Artificial Intelligence for Advance Requesting of Immunohistochemistry in Diagnostically Uncertain Prostate Biopsies
Authors:
- Andrea ChatrianACAndrea Chatrian
- Richard T CollingRCRichard T Colling
- Lisa BrowningLBLisa Browning
- Nasullah Khalid AlhamNANasullah Khalid Alham
- Korsuk SirinukunwattanaKSKorsuk Sirinukunwattana
- Maryam HaghighatMHMaryam Haghighat
- Alan AberdeenAAAlan Aberdeen
- Benjamin Moxley-WylesBMBenjamin Moxley-Wyles
- Emad RakhaEREmad Rakha
- David R J SneadDSDavid R J Snead
Preprint
Abstract
The use of immunohistochemistry in the reporting of prostate biopsies is an important adjunct when the diagnosis is not definite on haematoxylin and eosin (H&E) morphology alone. The process is however inherently inefficient with delays while waiting for pathologist review to make the request and duplicated effort reviewing a case more than once. In this study, we aimed to capture the workflow implications of immunohistochemistry requests and demonstrate a novel artificial intelligence tool to identify cases in which immunohistochemistry (IHC) is required and generate an automated request. We conducted audits of the workflow for prostate biopsies in order to understand the potential implications of automated immunohistochemistry requesting and collected prospective cases to train a deep neural network algorithm to detect tissue regions that presented ambiguous morphology on whole slide images. These ambiguous foci were selected on the basis of the pathologist requesting immunohistochemistry to aid diagnosis. A gradient boosted trees classifier was then used to make a slide level prediction based on the outputs of the neural network prediction. The algorithm was trained on annotations of 219 immunohistochemistry-requested and 80 control images, and tested by 3-fold cross-validation. Validation was conducted on a separate validation dataset of 212 images. Non IHC-requested cases were diagnosed in 17.9 minutes on average, while IHC-requested cases took 33.4 minutes over multiple reporting sessions. We estimated 11 minutes could be saved on average per case by automated IHC requesting, by removing duplication of effort. The tool attained 99% accuracy and 0.99 Area Under the Curve (AUC) on the test data. In the validation, the average agreement with pathologists was 0.81, with a mean AUC of 0.80. We demonstrate the proof-of-principle that an AI tool making automated immunohistochemistry requests could create a significantly leaner workflow and result in pathologist time savings.Keywords
All Related Versions
- Published version: Laboratory Investigation, 34 (9), 1780.
This publication has 24 references indexed in Scilit:
- Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic studyThe Lancet Oncology, 2020
- Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic studyThe Lancet Oncology, 2020
- Development and validation of a deep learning algorithm for improving Gleason scoring of prostate cancernpj Digital Medicine, 2019
- Artificial intelligence in digital pathology: a roadmap to routine use in clinical practiceThe Journal of Pathology, 2019
- Digital pathology and artificial intelligenceThe Lancet Oncology, 2019
- Automated Gleason grading of prostate cancer tissue microarrays via deep learningScientific Reports, 2018
- The 2016 WHO Classification of Tumours of the Urinary System and Male Genital Organs—Part B: Prostate and Bladder TumoursEuropean Urology, 2016
- Statistical shape model for manifold regularization: Gleason grading of prostate histologyComputer Vision and Image Understanding, 2013
- Complications After Prostate Biopsy: Data From SEER-MedicareJournal of Urology, 2011
- Automated Gland Segmentation and Classification for Gleason Grading of Prostate Tissue Images2010