Computer assisted pathology – past experiences and future prospects
Visual examination of histopathology specimens is an essential component of medical diagnosis; cancer malignancy grade is always determined by visual inspection of biopsy material. Ever since the advent of digital computers, there have been attempts to use digital image analysis as a supplement to visual examination. But despite promising research in computer assisted analysis, clinical histopathology has been slow to adopt the results. With the introduction of slide scanners, one impediment to the adoption of computer assisted methods is removed; digital slides have paved the way for computer assisted analysis of the histopathology specimens. In this presentation I will outline the historical developments of digital pathology and discuss some key aspects of histopathology image analysis, such as the utilization of spectral information provided by stains for image segmentation, classification of tissue components, interpretation of different architectural patterns, and classification or grading of the specimens according to established standards or related to other clinically relevant end-points. The presentations will mainly be illustrated by examples from the author’s own experiences during the past 40 years, with examples of the analysis of most common cancers, e.g., prostate, cervical and breast. The talk will also make some comparisons of similarities and differences between image analysis in cytology and histopathology.
Ewert Bengtsson received his MSc and PhD in engineering physics from Uppsala University in 1974 and 1977 respectively. He continued the work on cell image analysis from his thesis first as a researcher at Uppsala University and then in spin off companies from 1983 through 1988. Then he returned to Uppsala University as an adjunct professor and took the initiative of establishing the Centre for image analysis where he since 1995 is active as a full professor. His research interests include all kinds of biomedical image analysis but his main focus is on cancer detection and grading. He has published about 200 papers in the field and supervised more than 30 PhD students. He has served as Dean of mathematics and computer science and as Vice-rektor for information technology for Uppsala University. He is member of two academies, the Royal Society of Sciences at Uppsala and the Royal Academy of Engineering Sciences.