Abstract
ABSTRACT Unlike the eye, automated microscopy workstations with image processing and continuous class pattern recognition capabilities can consistently and precisely subtype lymph node cancers (follicular lymphomas). In contrast, pathologists agree with their own subtype less than 70% of the time, and exhibit major disagreements more than 35% of the time. As a result, over 10,000 of the 41,000 new follicular lymphoma cases estimated for 1992 may be incorrectly subtyped, and therefore subject to inappropriatetreatment. A Coulter diff3/50 Research microscope under computer control was used to digitize biopsyslides at differing resolutions and in full color. Then, a broad set of candidate features were extractedusing color analysis, template matching, statistical texture analysis, frequency domain techniques, andsurface modeling by both cellular logic filters and relative extrema analysis. In all, over 600 candidate features were measured for selection and classifier design. Standard pattern. recognition techniques for classifier design assume that objects cluster into
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.