Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges
Machine learning applications promise to augment clinical capabilities and at least 64 models have already been approved by the US Food and Drug Administration. These tools are developed, shared, and used in an environment in which regulations and market forces remain immature. An importa...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
JMIR Publications
2022-04-01
|
Series: | JMIR Formative Research |
Online Access: | https://formative.jmir.org/2022/4/e33970 |
_version_ | 1797735204831363072 |
---|---|
author | Keerthi B Harish W Nicholson Price Yindalon Aphinyanaphongs |
author_facet | Keerthi B Harish W Nicholson Price Yindalon Aphinyanaphongs |
author_sort | Keerthi B Harish |
collection | DOAJ |
description |
Machine learning applications promise to augment clinical capabilities and at least 64 models have already been approved by the US Food and Drug Administration. These tools are developed, shared, and used in an environment in which regulations and market forces remain immature. An important consideration when evaluating this environment is the introduction of open-source solutions in which innovations are freely shared; such solutions have long been a facet of digital culture. We discuss the feasibility and implications of open-source machine learning in a health care infrastructure built upon proprietary information. The decreased cost of development as compared to drugs and devices, a longstanding culture of open-source products in other industries, and the beginnings of machine learning–friendly regulatory pathways together allow for the development and deployment of open-source machine learning models. Such tools have distinct advantages including enhanced product integrity, customizability, and lower cost, leading to increased access. However, significant questions regarding engineering concerns about implementation infrastructure and model safety, a lack of incentives from intellectual property protection, and nebulous liability rules significantly complicate the ability to develop such open-source models. Ultimately, the reconciliation of open-source machine learning and the proprietary information–driven health care environment requires that policymakers, regulators, and health care organizations actively craft a conducive market in which innovative developers will continue to both work and collaborate. |
first_indexed | 2024-03-12T12:54:39Z |
format | Article |
id | doaj.art-81d6c8258ab34b73bf67c386ec0ed676 |
institution | Directory Open Access Journal |
issn | 2561-326X |
language | English |
last_indexed | 2024-03-12T12:54:39Z |
publishDate | 2022-04-01 |
publisher | JMIR Publications |
record_format | Article |
series | JMIR Formative Research |
spelling | doaj.art-81d6c8258ab34b73bf67c386ec0ed6762023-08-28T21:22:39ZengJMIR PublicationsJMIR Formative Research2561-326X2022-04-0164e3397010.2196/33970Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and ChallengesKeerthi B Harishhttps://orcid.org/0000-0001-9244-7253W Nicholson Pricehttps://orcid.org/0000-0003-0729-290XYindalon Aphinyanaphongshttps://orcid.org/0000-0001-8605-5392 Machine learning applications promise to augment clinical capabilities and at least 64 models have already been approved by the US Food and Drug Administration. These tools are developed, shared, and used in an environment in which regulations and market forces remain immature. An important consideration when evaluating this environment is the introduction of open-source solutions in which innovations are freely shared; such solutions have long been a facet of digital culture. We discuss the feasibility and implications of open-source machine learning in a health care infrastructure built upon proprietary information. The decreased cost of development as compared to drugs and devices, a longstanding culture of open-source products in other industries, and the beginnings of machine learning–friendly regulatory pathways together allow for the development and deployment of open-source machine learning models. Such tools have distinct advantages including enhanced product integrity, customizability, and lower cost, leading to increased access. However, significant questions regarding engineering concerns about implementation infrastructure and model safety, a lack of incentives from intellectual property protection, and nebulous liability rules significantly complicate the ability to develop such open-source models. Ultimately, the reconciliation of open-source machine learning and the proprietary information–driven health care environment requires that policymakers, regulators, and health care organizations actively craft a conducive market in which innovative developers will continue to both work and collaborate.https://formative.jmir.org/2022/4/e33970 |
spellingShingle | Keerthi B Harish W Nicholson Price Yindalon Aphinyanaphongs Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges JMIR Formative Research |
title | Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges |
title_full | Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges |
title_fullStr | Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges |
title_full_unstemmed | Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges |
title_short | Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges |
title_sort | open source clinical machine learning models critical appraisal of feasibility advantages and challenges |
url | https://formative.jmir.org/2022/4/e33970 |
work_keys_str_mv | AT keerthibharish opensourceclinicalmachinelearningmodelscriticalappraisaloffeasibilityadvantagesandchallenges AT wnicholsonprice opensourceclinicalmachinelearningmodelscriticalappraisaloffeasibilityadvantagesandchallenges AT yindalonaphinyanaphongs opensourceclinicalmachinelearningmodelscriticalappraisaloffeasibilityadvantagesandchallenges |