Weakly supervised temporal model for prediction of breast cancer distant recurrence

• Type of Material: Article
• Authors: Josh Sanyal, Stanford University; Amara Tariq, Emory University; Allison W Kurian, Stanford University; Daniel Rubin, Stanford University; Imon Banerjee, Emory University
• Language: English
• Date: 2021-05-04
• Publisher: NATURE RESEARCH
• Publication Version: Final Published Version
• Copyright Statement: © The Author(s) 2021
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1038/s41598-021-89033-6
• Title of Journal or Parent Work: SCIENTIFIC REPORTS
• Volume: 11
• Issue: 1
• Start Page: 9461
• End Page: 9461
• Keywords: Science & Technology - Other Topics; DIAGNOSIS; MEDICARE CLAIMS; Multidisciplinary Sciences; Science & Technology
• Research Categories: Health Sciences, Medicine and Surgery; Health Sciences, Epidemiology

Abstract

Efficient prediction of cancer recurrence in advance may help to recruit high risk breast cancer patients for clinical trial on-time and can guide a proper treatment plan. Several machine learning approaches have been developed for recurrence prediction in previous studies, but most of them use only structured electronic health records and only a small training dataset, with limited success in clinical application. While free-text clinic notes may offer the greatest nuance and detail about a patient’s clinical status, they are largely excluded in previous predictive models due to the increase in processing complexity and need for a complex modeling framework. In this study, we developed a weak-supervision framework for breast cancer recurrence prediction in which we trained a deep learning model on a large sample of free-text clinic notes by utilizing a combination of manually curated labels and NLP-generated non-perfect recurrence labels. The model was trained jointly on manually curated data from 670 patients and NLP-curated data of 8062 patients. It was validated on manually annotated data from 224 patients with recurrence and achieved 0.94 AUROC. This weak supervision approach allowed us to learn from a larger dataset using imperfect labels and ultimately provided greater accuracy compared to a smaller hand-curated dataset, with less manual effort invested in curation.