Weakly supervised temporal model for prediction of breast cancer distant recurrence

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...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Josh Sanyal, Amara Tariq, Allison W. Kurian, Daniel Rubin, Imon Banerjee
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/621d9a9af23c4bb7b2dde427d5dfbe39
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:621d9a9af23c4bb7b2dde427d5dfbe39
record_format dspace
spelling oai:doaj.org-article:621d9a9af23c4bb7b2dde427d5dfbe392021-12-02T14:49:26ZWeakly supervised temporal model for prediction of breast cancer distant recurrence10.1038/s41598-021-89033-62045-2322https://doaj.org/article/621d9a9af23c4bb7b2dde427d5dfbe392021-05-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-89033-6https://doaj.org/toc/2045-2322Abstract 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.Josh SanyalAmara TariqAllison W. KurianDaniel RubinImon BanerjeeNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Josh Sanyal
Amara Tariq
Allison W. Kurian
Daniel Rubin
Imon Banerjee
Weakly supervised temporal model for prediction of breast cancer distant recurrence
description 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.
format article
author Josh Sanyal
Amara Tariq
Allison W. Kurian
Daniel Rubin
Imon Banerjee
author_facet Josh Sanyal
Amara Tariq
Allison W. Kurian
Daniel Rubin
Imon Banerjee
author_sort Josh Sanyal
title Weakly supervised temporal model for prediction of breast cancer distant recurrence
title_short Weakly supervised temporal model for prediction of breast cancer distant recurrence
title_full Weakly supervised temporal model for prediction of breast cancer distant recurrence
title_fullStr Weakly supervised temporal model for prediction of breast cancer distant recurrence
title_full_unstemmed Weakly supervised temporal model for prediction of breast cancer distant recurrence
title_sort weakly supervised temporal model for prediction of breast cancer distant recurrence
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/621d9a9af23c4bb7b2dde427d5dfbe39
work_keys_str_mv AT joshsanyal weaklysupervisedtemporalmodelforpredictionofbreastcancerdistantrecurrence
AT amaratariq weaklysupervisedtemporalmodelforpredictionofbreastcancerdistantrecurrence
AT allisonwkurian weaklysupervisedtemporalmodelforpredictionofbreastcancerdistantrecurrence
AT danielrubin weaklysupervisedtemporalmodelforpredictionofbreastcancerdistantrecurrence
AT imonbanerjee weaklysupervisedtemporalmodelforpredictionofbreastcancerdistantrecurrence
_version_ 1718389448078524416