Histopathology Slide Indexing and Search — Are We There Yet?

The search and retrieval of digital histopathology slides are important tasks that have yet to be solved. In this case study, we investigate the clinical readiness of four state-of-the-art histopathology slide search engines — Yottixel ("one yotta pixel"), SISH (self-supervised image search for histology), HSHR (High-Order Correlation-Guided Self-Supervised Hashing-Encoding Retrieval), and RetCCL (Retrieval with Clustering-Guided Contrastive Learning) — on both unseen datasets and several patient cases. We provide a qualitative and quantitative assessment of each model's performance in providing retrieval results that are reliable and useful to pathologists. We found high levels of performance across all models using conventional metrics for tissue and subtyping search. Upon testing the models on real patient cases, we found that the results were still less than ideal for clinical use. On the basis of our findings, we propose a minimal set of requirements to further advance the development of accurate and reliable histopathology image search engines for successful clinical adoption. (Funded by The University of Texas Rising STARs [Science and Technology Acquisition and Retention] Program and The Cancer Prevention & Research Institute of Texas.)