r i z o n t a l D/A  converter  707004  not used  707021  s t o p the c l o c k and d i s a b l e the v e r t i c a l c o u n t e r g a t e s  707022  ,  l o a d the  i / o data l i n e s i n t o  the v e r t i c a l D/A  converter  707024  not u s e d  707041  s t a r t conversion of  707042  start  707044  c l e a r the system  707061  check i f A/D c o n v e r s i o n i s complete ( i / O s k i p )  707062 707064  The  c l e a r the A / D c o n v e r s i o n comp l e t e f l a g and r e s e t the A/D converter I n s t r u c t i o n s executable  full  the e x t e r n a l c l o c k  r e a d the A/D code onto the i/O d a t a l i n e s  .  F i g u r e 2.3  A/D  on the v i s u a l - i n p u t d e v i c e .  10 b i t c a p a b i l i t y o f the D/A  the d i s p l a y equipment, i s u s e d f o r d e c o d i n g  c o n v e r t e r s , a l s o employed by  the d e f l e c t i o n c o o r d i n a t e s  sent  20 by the  PDP-9.  i n the scan  The computer i s then a b l e t o i n t e r r o g a t e any o f the  area.  2  points  A  successive approximation A / D  c o n v e r t e r i s used  to t r a n s f o r m the  c o n t i n u o u s v i d e o s i g n a l v o l t a g e i n t o a 6 b i t b i n a r y number r a n g i n g from 0  F o r c l e a r e r comprehension a typical  data-fetch cycle are outlined.  r e a d s the o p t i c a l  READ  / E N T E R  707001 LAC  X I  LAC  Y I  / S T O P  707002  C L O C K  the P D P - 9 .  (x.,y.) i n t o  SUBROUTINE T H E  mode, the s t e p s o f  The p a r t i a l program shown below  d e n s i t y o f the g r i d p o i n t  0  TO  &  I N T E R R O G A T E  INTERRUPT  T H E  / T R A N S F E R  X  COORDINATE  TO  X  D / A  707022  / T R A N S F E R  Y  COORDINATE  TO  Y  D / A  JMS  / W A I T  F I L T E R  D E L A Y  JMP  F O R  PROGRAMMED  707041 707061  D E L A Y  o f the computer-read  / S T A R T / I S  A / D  PDP-9  / N O , A S K  .-1  S I G N A L  DELAY  -  A  P O I N T  D I S P L A Y  GO  TO  A  D E L A Y C O N V E R S I O N  READY  TO  A C C E P T  I N P U T  D A T A ?  A G A I N  707076  / Y E S ,  READ  J M P * R E A D  / E X I T  FROM  0  / A  100  to  O P T I C A L  D E N S I T Y  CODE  INTO  PDP-9  SUBROUTINE  MICROSECOND  LOOP  D E L A Y  LAC(777740 DAC D# ISZ D JMP JMP*DELAY  .-1  •  It  s h o u l d be n o t e d t h a t the d e l a y can e i t h e r be a '•'do-nothing" l o o p , as shown,  o r , more e f f i c i e n t l y , is  the program can be r e s t r u c t u r e d so t h a t the d e l a y time  taken up by some o t h e r i n s t r u c t i o n s which p r o c e s s d a t a .  3.  3.1  T H E  S Y S T E M  SOFTWARE  T e s t Programs A  program c a l l e d  "TEST70" was w r i t t e n f o r the PDP-9 t o check t h e  o p e r a t i o n o f the computer-device F i g u r e 2.3  interface.  Each o f the commands t a b l e d i n  c o u l d be d i s p a t c h e d , upon t e l e t y p e command, i n a l o o p so t h a t the  16  e x e c u t i o n and  t i m i n g c o u l d be  on the proper IOP timing  line.  seen on an o s c i l l o s c o p e , by  T h i s program f a c i l i t a t e s  sequences of a l l l o g i c a l f u n c t i o n s  the  performed by  t r i g g e r i n g the  c h e c k i n g of the the system.  scope  proper  But,  to  check the l i m i t a t i o n s of system parameters such as r e s o l u t i o n , a second major program c a l l e d "RID"  was  Chapter 1.  The  use  neutrophils  was  also  3.2  The  Simulation The  written;  i t produced the r i d g e f u n c t i o n d i s c u s s e d  in  of the c a l c u l a t e d r i d g e f u n c t i o n i n the c l a s s i f i c a t i o n  of  investigated.  of the A r e a Operator and  first  surrounding erythrocytes  and  o p t i c a l density  o t h e r p a r t of the c e l l p l o t s were made, u s i n g  to t r y to d e f i n e a s p e c i f i c g r e y - l e v e l which was  o f the n u c l e u s . Figure  to l o c a t e the l e u k o c y t e n u c l e u s .  the nucleus i s more opaque than any  input device,  Function  problem encountered i n implementing the r i d g e f u n c t i o n  a d e s c r i p t o r of the shape o f n u c l e i was tunately,  Ridge  A t y p i c a l g r e y - l e v e l p r o f i l e of a n e u t r o p h i l  For-  or the  picture  characteristic i s shown i n  3.1.  plane of  cross-section  cytoplasm white  nucleus black  cross-section  photomicrograph Figure  3.1  The g r e y - l e v e l c r o s s - s e c t i o n of a neutrophil.  as  The  transmittance  o f the photomicrograph was q u a n t i z e d  linearly  i n t o 8 l e v e l s and the n u c l e u s o f the n e u t r o p h i l was found to have a g r e y - l e v e l o f 0.  (Only  t h r e e b i t s o f o p t i c a l d e n s i t y r e s o l u t i o n , over a dynamic range  o f 5 v o l t s , were used because o f the s i g n a l - t o - n o i s e r a t i o  l i m i t a t i o n s of  the image d i s s e c t o r camera). To d i s t i n g u i s h a form o f i n t e r e s t from the r e s t o f the p i c t u r e , the program "RID" d e f i n e s a g r e y - l e v e l window; o p t i c a l d e n s i t i e s l y i n g t h i s window a r e c a l l e d " b l a c k "  and those o u t s i d e ,  "white".  v a r i a b l e w i d t h and v a r i a b l e o p t i c a l d e n s i t y p o s i t i o n .  within  The window has a  For instance,  defining  the l e v e l 0 t o be b l a c k and o t h e r s white, s u f f i c e s to d i s t i n g u i s h a b l a c k nucleus.  I t i s a l s o p o s s i b l e to d i s t i n g u i s h the c y t o p l a s m o f the n e u t r o p h i l  by h a v i n g  a window w i t h g r e y - l e v e l s from 2 to 5 i n c l u s i v e . To l o c a t e the n u c l e u s i n the scan f i e l d  used.  S t a r t i n g a t a programmed p o i n t near the lower l e f t  a scan i s made from bottom to top, spacing,  stepping  o f the image,  Upon l o c a t i n g the f i r s t  ( x ^ , y ^ ) , the program s w i t c h e s from a scan to a c o n t o u r - t r a c e  x + 2GS A f t e r t h i r t y trace steps, mode w i l l  point  GS; i f a b l a c k p o i n t i s n o t found the scan i s r e s e t to the bottom  t h i s time, a neighbourhood around the p o i n t  one  corner  i n i n c r e m e n t s o f the g r i d  o f the image and i n c r e m e n t e d GS h o r i z o n t a l l y . point,  a t e l e v i s i o n - t y p e scan was  be h a l t e d .  and  y ^  2GS  the t r a c e  One complete c o n t o u r o f a c l o s e d f i g u r e w i l l  i t cannot be ensured t h a t the p o i n t  o f the s t a r t i n g p o i n t i s d e f i n e d  encounter  (With  jitter  (-j_»y-j_) w i l l x  the n u c l e u s and thus a s m a l l  to s t o p  At  (3.1)  i f a p o i n t i s found w i t h i n t h i s r e g i o n ,  countered again a f t e r c o n t o u r - t r a c i n g  mode.  (x ,y ) i s d e f i n e d by the r e g i o n s  p o i n t w i t h i n t h i s neighbourhood and h a l t the t r a c e mode.  n o i s e i n the d e v i c e  black  be en-  neighbourhood  the t r a c e mode).  F o r a b l a c k f i g u r e , the n e u t r o p h i l n u c l e u s i n t h i s case,  and a white  environment," a c l o c k w i s e step algorithm.  t r a c e i s made around the f i g u r e i n a simple  square  The a l g o r i t h m , as used by Mason and Clemens [ l 3 ] , i s g i v e n  by the f o l l o w i n g r u l e s : first  x  y  = x  (3.2)  1  o  n  = y - cs  0  (3.3)  x  and upon i n t e r r o g a t i n g (x ,»y, ) and f i n d i n g i t i s bla.ck: x  k i  =  x  +  k  +  C y  y-k+l  k-i  " k y  t\  +  ]  (  3  '  4  )  ~ x _ ]  (3-5)  - y ^ ]  (3.6)  k  x  o r i f (x ,y ) i s a white p o i n t : x _ = x k+l 'k+l and  =  y  k  [y  +  +  f e  V i " k  [  x  k  ]  i f t h r e e white p o i n t s o r t h r e e b l a c k p o i n t s a r e encountered  the next p o i n t i s t r e a t e d as i f i t were b l a c k o r white It thus  i n succession,  respectively.  i s assumed t h a t a l l p a r t s o f the f i g u r e a r e w i d e r than GS and  the a l g o r i t h m s h o u l d encounter  every p o i n t w i t h i n GS o f the edge o f t h e  figure. A t each white p o i n t (x.,y.) based on e q u a t i o n diameter.  (l.l),  was formed - a q u a n t i z e d d i s c ,  Each p o i n t i n s i d e  which a r e w h i t e , a r e '  the  operator,  31 g r i d s p a c i n g s i n  the a r e a o p e r a t o r i s i n t e r r o g a t e d and a l l those  summed.  Thus A  J  where P  a l o n g the t r a c e , an a r e a  1  is  found.  J  white p o i n t i n s i d e the a r e a  operator.  Now A^ i s known t o be 725 - c a l c u l a t e d from the known s i z e o f the d i s c . Analytically A„  =  £ P  4  (3.9)  19 where Therefore  = the i  th  p o i n t i n the a r e a  operator.  r For V  r  , where  the r i d g e f u n c t i o n a t (x.,y.) i s V  each white p o i n t w i t h i n one  i s c a l c u l a t e d and  w  t  grid  3.1,  The with respect  of the n u c l e a r  The c o n t o u r - p a t h and nucleus.  s i z e of the a r e a o p e r a t o r  c o n t o u r t r a c e can  to the n e u t r o p h i l image, by v a r y i n g GS  the programmer to v a r y  GS  from 1 up  to 15  function  3.2  .  function  r i d g e f u n c t i o n of a  and  boundary  ridge  i s shown i n F i g u r e  ridge  contour-trace  3.2  spacing  s t o r e d i n core memory; the d i s c r e t e - v a l u e d  f o r the nucleus' o f the n e u t r o p h i l i n F i g u r e  Figure  (3.10)  r -  1  f o r each.  neutrophil  be  changed,  "RID"  fundamental g r i d p o i n t s of  allows the  image.  3.3  Problems Encountered i n Computing the Ridge Since  the r i d g e f u n c t i o n was  Function  implemented on d a t a c o n s i s t i n g of  actual  leukocyte  images, problems i n f o r m i n g  t h e a r e a o p e r a t o r and c a l c u l a t i n g t h e  r i d g e f u n c t i o n arose, which would n o t o c c u r i f t e s t p a t t e r n s such as the one i n F i g u r e 1.4 were used.  I t was the purpose o f t h i s t h e s i s ,  though, to d i s -  c o v e r problems a r i s i n g from working w i t h m i c r o s c o p i c images so t h a t an eval u a t i o n o f the u s e f u l n e s s o f the v i s u a l - i n p u t d e v i c e , i n f u t u r e work, c o u l d be made. I d e a l l y , the contour-trace  g r i d s p a c i n g s h o u l d be kept as s m a l l as  p o s s i b l e t o m i n i m i z e q u a n t i z a t i o n n o i s e caused  when a r c l e n g t h i s segmented.  In a c t u a l p r a c t i c e , when t h e g r i d s p a c i n g was k e p t tended  small, the contour  trace  to f o l l o w the s m a l l i r r e g u l a r i t i e s i n the boundary o f t h e n e u t r o p h i l  n u c l e u s - t h i s caused  the t r a c e to g e t caught i n l o o p s and, i n g e n e r a l , produced  a poor boundary t r a c e . parencies i s approaching  S i n c e the m a g n i f i c a t i o n o f t h e n e u t r o p h i l s i n the t r a n s the l i m i t of o p t i c a l r e s o l u t i o n ,  c e l l n u c l e u s i s bound t o be ragged  and " g r a i n y " .  t h e boundary o f t h e  As the g r i d s p a c i n g was  i n c r e a s e d to above GS=8, t h e t r a c e d i d n o t f o l l o w t h e i r r e g u l a r i t i e s and appeared to f o l l o w the boundary i n a more c o n t i n u o u s  GS=8  GS=4  F i g u r e 3-3  Contour-traces  manner.  (See F i g u r e 3-3).  GS=12  showing t h e e f f e c t s o f i n c r e a s i n g t h e g r i d  spacing.  As  the g r i d s p a c i n g was i n c r e a s e d above GS=2, the f i x e d s i z e  aperture  of the image d i s s e c t o r gave a poor average o f the t r a n s m i t t a n c e i n the element o f a r e a which the g r i d p o i n t d e f i n e s . the a p e r t u r e w i l l mil  2  s t e p to e v e r y  F o r i n s t a n c e , i f GS-£, where 3 £ $ ^ 15,  fundamental image g r i d p o i n t and the 1  (2  i n t e r r o g a t i o n area, r e p r e s e n t s a s p a t i a l a r e a o f o  2  mil .  This  effect  l e a d s to e r r o r s i n the c o n t o u r - t r a c e .  (2 By s i m u l a t i n g a l a r g e r a p e r t u r e which c o v e r s the e n t i r e  2  o mil.  a r e a , t h i s problem was overcome. The  v i d e o s i g n a l v o l t a g e r e p r e s e n t s an average o f the o p t i c a l d e n s i t y  over the a r e a o f the e x i s t i n g a p e r t u r e . area, a l l t h a t i s r e q u i r e d i s t o extend  Then, to s i m u l a t e a l a r g e r  aperture  the a v e r a g i n g p r o c e s s over a l a r g e r  area. A  square  aperture with a v a r i a b l e side i  was s i m u l a t e d .  i n t e r r o g a t e d a l l fundamental g r i d p o i n t s w i t h i n the a r e a o p t i c a l d e n s i t y codes i n i t s CPU. c l o s e approximation  S  The PDP-9  and averaged the  The r e s u l t a n t o p t i c a l d e n s i t y code i s a  to t h a t o f an a c t u a l square  a p e r t u r e w i t h an a r e a o .  I n a c o l o r l e s s t r a n s p a r e n c y o f the a c t u a l n e u t r o p h i l , i t was t h a t the boundary between the n u c l e u s  and t h e cytoplasm was, a t times,  d e f i n e d , e s p e c i a l l y when the cytoplasm  c o n t a i n e d dark g r a n u l e s .  poorly  The l a c k o f  g r e y - l e v e l c o n t r a s t between the two p a r t s o f the l e u k o c y t e c o m p l i c a t e d distinction.  found  their  To overcome t h i s problem, a l o g a r i t h m i c a m p l i f i e r was used to  expand the c o n t r a s t between the two r e g i o n s .  S i n c e the A / D  a n a l o g range, as  w e l l as the dynamic range o f the v i d e o s i g n a l v a r i e d from 0 t o -5 v o l t s , a l o g a r i t h m i c a m p l i f i e r w i t h the f o l l o w i n g c h a r a c t e r i s t i c s had to be b u i l t :  y(t) =  m(-x(t)+i)  (3.11)  where  y ( t ) = the output  o f the n o n - l i n e a r a m p l i f i e r ,  and  x ( t ) = the i n p u t v i d e o s i g n a l t o the a m p l i f i e r .  22  A schematic  diagram o f the l o g a r i t h m i c a m p l i f i e r i s shown i n Appendix I I I . The  t r a n s f o r m a t i o n from a l i n e a r to a l o g a r i t h m i c o p t i c a l density-  s c a l e i s d e s i r a b l e s i n c e human b r i g h t n e s s p e r c e p t i o n i s a l s o l o g a r i t h m i c and the PDP-9 then works w i t h the same o p t i c a l d e n s i t i e s t h a t a human o b s e r v e r would see. A c c o r d i n g to the d e f i n i t i o n o f the a r e a o p e r a t o r i n Chapter  1, i t  i s r e q u i r e d t h a t , a t each p o i n t on the curve C, the d i s c s h o u l d e n c l o s e o n l y one  continuous  c u r v a t u r e segment.' Thus, the d i s c s h o u l d n o t o v e r l a p any o t h e r  p o r t i o n o f the n u c l e u s w h i l e o p e r a t i n g on a p a r t i c u l a r c u r v a t u r e segment, as shown i n F i g u r e 3.4.  The a r e a o p e r a t o r s h o u l d a l s o have a r a d i u s s m a l l e r  than the t h i n n e s t p a r t o f the f i g u r e i t i s p r o c e s s i n g . Both the above c o n d i t i o n s were, a t times, u n a v o i d a b l y  violated  because o f the n a t u r e o f the shape o f the a c t u a l n e u t r o p h i l s p r o c e s s e d .  area operator r a d i u s i s too large  nuclear  area operator encloses two segments o f the n u c l e u s a t one time  outline  F i g u r e 3.4  T y p i c a l e r r o r s made when o p e r a t i n g on a c t u a l neutrophil nuclei.  As the s i z e o f the a r e a o p e r a t o r was decreased, occurrences  o f the above mentioned e r r o r s ,  to reduce the  the o p e r a t o r became too s e n s i t i v e  23  to c u r v a t u r e .  Q u a n t i z a t i o n n o i s e was a l s o added t o t h e r i d g e f u n c t i o n s i n c e  the s i z e o f t h e c o n t o u r - t r a c e s t e p s i s i n c r e a s e d w i t h r e s p e c t t o t h e a r e a operator s i z e . sensitivity  F i g u r e 3.5 shows the e f f e c t s o f i n c r e a s e d r i d g e f u n c t i o n  t o c u r v a t u r e , and i n c r e a s e d q u a n t i z a t i o n n o i s e .  a r e a o p e r a t o r GS=8  F i g u r e 3.5  a r e a o p e r a t o r GS=5  Ridge f u n c t i o n s d e r i v e d u s i n g d i f f e r e n t a r e a o p e r a t o r  4.  sizes.  A NEUTROPHIL CLASSIFICATION STUDY  H a v i n g t h e means t o produce a r i d g e f u n c t i o n which i s a shape d e s c r i p t o r , i t i s then p o s s i b l e to a c t u a l l y t e s t i f the f u n c t i o n c a n be used f o r l e u k o c y t e c l a s s i f i c a t i o n .  The f o l l o w i n g problem was posed:  determine t h e age and l o b u l a r i t y o f n e u t r o p h i l s .  Of the f i v e major  types o f l e u k o c y t e s , n e u t r o p h i l s were chosen as a t e s t c l a s s because seem to have a f a i r l y w e l l d e f i n e d boundary between  the n u c l e u s  to morphological they  and c y t o -  plasm. The m o r p h o l o g i c a l  ages o f i n t e r e s t f a l l  the metamyelocyte, t h e banded and t h e segmented  into  cells.  t h r e e b a s i c groups: Banded and segmented  c e l l s a r e abundant i n h e a l t h y p e r i p h e r a l b l o o d , w h i l e metamyelocytes a r e r a r e l y  24  found.  The t y p i c a l n u c l e a r s h a p e o f e a c h o f t h e t h r e e g r o u p s i s shown i n F i g u r e  4.1.  F i g u r e 4.1  The g e n e r a l n u c l e a r s h a p e s o f n e u t r o p h i l s .  Although aging i s a continuous process, f o rquantitative made, s h a r p d e c i s i o n " b o u n d a r i e s m u s t be d r a w n . k i d n e y o r bean shaped;  t h e ba:id c e l l  decisions to  The m e t a m y e l o c y t e n u c l e u s i s  n u c l e u s i s horse-shoe shaped and o c c u p i e s  a s m a l l e r a r e a t h a n t h a t o f t h e metamyelocyte; and, t h e segmented  nucleus has  l o b e s o r i s l a n d s i n t e r c o n n e c t e d by f i l a m e n t s .  are para-  phased from a hematology manual s h o u l d be r e l a t i v e l y s h o u l d have that, peak.  smooth  one m a i n peak  f o r t h e segmented  [l4]).  (These d e f i n i t i o n s  The r i d g e f u n c t i o n f o r m e t a m y e l o c y t e s  w i t h a s m a l l hump; t h a t , f o r t h e b a n d n u c l e u s  corresponding'to  the concave p a r t o f the horseshoe;  n u c l e u s s h o u l d be i r r e g u l a r a n d h a v e more t h a n one  F i g u r e 4.2 s h o w s a t y p i c a l  r i d g e f u n c t i o n f o r e a c h o f t h e age g r o u p s .  25  metamyelocyte  banded  F i g u r e 4.2  segmented  Typical ridge functions.  A peak of c u r v a t u r e i s d e f i n e d , here, line.  Two  to be above the t h r e s h o l d  peaks cannot l i e a d j a c e n t w i t h a s p a c i n g l e s s than one  l e n g t h o f the r i d g e f u n c t i o n .  ( T h i s d e f i n i t i o n o f a peak t r i e s to  q u a n t i z a t i o n n o i s e and n u c l e a r i r r e g u l a r i t i e s which a r e not Looking is  a t F i g u r e 4.1,  i t may  becomes more i r r e g u l a r as the c e l l ' s age  increases.  o f the  A = the a r e a o c c u p i e d  i s s i z e and  a lower bound of 24%  o r i e n t a t i o n independent.  dominant). aging  G e n e r a l l y , the n u c l e a r boundary A parameter which g i v e s  a measure o f the i r r e g u l a r i t y o f a f i g u r e boundary i s T/TA] P = the p e r i m e t e r  the  exclude  be n o t i c e d t h a t a p r o p e r t y of  the f o r m a t i o n of l o b e s i n the n u c l e u s .  T h i s parameter, which has  fifth  P was  where  figure.  by the  figure.  i n the case  found  of a p e r f e c t c i r c l e ,  by c o u n t i n g  the number of g r i d  p o i n t s i n the counting  contour-trace,  a l l black g r i d  and  the a r e a  points.  o f the n u c l e u s ,  Thus the f e a t u r e  A,  can be  was  found  by  extracted  q u a n t i t a t i v e l y on the PDP-9. T h i r t y transparencies o f the was  t h r e e age  two and  groups were p r o c e s s e d  determined by a q u a l i f i e d Figure  of c a r e f u l l y chosen n e u t r o p h i l s ,  4.3  on the PDP-9.  The  age  ten o f each  of each  cell  specialist [ l 5 J .  shows a two-dimensional p l o t i n measurement space;  f e a t u r e s used to c l a s s i f y n e u t r o p h i l s are the number of peaks per  the  nucleus  P/VT. Figure  4.3  shows t h a t the  three  cell  age  groups are d i s t i n g u i s h a b l e  i n measurement space - t h e r e i s d e f i n i t e c l u s t e r i n g which a l l o w s  the  implemen-  t a t i o n of d e c i s i o n b o u n d a r i e s . It cessed the age  i s then r e a s o n a b l e  to assume t h a t the shape of the n u c l e u s ,  from the r i d g e f u n c t i o n and  the parameter P/7A~is a means o f  determining  of n e u t r o p h i l s . A function of curvature  with respect  as a d e s c r i p t o r of the shape of n u c l e i and the use  pro-  o f the a r e a o p e r a t o r  to a r c l e n g t h was  found u s e f u l  i n turn t h e i r c l a s s i f i c a t i o n ,  to d e r i v e t h i s f u n c t i o n has  four d i s t i n c t  but  dis-  advantages: a)  to compute one which make up the p r o c e s s  b)  p o i n t on the r i d g e f u n c t i o n , 725 the a r e a o p e r a t o r  grid  points  must be i n t e r r o g a t e d -  therefore  i s i n h e r e n t l y slow.  the a r e a o p e r a t o r  overlapped  the n u c l e u s a t t h i n r e g i o n s  giving  e r r o r s i n the r i d g e f u n c t i o n . c) the a r e a o p e r a t o r n u c l e u s a t one  impinged on more than one  time, e s p e c i a l l y w i t h  segment of  the  segmented n u c l e i , g i v i n g  another form o f e r r o r i n the r i d g e f u n c t i o n .  27  10 P -  M  METAMYELOCYTE  B  BANDED CELL  S  SEGMENTED  CELL  S  s  S  8  B B  S s B B  B  S 5  B B B B M  S  S  S  M  M MM M MM J 0  1  J  L 2  3  4  I 5  NUMBER OF PEAKS OF RIDGE FUNCTION  F i g u r e 4.3  A measurement-space graph showing the c l u s t e r i n g of n e u t r o p h i l age g r o u p s .  d) A l i n e a r r e a d i n g o f c u r v a t u r e was not made i n most cases s i n c e the r a d i u s o f c u r v a t u r e o f the n u c l e u s was r a r e l y l a r g e r the r a d i u s o f the a r e a The  statements  than  operator.  b) and c) above, have been i l l u s t r a t e d i n F i g u r e 3-4.  F o r the g e n e r a l p a t t e r n c l a s s o f l e u k o c y t e s , the a r e a o p e r a t o r p r o v i d e s a poor method o f e x t r a c t i n g n u c l e a r c u r v a t u r e .  At small operator r a d i i ,  e r r o r s o u t l i n e d i n b) and c) a r e n o t encountered, i n the r i d g e f u n c t i o n becomes too h i g h . increased,  the q u a n t i z a t i o n n o i s e  level  When the s i z e o f the o p e r a t o r i s  the r e q u i r e d c u r v a t u r e s e n s i t i v i t y i s decreased  f u n c t i o n becomes erroneous.  where the  and the r i d g e  Then,.to overcome the d i s a d v a n t a g e s  mentioned  above, an a p p r o x i m a t i o n  method o f c u r v a t u r e measurement, as used by L e d l e y ,  s h o u l d be implemented.  The a p p r o x i m a t i o n  from the c o n t o u r  itself,  method r e q u i r e d i n f o r m a t i o n , o n l y  n o t from an a r e a around the contour,  because i t i s  based on the d e f i n i t i o n o f c u r v a t u r e dt ds  K = where  (4.1)  d t = the d i f f e r e n t i a l change i n tangent  v e c t o r on  the curve C. ds = the d i f f e r e n t i a l change i n a r c l e n g t h a l o n g C. Assume t h a t the curve C i s q u a n t i z e d i n t o e q u a l l e n g t h segments a t p o i n t s s^, S g , s ^ , . . . s ^ , . . . ,  then the c u r v a t u r e a t a p o i n t o f a r c l e n g t h s^  can be approximated by the e x p r e s s i o n :  K  |  l  (  s  S  s • 1  s. ) - ( s . - s .)  k - i  - .i S  )  }  s. - s .  I  1  (4.2)  J  where  s, = s. + nAs and s . = s. - nAs. k x j x  and  As = b a s i c a r c l e n g t h q u a n t i z a t i o n ( i . e . A s = s ^ ^ - s ^ ) . +  n = a non-zero, p o s i t i v e i n t e g e r .  29 The  c u r v a t u r e , a c c o r d i n g to e q u a t i o n 4.2,  u s i n g the c o n t o u r - t r a c e as a segmented c u r v e .  can r e a d i l y be  T h i s a l g o r i t h m would produce  a f u n c t i o n g i v i n g c u r v a t u r e w i t h r e s p e c t to arc l e n g t h and l e s s time consuming than the f o r m a t i o n of an a r e a  5.  5.1  Noise  THE  calculated  i s o b v i o u s l y much  operator.  SYSTEM PARAMETERS  C h a r a c t e r i s t i c s of the Image D i s s e c t o r Tube The  t h e o r y of the image d i s s e c t o r tube, u n l i k e the o r t h i c o n o r  v i d i c o n tubes,  i s simple and  noise l i m i t a t i o n s .  thus i t i s r e l a t i v e l y easy to f o r m u l a t e b a s i c  T h i s t h e o r y p l a y s an i m p o r t a n t  r o l e i n the d e s i g n of  v i s u a l - i n p u t d e v i c e , s i n c e the n o i s e l e v e l l i m i t s the d e t e c t a b i l i t y of s i g n a l and  thus  the number of l e v e l s i n t o which the v i d e o s i g n a l can  quantized f o r d i g i t a l The  the  the  be  processing.  d e t e c t i o n of a s i g n a l i s hampered by the f o l l o w i n g n o i s e  sources:  a) R e s i d u a l dark n o i s e or the t h e r m i o n i c e m i s s i o n of c u r r e n t , d e t e c t e d w i t h no f l u x i n c i d e n t on the b) M o n i t o r i n g c i r c u i t  photocathode,  dark n o i s e ,  c ) I n h e r e n t random f l u c t u a t i o n s o f the s i g n a l f l u x , t i c a l v a r i a t i o n s i n the s i g n a l c u r r e n t ,  causing  statis-  and  d) Secondary e m i s s i o n n o i s e i n t r o d u c e d by the p h o t o m u l t i p l i e r . F o r t h i s a p p l i c a t i o n of the image d i s s e c t o r tube, was  found  to be n e g l i g i b l e ,  s i n c e i t i s a t l e a s t f i v e o r d e r s o f magnitude  s m a l l e r than the minimum s i g n a l c u r r e n t . The frequency  monitoring  the tube dark n o i s e  (See r e f e r e n c e  [l7]).  circuit  o r wide-band v i d e o a m p l i f i e r i n t r o d u c e d  low  n o i s e i n t o the system.  Under a c t u a l working c o n d i t i o n s , 60 Hz  hum  w i t h some t h i r d harmonic d i s t o r t i o n was  observed.  v o l t a g e superimposed on the v i d e o s i g n a l was  The  about .25  peak-to-peak n o i s e v o l t s over a dynamic  r a n g e o f 5.5 v o l t s .  T h i s n o i s e c a n be r e d u c e d b y c a r e f u l p l a c e m e n t o f t h e  power s u p p l i e s a n d s h i e l d i n g o f t h e image d i s s e c t o r t u b e a n d v i d e o  amplifier.  A notch  a t 60 H z .  filter  c o u l d be e m p l o y e d t o r e d u c e t h e n o i s e  power p r e s e n t  The f u n d a m e n t a l n o i s e , c a u s e d b y s t a t i s t i c a l signal is  c u r r e n t i s shot n o i s e , which i s "white".  f l u c t u a t i o n s of the  The mean s q u a r e n o i s e  current  g i v e n by i  where  2  = 2qI Af  (5.1)  k  . q = t h e charge on an e l e c t r o n . I  = t h e d.c. s i g n a l c u r r e n t e n t e r i n g t h e a p e r t u r e .  Af = the bandwidth of the video  signal .  The e f f e c t o f t h i s f u n d a m e n t a l n o i s e i n c o n j u n c t i o n w i t h emission noise ratio,  secondary  gives the f o l l o w i n g peak-to-peak s i g n a l - t o - n o i s e current  derived i n [ l 8 ] : S N  where  J  , J A(cr-i) ' 7 2CTq Af •' 1  P-P  = t h e d.c. c u r r e n t d e n s i t y i m p i n g i n g aperture  A = the area dAs equation largest value  of I  (5.2)  k  on t h e  plane . of the aperture.  the average g a i n per stage  of the photomultiplier.  (5«l) p o i n t s o u t , t h e n o i s e a m p l i t u d e  i s largest at the  and t h e number o f s i g n a l q u a n t i z a t i o n l e v e l s  t h a t c a n be  .K. used  must be c a l c u l a t e d a t t h i s c o n d i t i o n .  d e s i g n o f t h e image d i s s e c t o r tube, to  Other than  changing the i n t e r n a l  t h e o n l y way t o r e d u c e t h e n o i s e l e v e l i s  decrease the bandwidth,Af, o f the e x t e r n a l low-pass f i l t e r .  be t h e o r e t i c a l l y made a r b i t r a r i l y  s m a l l , t h e A.&fo m o n i t o r i n g  p r o v e d t o be t h e l i m i t i n g f a c t o r f o r n o i s e . meaningfully  quantized  Then, t h e v i d e o  Since.Af can  circuit  dark  noise  s i g n a l c a n be  t o , a t most, e i g h t l e v e l s o r t h r e e b i n a r y b i t s .  This  was  adequate g r e y - l e v e l r e s o l u t i o n f o r p r e l i m i n a r y r e s e a r c h work.  e i g h t l e v e l s can  be r e s o l v e d by r e d u c i n g the m o n i t o r i n g c i r c u i t n o i s e  d e c r e a s i n g the low-pass f i l t e r  5.2  The  and  bandwidth.  S p a t i a l Frequency Spectrum Before  necessary image.  More than  the bandwidth of the v i d e o s i g n a l can  be found,  to c o n s i d e r the s p a t i a l f r e q u e n c i e s w h i c h . e x i s t on the  A s p a t i a l frequency  i t is optical  can be d e f i n e d as the r a t e of change of  o p t i c a l d e n s i t y i n the two-dimensional f i x e s the s p a t i a l f r e q u e n c i e s and  image p l a n e .  they are s a i d  The  image m a g n i f i c a t i o n  to e x i s t i n the s p a t i a l domain.  When the o p t i c a l d e n s i t y a l o n g some curve i n the two-dimensional scanned by the a p e r t u r e of the image " d i s s e c t o r tube,  image i s  a time v a r y i n g v i d e o  f u n c t i o n i s produced which i s p r o p o r t i o n a l to the o p t i c a l d e n s i t y . s c a n n i n g p r o c e s s i s s a i d to t r a n s f o r m  the  the s p a t i a l v a r i a t i o n s i n t o  The the  time  domain. Assume the s p a t i a l o p t i c a l d e n s i t y v a r i a t i o n s on the image, are g i v e n by the t w o - v a r i a b l e f u n c t i o n f ( x , y ) .  C o n s i d e r a scan a l o n g an  curve, X" ; the x and y c o o r d i n a t e s can be e x p r e s s e d v a r i a b l e a r c l e n g t h a l o n g Y.  i n terms o f the  arbitrary  single  That i s ,  f ' ( x , y ) = f(s)  (5.3)  where s i s a r c l e n g t h measured a l o n g the c u r v e The  a p e r t u r e a c t u a l l y scans  quantized increments  As.  the image i n a d i g i t a l manner, u s i n g  Now  (5.4)  As = r A t where  Y.  As = s p a t i a l ,  a r c l e n g t h increments  about 1 m i l i n t h i s  case,  r = the r a t e of scan a l o n g  X.  - f i x e d at  At = the ''dwell aperture S i n c e As  i s very  time" or the time r e q u i r e d by  to i n t e r r o g a t e a  s m a l l compared to the  o f the image, e q u a t i o n ( 5 . 4 ) can  the  point.  .8 i n c h by  .8 i n c h expanse  be approximated, w i t h l i t t l e  error,  by  s = rt  (5.5)  Using equation ( 5 - 5 ) , be s t a t e d  the  time domain r e p r e s e n t a t i o n  of f ( s )  as v(t) = f (J)  v(t),  then, i s the  the v i d e o  (5.6)  time domain r e p r e s e n t a t i o n  s i g n a l produced by a scan The  F o u r i e r transform v ( t ) —  A s i m i l a r d e f i n i t i o n can be  of the f u n c t i o n f ( s ) and  of v ( t ) i s defined  to be V ( f ^ ) .  That i s ,  V(f ) ^ F{v(t)]  (5.7)  t  a p p l i e d to f ( s ) , (5.8)  s  since v(t) = f  and  is  along  f ( s ) — F ( f ) ^ F{f(s)} Now,  can  r has  (5,9)  some non-zero, p o s i t i v e v a l u e f(J) —  by the  (f)  s c a l i n g property  r F(r.f )  (5.10)  of F o u r i e r transforms. V(f ) t  = r F(r.f )  (5.11)  g  E q u a t i o n ( 5 . 1 l ) c l e a r l y shows t h a t the of f ( s ) , V ( f ^ ) , V(f^_) can  Therefore  time-domain F o u r i e r  i s f r e q u e n c y s c a l e d by the r a t e c o n s t a n t  be compressed.  trum of the v i d e o  The  By d e c r e a s i n g  a b i l i t y to frequency-compress the F o u r i e r  s i g n a l v ( t ) i s important,  the e x t e r n a l bandwidth.  r.  spectrum  since t h i s allows  Then, the fundamental n o i s e  spec-  a decrease i n  power can be  r,  reduced.  The  n e u t r o p h i l image, p r e s e n t  on the photocathode, r e p r e s e n t s an  o p t i c a l m a g n i f i c a t i o n o f about 1400 from the a c t u a l c e l l .  At t h i s  tion,  of o p t i c a l r e s o l u -  tion.  the image c o n t a i n s i n f o r m a t i o n a p p r o a c h i n g the l i m i t  T h i s l i m i t i s dependent on the wavelength o f the l i g h t i l l u m i n a t i n g  the c e l l  and the i m p e r f e c t i o n s  p o i n t source The  limit  o f the m a g n i f y i n g  of l i g h t i s transformed  lenses.  o f r e s o l u t i o n i s u s u a l l y d e f i n e d t o be twice  expressed  i n the f o l l o w i n g  where  Theoretically, a  i n t o an A i r y d i f f r a c t i o n p a t t e r n [20].  maxima t o t h e f i r s t minima o f the A i r y p a t t e r n . is  magnifica-  the d i s t a n c e from the  Analytically,  this  limit  equation:  R = the r e s o l u t i o n l i m i t  distance  (expressed  i n the  • same u n i t s as A.). X - the wavelength o f the i l l u m i n a t i n g l i g h t . N.A. = the n u m e r i c a l  aperture  o f the m i c r o s c o p e  objective . For green l i g h t ,  the r e s o l u t i o n l i m i t i s about .25_/t.  A n e u t r o p h i l has an average d i a m e t e r o f about 12^ and i t s m a g n i f i e d image on the photocathode w i l l have a d i a m e t e r o f 16.8 mm  • 25yit r e s o l u t i o n l i m i t i s m a g n i f i e d to .35 mm. o p t i c a l density information i s present image.  I t i s then e v i d e n t  upper bound.  ( m i l l i m e t e r s ) . The  T h i s i m p l i e s t h a t no m e a n i n g f u l  a t d i s t a n c e s s m a l l e r than .35 mm on the  t h a t s p a t i a l o p t i c a l d e n s i t y f r e q u e n c i e s have an  The maximum s p a t i a l f r e q u e n c y  of F ( f ) , g  f _ s  m a x  >  c  a  n  then be  d e f i n e d as  f  =7 s-max  Since F ( f ) i s band-limited s bound f r e q u e n c y  of f, , where t-max  (5.13)  R  w  to f  s-max  , V ( f , ) w i l l a l s o have an upper t  34 f\ = r f t-max s-max  5.3 The S p a t i a l  (5.14)  Resolution  F o r l e u k o c y t e c l a s s i f i c a t i o n , the image d i s s e c t o r  tube s h o u l d be  r e q u i r e d to r e s o l v e the d i s t a n c e between two l i n e s , b l a c k and white, depth  of modulation  traversing  o f 100$.  I n o t h e r words, an o p t i c a l d e n s i t y  the complete dynamic range w i t h i n two fundamental g r i d  at a  signal points  s h o u l d be d e t e c t a b l e . Using this d e f i n i t i o n , can be f o r m u l a t e d .  Firstly,  three d i f f e r e n t  forms o f s p a t i a l  s i n c e the computer v i s u a l - i n p u t  f o r a s p e c i f i c u s e , o p e r a t i n g on a r e s t r i c t e d c l a s s frequency  spectrum o f the images may be f o r m u l a t e d .  resolution  x  to  system i s d e s i g n e d  o f images, the s p a t i a l Then, a minimum  spatial  c a n be d e f i n e d . I f P ( f ) i s found s'  signal  resolution  resolution,  R  sp  to have an upper bound o f f  s-max  , then the s p a t i a l  , is  R  sp  =-7  (5.15)  f s-max  In t h i s p a r t i c u l a r resolution  as shown i n F i g u r e  e*—  u(x,y)  the s p a t i a l s i g n a l r e s o l u t i o n  i s s i m p l y the o p t i c a l  l i m i t which i s 14 m i l s , on the image. Secondly,  input picture  case,  the d e g r a d a t i o n o f the image by the d e v i c e c a n be modelled 5-1  h'(x,y) o p t i c a l image focus  F i g u r e 5.1  optically f o c u s e d image  g(x,y)  output image  • (x,y)  magnetic image focus  r(x,y)  A model o f the d e g r a d a t i o n  o f an image.  ®>  35 Assume the i n p u t p i c t u r e i s a s t e p f u n c t i o n i n o p t i c a l d e n s i t y , from b l a c k to white i n t h e x d i r e c t i o n .  Then the i n p u t i s o f the form  u(x,y) = u C x , ^ )  (5.16)  The l e n s e s f o c u s u ( x , y ) onto the photocathode g i v i n g r i s e pattern.  going  to an A i r y d i f f r a c t i o n  The o p t i c a l image undergoes a t r a n s f o r m a t i o n , g i v e n by the impulse  response h ( x , y ) , which i s o f the form  h(x,y) „ k  0  sin k x ( — ^ - )  o p t i c a l l y f o c u s e d image i s c o n v e r t e d  photocathode.  (5.17)  2  y  i n t o an e l e c t r o n i c image by the  Sharp edges o f the e l e c t r o n i c image a r e defocused  manner because o f the p r o b a b i l i s t i c n a t u r e photocathode.  2  y  2 The  sin k y „ ( — ± - ) k  0  o f e l e c t r o n emission  Magnetic f o c u s i n g can be m o d e l l e d by the s p a t i a l  i n a Gaussian from the impulse  response g(x,y) = I t i s formulated edge, t r a n s f o r m e d a c t u a l extent The  k e~V -V 2  2  (5.18)  5  t h a t an o p t i c a l s t e p f u n c t i o n i s defocused  i n g e n e r a l form a c c o r d i n g e q u a t i o n s  ( t h e v a l u e s o f the c o n s t a n t s )  (5.17) and (5.18); i t s  can o n l y be found  edge o f a r a z o r blade was f o c u s e d as b e s t p o s s i b l e , f i r s t  experimentally. optically  the photocathode and then m a g n e t i c a l l y onto t h e a p e r t u r e plane, o p t i c a l density step f u n c t i o n . s t e p response.  i n t o an  F i g u r e 5.2 i s the q u a n t i z e d  onto  t o form an  optical density  Each l e v e l c o r r e s p o n d s  to the o p t i c a l d e n s i t y o f an i n t e r -  r o g a t i o n o f a fundamental g r i d p o i n t .  A minimum b l a c k to white, t r a n s i t i o n  o f f o u r fundamental g r i d p o i n t s , o r about 4 m i l s on the image, c a n be seen.  36  ^-edge o f r a z o r  F i g u r e 5.2  e n l a r g e d scan o f black-white t r a n s i t i o n  A scan a c r o s s the edge o f a r a z o r b l a d e .  T h i r d l y , when the v i s u a l - i n p u t d e v i c e i s u s e d under PDP-9 c o n t r o l , the minimum s p a t i a l r e s o l u t i o n which c a n be d e t e c t e d i s the d i s t a n c e between two  fundamental g r i d p o i n t s .  I t i s then o b v i o u s  d e t e c t a s i g n a l , a t a depth o f m o d u l a t i o n  t h a t the computer cannot  o f 100$, whose o p t i c a l density-  v a r i e s from b l a c k t o white i n a d i s t a n c e l e s s than about 2 m i l s , on t h e image.  5.4 Bandwidth Requirements The  frequency  spectrum o f the v i d e o s i g n a l emerging from the image  d i s s e c t o r tube c o n s i s t s b a s i c a l l y o f the sum o f the s i g n a l spectrum V ( f ^ ) and  a c o n s t a n t amplitude  w h i t e n o i s e spectrum.  Low-pass f i l t e r i n g a t the  o p t i m a l bandwidth s h o u l d e l i m i n a t e as much n o i s e power w i t h o u t  a substantial  l o s s o f s i g n a l power. In the d i s p l a y mode, the image i s scanned by a d j a c e n t the a p e r t u r e .  steps of  S i n c e the bandwidth o f the d e v i c e s h o u l d be made as s m a l l as  p o s s i b l e to reduce the n o i s e power, A f i s made e q u a l to f ,  .  T h i s band-  37  width i s sub-optimal If  but adequate, s i n c e the e x a c t form o f "V(f^) i s not known.  the d e t e c t i o n o f the h i g h e s t f r e q u e n c y  Af s h o u l d be wide enough so t h a t each a p e r t u r e o r i n o t h e r words, so t h a t o n l y the f i r s t filtered  out.  f  =  then  i n t e r r o g a t i o n i s b a r e l y "seen",  harmonic o f the s i g n a l i s n o t  Then, a c c o r d i n g to the sampling  A  possible i s required,  theorem  2 A T  (5.19)  F o r the computer-read mode, the e x t e r n a l bandwidth s h o u l d be as s m a l l as p o s s i b l e to reduce the n o i s e but n o t so s m a l l t h a t the i n h e r e n t caused by the low-pass f i l t e r time-consuming. to  will  Since a f i n i t e  be so l o n g t h a t the f e t c h c y c l e i s e x c e s s i v e l y  time i s r e q u i r e d f o r the d e f l e c t i o n a m p l i f i e r s  s e t t l e and s i n c e a s i g n a l p r o p a g a t i o n  filter,  the A / D c o n v e r s i o n o f the v i d e o  reaches  a d.c.  least  level  delay  d e l a y i s i n h e r e n t t o a low-pass s i g n a l s h o u l d s t a r t a f t e r the s i g n a l  to w i t h i n the a n a l o g v o l t a g e e q u i v a l e n t o f o n e - h a l f the  significant b i t of quantization.  A t the maximum d.c.  l e v e l o f the v i d e o  s i g n a l , Af was extended to the p o i n t where the peak-to-peak s i g n a l - t o - n o i s e r a t i o was no s m a l l e r than  16:1.  This s e t t i n g provides  p o s s i b l e , before A / D conversion begins,  the l e a s t f i l t e r  because i t i s the w i d e s t  a l l o w i n g three b i t g r e y - l e v e l r e s o l u t i o n .  delay  bandwidth  38 6.  The  CONCLUSIONS  v i s u a l - i n p u t system i s a g e n e r a l  graphical input  t o the PDP-9.  The system was d e s i g n e d t o be a d a p t a b l e t o a  PDP-9 computer f o r the purpose o f e v e n t u a l l y but  purpose t h r e e - d i m e n s i o n a l  automating l e u k o c y t e  classification  i t can be used f o r any form o f computerized image p r o c e s s i n g .  a p p l i e d e q u a l l y w e l l t o b i o l o g i c a l image p r o c e s s i n g , picture processing The using  research  recognition,  or a r t i f i c i a l i n t e l l i g e n c e studies.  system can m e a n i n g f u l l y r e s o l v e  e i t h e r a l i n e a r or logarithmic  g r i d points  character  I t can be  density,  A p i c t u r e m a t r i x o f 1024 by 1024  scale.  c a n be randomly a c c e s s e d .  eight l e v e l s of o p t i c a l  O p t i c a l d e n s i t i e s a r e r e a d i n t o the  PDP-9 w i t h an i n t e r r o g a t i o n time dependent on t h e low-pass f i l t e r bandwidth. (An i n t e r r o g a t i o n time o f 100 microseconds i s t y p i c a l ) . p r o c e s s e d from a 23 mm by 23 mm square on a t r a n s p a r e n t  A picture i s slide;  the image  f i e l d can be viewed on an e x t e r n a l l y c o n t r o l l e d d i s p l a y u n i t . The reading  programmer has the o p t i o n  the p i c t u r e i n f o r m a t i o n  with data processing The  of on-line processing  or systematically  onto magnetic tape (D.E.C. tape) and p r o c e e d i n g  from t h a t medium.  image d i s s e c t o r tube a p e r t u r e ,  found t o be n e e d l e s s l y  small.  which was 1 m i l i n diameter, was  I t i s a major f a c t o r c o n t r i b u t i n g to a s m a l l  low-pass f i l t e r bandwidth and thus' a l o n g g r i d p o i n t i n t e r r o g a t i o n time, and i t provided  u n n e c e s s a r y s p a t i a l r e s o l u t i o n , below an i n h e r e n t  4 mils.  The s i m u l a t i o n  focusing  o f l a r g e r a p e r t u r e s showed t h a t an a p e r t u r e l a r g e r  than 4 m i l s proved to be a b e t t e r p i c t u r e t r a n s d u c e r , r e d u c i n g The nuclei,  jitter.  s u c c e s s f u l c l u s t e r i n g , i n measurement space, o f the n e u t r o p h i l  strongly i n d i c a t e s that c l a s s i f i c a t i o n of leukocytes using  functions  l i m i t of  s h o u l d be f u r t h e r i n v e s t i g a t e d .  though, a c u r v a t u r e  Instead  o f Connor's a r e a  curvature operator,  e x t r a c t i o n method o f the type used by L e d l e y s h o u l d be  implemented. that  More i m p o r t a n t ,  the v i s u a l - i n p u t  the  system can  c l a s s i f i c a t i o n of l e u k o c y t e s ,  neutrophil be  classification  s t u d y has  s u c c e s s f u l l y used f o r the  making f e a s i b l e the  shown  morphological  automation of the  differential  b l o o d count. F u t u r e l e u k o c y t e c l a s s i f i c a t i o n work s h o u l d i n c o r p o r a t e features  such as c o l o r , n u c l e a r d e n s i t y ,  and  g r a n u l a r i t y o f the  may  be  the  r e l a t i v e frequency of c e l l  cytoplasm, as w e l l as m o r p h o l o g i c a l shape.  e n c o u n t e r e d i n the  detection  o f f o l d s i n n u c l e i and  also  cytoplasm.  Inconsistent  l e v e l ranges f o r c e l l s of the adjusted.  cell  Problems  between  s t a i n i n g w i l l cause v a r y i n g  same type, and  the  types,  the d i s t i n c t i o n  o f n u c l e i themselves where t h e r e i s a l a c k of g r e y - l e v e l c o n t r a s t n u c l e u s and  descriptive  the  grey-  i l l u m i n a t i o n w i l l have to  be  40  APPENDIX I S p e c i f i c a t i o n s and. Performance of a F i v e m i l A p e r t u r e Image D i s s e c t o r Tube  Since design,  the m o n i t o r i n g  circuit  assume, f o r t h i s d i s c u s s i o n , t h a t the  damental n o i s e , w i t h  aperture  be r e d u c e d by c a r e f u l  l i m i t i n g noise  preprocessing,  proved to be n e e d l e s s l y s m a l l .  of the  can  f a c t o r i s fun-  (5.2).  a s i g n a l - t o - n o i s e r a t i o g i v e n by e q u a t i o n  D u r i n g c o n t o u r - t r a c i n g and  limit  dark n o i s e  I t was  the e x i s t i n g 1 m i l d i a m e t e r  smaller  than the r e s o l u t i o n  f o c u s i n g system of the image d i s s e c t o r tube.  O n l y t h r e e b i t s of g r e y - l e v e l q u a n t i z a t i o n were u s e d and t h u s a signal-to-noise ratio The  s l i g h t l y s m a l l e r than 16:1,  low-pass bandwidth was  adjusted  at maximum I , i s r e q u i r e d .  accordingly.  Then, to improve b o t h the s i g n a l - t o - n o i s e r a t i o and t i o n , a l a r g e r aperture 5 m i l s o r .127 The  mm  is  be purchased.  An  aperture  diameter of  appropriate.  aperture  of an edge and  tube s h o u l d  spatial resolu-  diameter w i l l  signals with  then be  j u s t l a r g e r t h a n the  a depth o f m o d u l a t i o n of 100^  mental g r i d p o i n t s , spaced a t about 5 m i l i n t e r v a l s , w i l l  "best-focus"  between two be e a s i l y  fundadetected.  2 Because the new area,  aperture  j i t t e r w i l l be The  would average the o p t i c a l d e n s i t y o v e r a 19.5  less noticeable.  optical resolution limit  of a c t u a l leukocytes  to 5 m i l s a t an image m a g n i f i c a t i o n of about 500. at  t h i s m a g n i f i c a t i o n w i l l r e s u l t i n no  (The  image emerging from the e y e p i e c e  projected  mil  w i l l be  enlarged  Grid point i n t e r r o g a t i o n  l o s s of meaningful s p a t i a l r e s o l u t i o n .  o f a 500  power microscope can  then  be  d i r e c t l y onto the photocathode o f the image d i s s e c t o r t u b e ) . With a 5 m i l a p e r t u r e ,  by a f a c t o r o f  19«5>  the s i g n a l - t o - n o i s e r a t i o w i l l  be  increased  i f the e x t e r n a l bandwidth remains the same as above.  Then a s i g n a l - t o - n o i s e r a t i o quantization  are  of 64-'. 1 can  achieved  and  6 b i t s of g r e y - l e v e l  meaningful.  Alternately, maintaining e x t e r n a l bandwidth can be i n a smaller  be  3 b i t s of g r e y - l e v e l r e s o l u t i o n ,  extended by a f a c t o r o f 19.5.  low-pass f i l t e r  delay  time and  the  This w i l l r e s u l t  image p r o c e s s i n g  will  be f a s t e r .  APPENDIX I I D e t a i l s o f the Hardware Layout  F i g u r e 2.1  shows a b l o c k diagram o f the l a y o u t o f the i n t e r f a c e  between the image d i s s e c t o r camera and the PDP-9, and the hardware to- produce a d i s p l a y . modules.  The l o g i c a l f u n c t i o n s were w i r e d u s i n g D.E.C. p l u g - i n  The d e t a i l s o f the l o g i c networks which a r e c o n t a i n e d  b l o c k o f F i g u r e 2.1,  required  are g i v e n i n t h i s  PiTL LOGIC  LOGIC  0V--+3V  appendix.  i n each  E N A B L E  POWER  DATA  C L E A R  B»—  (IOT1  DO)  STOP  C L O C K  LOAD  X  D E V I C E S E L E C T I O N (NOT  L I N E S  U S E D )  (IOT1 STOP  D2) C L O C K  LOAD  (NOT  Y  U S E D )  S U B D E V I C E S E L E C T I O N  - | START  L I N E S  A / D  START  CLEAR  IOP  CONV.  C L O C K  S Y S T E M  1  L I N E S  (_MAN. 1  CLEAR  READ I / O  S K I P  A / D  RQ C L E A R  A / D  CONV.  COMPLETE  F i g u r e A2.1  Device c o n t r o l .  A / D  F L A G  D/A INPUT CONTROL (I0T1 DO) STOP CLOCK  (IOT1 D2) STOP CLOCK  MAN. STOP CLOCK MAN. START CLOCK  A H 4"  I  SPEED ' SET  H 1 CLOCK ENABLE  START CLOCK  Figure  A2.2  Mode c o n t r o l .  VARIABLE SPEED CLOCK  CLOCK OUTPUT  D70  <3^  SLOO S L 0 2  0  1 8  •o— L> :=;  .-—.ex  O 10  8 Y  1  2  .9X -  i>  9  TO  Y  'CONVERTERS  I / O  3  BUS  10X  DATA  1 0 Y  L I N E S  11  4  .11X •11Y  12  D / A  4>  5  12X  f>  1 2 Y  =3>  13  DL12  ^  DL13  13X  I  13  Y  DL14  -14Y  15  7  FROM A/D CONVERTER  .14X  14  6  —X —X  ~H>— -0-  —K —X NOT ' U S E D  —X  H>-t>-  —X  —X  —K  DL15  .15X -15Y  16  DL16  .16X -16Y  17  DL17  17X 1 7 Y  READ REQUEST  READ  F i g u r e A 2 . 3 Data t r a n s f e r  A / D  Ul  control.  C L E A R  A / D  C L E A R  F L A G  S Y S T E M  L E V E L  L E V E L  CONVERTER  C O N V E R T E R  D E C - R T L  R T L - D E C  L E V E L  s  CONVERTER  1  R T L - D E C R  0 L E V E L C O N V E R T E R  A / D  C O O T .  R T L - D E C  COMP.  L E V E L  L E V E L  R T L  C O N V E R T E R  CONVERTER-  C L O C K  D E C - R T L  R T L - D E C  L E V E L  s  **i  1  R T L - D E C  L E V E L R  0  CONVERTER  CONVERTER R T L - D E C  1  S A / D  1  L E V E L C O N V E R T E R  DONE  0  C  S T A R T  S  A / D  C  COMPARATOR  0  R T L - D E C  R-2R  C O N V E R S I O N  LADDER V I D E O  S I G N A L  F i g u r e A2.4  A/D  converter.  Y X  R I P P L E  COUNTER  COUNTER TOGGLE  MAN. C L E A R  \  S Y S T E M  s  C L O C K OUTPUT  1  S  T C  1  T  0  C  0  IX  I n  s  s  1  T C  s  1  T  0  C  T  0  C  0  7X  t J  I TO  CLEAR  Y  DATA  R I P P L E  G A T E S  COUNTER  S Y S T E M  COUNTER  T O G G L E  JT  s  s  1  T C  1  T  c  0  s  s  1  T  0  c  1  T  c  0  6Y  0  7Y  J TO  F i g u r e A2.5  Y COUNTER  1  2X  Y  S T O P  DATA  G A T E S  8 B i t X (Y) r i p p l e counter.  X  LOAD  P U L S E  CONTROL  LOAD  C L O C K  OUTPUT  MONO-  >  P U L S E  S T A B L E  CONVERTER  Y  P U L S E  LOAD  L E V E L  ?  r  X  CONVERTER  D A T A  D E C - R T L  C L E A R - S Y S T E M  LOAD  Y  LOAD  P U L S E  X  CONTROL  COUNTER T O G G L E  MONOS T A B L E  L E V E L CONVERTER D E C - R T L  LOAD  Y  P U L S E  CLEAR  LOAD  S Y S T E M  Y  Figure A 2 . 6  X (Y) load pulse c o n t r o l .  LOAD  Y  D A T A  x D/A OUTPUT (Y D/A OUTPUT)  > 3.9K  3-9K  2.2K  2.2K  I DISPLAY X DEFLECTION (DISPLAY Y DEFLECTION)  F i g u r e A2.7  X (Y)  attenuator.  CAMERA X DEFLECTION (CAMERA Y DEFLECTION)  17X  (17Y)  L . C . D E C - R T L  FROM i / O  DATA  16X  L . C .  (16Y)  LINES  D E C - R T L  15X  L . C .  (15Y)  D E C - R T L  14X  (14Y)  L . C . D E C - R T L  13X  (13Y)  " L . C .  10  B I T  B I N A R Y  D E C - R T L R E G I S T E R 12X  L . C .  (12Y)  D E C - R T L  FROM DATA G A T E S  • 11X  (llY)  L . C . D E C - R T L  10X  (IOY)  L . C . D E C - R T L  9X  (9Y)  L . C . D E C - R T L  8X  (8Y)  L . C . D E C - R T L LOAD  X  D A T A  (LOAD  Y  DATA  Figure A 2 . 8  X ( Y ) D / A converter.  WEIGHTED LADDER  X  D / A  O U T P U T  (Y  D / A  OUTPUT)  3»-  FROM X RIPPLE COUNTER  (FROM Y RIPPLE COUNTER)  \ 4X (4Y)  D/A  5X  6X  7X  (5Y)  (6Y)  (7Y)  INPUT CONTROL  0  0  0  0  13X  12X  11X  10X  (131)  (12Y)  ( H Y )  (10Y)  0 9X (9Y)  x  D / A CONVERTER  F i g u r e A 2 . 9 . X ( Y ) data gates.  I  8 X (8Y)  y  V TO  I  (TO  Y  D / A  CONVERTER)  52  APPENDIX I I I The L o g a r i t h m i c A m p l i f i e r  A logarithmic a logarithmic  grey-level  a m p l i f i e r was used as an o p t i o n a l f e a t u r e scale.  A schematic diagram o f the a m p l i f i e r i s  shown below.  ALL AMPLIFIERS  F i g u r e A3.1  to p r o v i d e  NEXUS SQ-lOa  Schematic o f the l o g a r i t h m i c  amplifier.  53 BIBLIOGRAPHY  1.  S t e i n , P.G.,  L i p k i n , L.E.,  Purpose M i c r o s c o p e Input  .and S h a p i r o ,  H.M.  f o r a Computer"  "Spectre  Science,  II:  General-  1966, October 1969,  (328-333). 2.  .Ledley, R.S. Proceeding  3.  "Automatic P a t t e r n R e c o g n i t i o n f o r C l i n i c a l  of the IEEE, 5 7 , 1 1 , November 1 9 6 9 , ( 2 0 1 7 - 2 0 3 5 ) .  Bostrom, R.C.  and Holcomb, W.G.  IEEE I n t e r n a t i o n a l Convention 4.  Mawdesley - Thomas, L.E.  "CYDAC—A D i g i t a l Scanning Cytophotometer" Records, V I I , 1 9 6 3 ,  and Healey, P.  Changes i n H i s t o l o g i c a l S e c t i o n s " 5.  Rosenberg, S.A., and  Medicine"  Ledeen, K.S.,  "Automated A n a l y s i s o f  Science,  1 6 3 , March 1 9 6 9 ,  and K l i n e , T.  Measurement o f C e l l s by Computer"  (110-119).  "Automatic  Science,  Cellular  (1200).  Identification  1 6 3 , March 1 9 6 9 , ( 1 0 6 5 -  1067). 6.  T e c h n i c a l Brochure  7.  T r e t i a k , O.J.  "Picture Processing"  Cambridge, No.  8 3 , October 1 9 6 6 ,  8.  A n n a l s New Young, I.T.  Bourk, T.R.  M.I.T. Q u a r t e r l y P r o g r e s s  R i t a l a , W.M.  Report,  "The  A n a l y s i s of C e l l  Images"  128, 1966, ( 1 0 3 5 - 1 0 5 3 ) .  " B i o l o g i c a l Image P r o c e s s i n g - Automated Leukocyte Report, Cambridge, No.  "Leukocyte L o c a t i n g Procedure"  Report, Cambridge, No. 11.  I n t e r n a t i o n a l Inc.  (129-142).  York Academy of S c i e n c e s ,  M.I.T. Q u a r t e r l y P r o g r e s s 10.  Information  P r e w i t t , J.M.S. and Mendelsohn, M.L. ,  9.  "Computer Eye"  and  with Overlapping  Hsu,  97, A p r i l C.C.  Cells"  89, A p r i l  Recognition"  1968.  M.I.T. Q u a r t e r l y  Progress  1970.  "A F e a t u r e - D e t e c t i o n  Program f o r P a t t e r n s  IEEE T r a n s a c t i o n s , SSC-4, No.  1, March 1 9 6 8 ,  (16-23). 12.  Connor, D.J. Processing"  " L a t e r a l I n h i b i t i o n and Ph.D.  the A r e a Operator  T h e s i s , U.B.C., June 1969-  i n Visual Pattern  13.  Mason, S . J .  a n d Clemens, J.K.  "Character  Reading Machine f o r t h e B l i n d " Eden - ed.,  The M.I.T. P r e s s ,  14.  W i n t r o b e , M.M.  15.  Morphological  Clinical  Eberhardt,  E.H.  Eberhardt,  E.H.  I.T.T. I n d u s t r i a l 18.  Eberhardt,  E.H.  "Noise  Eberhardt,  E.H.  Longhurst, Co.  R.S.  1951.  1970.  R e s e a r c h Memo No. 3 0 9 ,  I960.  i n Image D i s s e c t o r T u b e s "  R e s e a r c h Memo No. 3 3 7 ,  L a b o r a t o r i e s , • I n d i a n a ,1961. " S i g n a l - t o - N o i s e R a t i o i n Image D i s s e c t o r s "  "Off-Focus  Geometrial  L t d . , London, 1964.  Research  L a b o r a t o r i e s , Indiana, 1964.  Beam S i z e i n Image D i s s e c t o r s "  Memo No. 4 2 6 , I . T . T . I n d u s t r i a l 20.  Philadelphia,  D. J o h n s o n , A.R.T., H e a d o f  i nM u l t i p l i e r Phototubes"  Memo No. 3 8 6 , I . T . T . I n d u s t r i a l 19-  (156-167).  H o s p i t a l , V a n c o u v e r , B.C.,  Laboratories, Indiana, "Noise  P a t t e r n s , P.A. K o l e r a n d M.  C a m b r i d g e M a s s . , 1968,  V e r i f i c a t i o n o f Data. Miss  I.T.T. I n d u s t r i a l 17.  Recognizing  Hematology, L e a and F e b i g e r ,  Hematology Dept., S t . P a u l ' s 16.  Recognition i nan Experimental  Research  L a b o r a t o r i e s , Indiana, 1965.  and P h y s i c a l O p t i c s .  Longmans, G r e e n a n d  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0102199/manifest

Comment

Related Items