Medical data sharing with ethics? Stanford unveils framework
Molecular and clinical data are vital to power predictive modeling in the development of new drugs and, essentially, vaccines.
While predictive modeling increases the probability of identifying compatible molecules to develop new drugs, high-quality data sets are required to ensure efficiency.
Especially amid the pandemic, companies, universities, and organizations are ramping up efforts to develop, trial, and deploy a vaccine.
On Monday, Johnson & Johnson announced it hopes to get the green light for a new coronavirus vaccine to be distributed globally early next year.
The US giant has selected a lead candidate vaccine after months of research and, by September, aims to begin human clinical studies. Clinical data of the studied vaccine would contribute greatly to the literature of science and health.
Big tech in healthcare
Though pursuing a noble cause, news of tech giants racing to get a share of clinical data tend to raise discomfort and skepticism of the applications of data.
As an effort to set foot into the healthcare industry, the partnership enables the tech giant to access a trove of medical data, incusing birth dates, diagnoses, lab results, and many more. The mass datasets are used to enhance and upgrade Google’s DeepMind and hatch new capabilities for the AI-powered tech.
With that said, a team of radiologists from Stanford University, School of Medicine, has proposed “an ethical framework for using and sharing clinical data for the development of artificial intelligence (AI) applications,” as stated in its research abstract.
The new framework stems from core principles such as patient privacy, transparency, and an obligation to not capitalize on clinical data, the team explained in Radiology.
“Now that we have electronic access to clinical data and the data processing tools, we can dramatically accelerate our ability to gain understanding and develop new applications that can benefit patients and populations,” David B. Larson, MD, said in a press release.
“But unsettled questions regarding the ethical use of the data often preclude the sharing of that information.”
Presently, the idea of sharing clinical data requires either the patient or the institution’s consent. However, Larson and his colleagues are suggesting a third option — when it comes to secondary use, the data does not belong to any party in the traditional sense.
Instead, the framework proposes a distillation of clinical data, in which data goes through processes of re-identification and clustering, suited for research and development. Meanwhile, those wishing to access and use the data are subjected to ethical data practices and must be transparent of their identities.
In essence, the decentralization of data ownership from patients, healthcare providers, and corporations that gain access to it, is proposed as a protective measurement.
The stripping of identifying information from data adds a protective layer on patient privacy and disrupts profit-driven enterprises to mass-target potential patients for products and services. At the same time, it does not deter medical institutions from accessing valuable data sets aimed at research and development.
In a bid to call for collective efforts in establishing ethical data practices, the framework will be released to the public for potential stakeholders to examine.
“We hope this framework will contribute to more productive dialogue, both in the field of medicine and computer science, as well as with policymakers, as we work to thoughtfully translate ethical considerations into regulatory and legal requirements,” Larson stated.
2 December 2022
1 December 2022