TY - GEN
T1 - Weakly Supervised Deep Learning for Predicting the Response to Hormonal Treatment of Women with Atypical Endometrial Hyperplasia
T2 - Medical Imaging 2023: Digital and Computational Pathology
AU - Kahaki, Seyed
AU - Hagemann, Ian S.
AU - Cha, Kenny
AU - Trindade, Christopher J.
AU - Petrick, Nicholas
AU - Kostelecky, Nicolas
AU - Chen, Weijie
N1 - Funding Information:
This work has not been submitted for publication or presentation elsewhere. This is a contribution of the U.S. Food and Drug Administration and is not subject to copyright. Python implementation of the methods described in this study will be freely available at https://github.com/mousavikahaki/response_prediction. We thank Dr. Marios Gavrielides for his efforts in developing this project concept and helpful discussions. We thank Dr. Alexej Gossmann for helpful discussions. This work was supported by the FDA Office of Women’s Health. This project was supported in part by an appointment to the ORISE Research Participation Program at the Center for Devices and Radiological Health, U.S. Food and Drug Administration, administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and FDA/CDRH.
Publisher Copyright:
© 2023 SPIE.
PY - 2023
Y1 - 2023
N2 - Endometrial cancer (EC) is the most common gynecologic malignancy in the US and complex atypical hyperplasia (CAH) is considered a high-risk precursor to EC. Treatment options for CAH and early-stage EC include hormone therapies and hysterectomy with the former preferred by certain patients, e.g., for fertility preservation or poor surgical candidates. Accurate prediction of response to hormonal treatment would allow for personalized and potentially improved recommendations for the treatment of these conditions. In this study, we investigate the feasibility of utilizing weakly supervised deep learning models on whole slide images of endometrial tissue samples for the prediction of patient response to hormonal treatment. We curated a clinical whole-slide-image (WSI) dataset of 112 patients from two clinical sites. We developed an end-to-end machine learning model using WSIs of endometrial specimens for the prediction of hormonal treatment response among women with CAH/EC. The model takes patches extracted from pathologist-annotated CAH/EC regions as input and utilizes an unsupervised deep learning architecture (Autoencoder or ResNet50) to embed the images into a low-dimensional space, followed by fully connected layers for binary prediction. Our autoencoder model yielded an AUC of 0.79 with 95% CI [0.61, 0.98] on a hold-out test set in the task of predicting a patient with CAH/EC as a responder vs non-responder to hormonal treatment. Our results, demonstrate the potential for using weakly supervised machine learning models on WSIs for predicting response to hormonal treatment of CAH/EC patients.
AB - Endometrial cancer (EC) is the most common gynecologic malignancy in the US and complex atypical hyperplasia (CAH) is considered a high-risk precursor to EC. Treatment options for CAH and early-stage EC include hormone therapies and hysterectomy with the former preferred by certain patients, e.g., for fertility preservation or poor surgical candidates. Accurate prediction of response to hormonal treatment would allow for personalized and potentially improved recommendations for the treatment of these conditions. In this study, we investigate the feasibility of utilizing weakly supervised deep learning models on whole slide images of endometrial tissue samples for the prediction of patient response to hormonal treatment. We curated a clinical whole-slide-image (WSI) dataset of 112 patients from two clinical sites. We developed an end-to-end machine learning model using WSIs of endometrial specimens for the prediction of hormonal treatment response among women with CAH/EC. The model takes patches extracted from pathologist-annotated CAH/EC regions as input and utilizes an unsupervised deep learning architecture (Autoencoder or ResNet50) to embed the images into a low-dimensional space, followed by fully connected layers for binary prediction. Our autoencoder model yielded an AUC of 0.79 with 95% CI [0.61, 0.98] on a hold-out test set in the task of predicting a patient with CAH/EC as a responder vs non-responder to hormonal treatment. Our results, demonstrate the potential for using weakly supervised machine learning models on WSIs for predicting response to hormonal treatment of CAH/EC patients.
KW - Weakly supervised learning
KW - atypical endometrial hyperplasia
KW - deep learning
KW - hormonal treatment
UR - http://www.scopus.com/inward/record.url?scp=85160565614&partnerID=8YFLogxK
U2 - 10.1117/12.2652912
DO - 10.1117/12.2652912
M3 - Conference contribution
C2 - 37159719
AN - SCOPUS:85160565614
T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE
BT - Medical Imaging 2023
A2 - Tomaszewski, John E.
A2 - Ward, Aaron D.
PB - SPIE
Y2 - 19 February 2023 through 23 February 2023
ER -