Abstract
We introduce PathoVA, a visual analytics system for computer aided pathology diagnosis. The diagnostic work of a pathologist involves characterizing cells and the appearance of histological sections of tissues, which are stained by techniques that highlight particular elements. Our system supports this work by recording activities in a digital tissue slide viewer, from which a diagnostic trace is constructed automatically. The visualization of the trace is enhanced with quantitative data about the tissue obtained by image analysis. Using the trace visualization, the pathologist can fill out the pathology report according to the required protocol. We demonstrate our approach on a use case of breast tissue examination. Our qualitative evaluation shows that PathoVA supports the diagnostic procedure and simplifies reporting.
Original language | English |
---|---|
Title of host publication | 2017 IEEE Workshop on Visual Analytics in Healthcare, VAHC 2017 |
Place of Publication | Piscataway |
Publisher | Institute of Electrical and Electronics Engineers |
Pages | 77-83 |
Number of pages | 7 |
ISBN (Electronic) | 9781538631874 |
DOIs | |
Publication status | Published - 15 Jun 2018 |
Event | 8th Annual IEEE Workshop on Visual Analytics in Healthcare, VAHC 2017 - Phoenix, United States Duration: 1 Oct 2017 → … |
Conference
Conference | 8th Annual IEEE Workshop on Visual Analytics in Healthcare, VAHC 2017 |
---|---|
Country/Territory | United States |
City | Phoenix |
Period | 1/10/17 → … |
Funding
The authors wish to thank Dr. Mitko Veta, from Eindhoven University of Technology, for the provided mitosis detection algorithm and Dr. Peet Nooijen from Jeroen Bosch Ziekenhuis for his time and the precious explanation regarding pathology practice. We thank Dirk Verhagen, from Philips Digital Pathology, for his feedback on tool usability and the different scenarios. This research was performed within the framework of the strategic joint research program on Data Science between TU/e and Philips Electronics Nederland B.V.
Keywords
- breast-cancer
- computational pathology
- computer assisted diagnosis
- Digital pathology
- Nottingham Score
- pathology informatics
- visual analytics
- workflow