Image of tissue sections recorded on the ultrafast laser scanner microscope (ULSM) will be segmented by means of software controlled by an expert systems module, and the subsequent diagnostic assessment will be guided by a separate diagnostic expert system module. Knowledge bases for both of these expert modules will be developed and extensively tested on clinical material from a limited number of body sites. A systematic for the segmentation expert module, now in its early stages of development will be perfected and tested. To assure that the rule based systems have an adequate """"""""understanding"""""""" of the submitted imagery, the required segmentation tasks, and the diagnostic assessment, a validation procedure will be implemented for each problem domain (i.e., tissue sections from a given body site). This project will bring concepts from knowledge engineering to bear on computer analysis of diagnostic imagery and involves the pathologists with their personal diagnostic experience as judges of system performance. The ULSM will continue to be applied to the computer processing of clinical materials involving uncoupled processing tasks such as ploidy determinations in lesions from the colon, prostate, and bladder. Software development for these tasks will continue as needed. The operational software for the high speed fluorescence scanner will be defined by the pathologists as eventual end users, and will be tested on selected clinical material from the prostate.
| Bartels, P H; Thompson, D; Weber, J E (1992) Expert systems in histopathology. IV. The management of uncertainty. Anal Quant Cytol Histol 14:1-13 |
| Bartels, P H (1992) Computer-generated diagnosis and image analysis. An overview. Cancer 69:1636-8 |
| Bartels, P H; Thompson, D; Weber, J E (1992) Expert systems in histopathology. V. DS theory, certainty factors and possibility theory. Anal Quant Cytol Histol 14:165-74 |
| Wied, G L; Bartels, P H; Bibbo, M et al. (1989) Image analysis in quantitative cytopathology and histopathology. Hum Pathol 20:549-71 |
| Bartels, P H (1989) The diagnostic pattern in histopathology. Am J Clin Pathol 91:S7-13 |
| Bartels, P H; Weber, J E; Duckstein, L (1988) Machine learning in quantitative histopathology. Anal Quant Cytol Histol 10:299-306 |
| Graham, A R; Paplanus, S H; Bartels, P H (1988) Micromorphometry of colonic lesions. Lab Invest 59:397-402 |
| Shack, R V; Bartels, P H; Buchroeder, R A et al. (1987) Design for a fast fluorescence laser scanning microscope. Anal Quant Cytol Histol 9:509-20 |
| Paplanus, S H; Graham, A R (1987) Morphometry in surgical pathology. Anal Quant Cytol Histol 9:455-8 |
| Paplanus, S H; Graham, A R; Layton, J M et al. (1985) Statistical histometry in the diagnostic assessment of tissue sections. Anal Quant Cytol 7:32-7 |
