ibul 2004 
the golgi 
ures" arc 
med into 
As the 
‘ent more 
r tof the 
mance of 
nt by the 
yutions of 
features; 
003): 
value is 
>s of the 
fulfilling 
s a four- 
MURPHY, 
sisting of 
' or equal 
1e image 
(1) 
1v, 2001). 
(2) 
(row and 
ce matrix, 
ccurrence 
11... 1973). 
(3) 
AS 
h a line 
istrate the 
- proteins 
tion, Fig. 
atures. 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5 Istanbul 2004 
87 % of the test data samples were correctly classified (c.f. 
STECKLING et al., 2003). In contrast to this several individuals 
who were first shown the training data and who consecutively 
classified the test data achieved classification accuracies of 95 
to 100 95. On the one hand, this result shows that the automatic 
classification procedure still has to be improved. On the other 
hand, the high classification rate of test persons who did not go 
through intensive training procedures indicates that it should be 
possible to reach this goal by an automatic procedure. 
  
white pixels 
  
  
Fig. 5: Feature space (reduced to three dimensions). 
The robustness of the classification primarily depends on the 
statistic. behaviour of the feature vector which is not only 
determined by the visual appearance of the proteins, i.e. the 
differences between the spatial structures of individual cells, 
but also by the variations caused by the chemical preparation of 
the cells and the conditions under which the imagery was 
acquired. Therefore, successful application of the method 
proposed here requires well-controlled laboratory procedures. 
5. REFERENCES 
BECKER-CARUS, C. (1981): Grundrifi der Physiologischen 
Psychologie, HEIDELBERG. 
BOLAND, MICHAEL V.; MARKEY, M.K.; MURPHY, ROBERT F.; 
(1997): Classification of Protein Localization Patterns Obtained 
via Fluorescence Light Microscopy, Proceedings of the 19th 
Annual International Conference of the IEEE Engineering in 
Medicine and Biology Society, 1997, pp. 594-597. 
BOLAND, MICHAEL V; MARKEY, M. K.; MURPHY, ROBERT F.; 
(1998): Automated Recognition of Patterns Characteristic of 
      
  
  
  
   
    
     
   
      
   
  
   
  
  
   
  
  
   
  
   
   
   
  
  
   
  
   
  
  
  
    
  
  
  
  
     
     
  
  
   
    
  
  
  
  
    
  
     
  
  
   
  
   
  
   
     
Subcellular Structures in Fluorescence Microscopy Images. 
Cytometry 33: 366-375, 1998. 
BOLAND, MICHAEL V.; MURPHY, ROBERT F.; (2001): A Neural 
Network Classifier Capable of Recognizing the Patterns of all 
Major Subcellular Structures in Fluorescence Microscope 
Images of HeLa Cells, Bioinformatics 17, 1213-1223, 2001. 
DANCKAERT, A.; (GONZALEZ-COUTO, E.; BOLLONDL L.; 
THOMPSON, N.; HAYES, B.; (2002): Automated Recognition of 
Intracellular Organclles in Confocal Microscope Images, 
Traffic, vol. 3, no 1, pp. 66-73, January 2002. 
HARALICK, R. M.; SHANMUGAN, R.; DINSTEIN, I; (1973): 
Textural Features for Image Classification, IEEE Trans Sys. 
Man Cyb., vol. SMC-3, no. 6, pp. 610-621, 1973. 
MARKEY, M. K.; BOLAND, MICHAEL V; MURPHY, ROBERT F.; 
(1999): Towards Objective Selection of Representative 
Microscope Images. Biophys. J. 76:2230-2237, 1999. 
MURPHY, ROBERT F.; BOLAND, MICHAEL V.; VELLISTE, MEEL; 
(2000). Towards a Systematics for Protein Subcellular 
Location: Quantitative Description of Protein Localization 
Patterns and Automated Analysis of Fluorescence Microscope 
Images. Proc Int Conf Intell Syst Mol Biol (ISMB 2000) 8: 
251-259, 2000. 
MURPHY, ROBERT F.; VELLISTE, MECEL; YAO, JIE; PORRECA, 
GREGORY; (2001): Searching Online Journals for Fluorescence 
Microscope Images Depicting Protein Subcellular Location 
Patterns, Proceedings of the 2nd IEEE International Symposium 
on Bio-Informatics and Biomedical Engineering (BIBE 2001), 
pp. 119-128. 
MURPHY, ROBERT F.; VELLISTE, MEEL; PORRECA, GREGORY; 
(2002): Robust classification of subcellular location patterns in 
fluorescence microscope images, Proseedings of the 2002 IEEE 
International Workshop on Neural Networks for Signal 
Processing (NNSP 12), , pp. 67-76, Proceedings of the jo" 
IEEE Workshop on 4.-6. September 2002. 
RoouES, E.JS.; MURPHY, RoBERT F.; (2002) Objective 
evaluation of differences in protein subcellular distribution. 
Traffic 3: 61-65, 2002. 
STECKLING, TANIA; KLOTZER, HARTMUT; (2003): 
Objekterkennung und Modellierung zellulärer Strukturen aus 
mikroskopischen Bildern; Diplomarbeit, Technische Universität 
Berlin, 2003. 
STECKLING, TANJA; KLÔTZER, HARTMUT, SUTHAU, TIM: 
WALTER, STEPHANIE; WANKER, ERICH; HELLWICH, OLAF; 
(2003): Objekterkennung und Modellierung — zellulárer 
Strukturen aus mikroskopischen Bildern; 23. Jahrestagung der 
DGPF, Bochum 2003. 
VELLISTE, MEEL; MURPHY, ROBERT F; (2002) Automated 
Determination of Protein Subcellular Locations from 3D 
Fluorescence Microscope Images. Proceedings of the 2002 
[EEE International Symposium on Biomedical Imaging (ISBI 
2002), pp. 867-870.