[Press release] JMIR Review: AI technologies playing a key role in the medical response to the COVID-19 pandemic
Zurich, Switzerland (March 12, 2021) Scientists from the Women’s Brain Project have collaborated on a new review that captures the key role played by artificial intelligence (AI) technologies in the medical response to the COVID-19 pandemic. The review, “AI-aided Precision Medicine against COVID-19: Strategic Areas of Research and Development”, was published by the Journal of Medical Internet Research on March 12.
AI-powered healthcare is essential to help public health systems efficiently manage future outbreaks and improve patient outcomes, the authors of the review have found – a collaboration between the Women’s Brain Project and scientists from Harvard Medical School, the Barcelona Supercomputing Center (BSC) and others.
The ongoing SARS-CoV-2 pandemic has highlighted the fragility of healthcare systems to unexpected events, testing the endurance of even the top-performing systems.
In their review of AI-aided healthcare, published by the Journal of Medical Internet Research, the international team of scientists observed that AI technologies are playing a key role in the medical response to the pandemic and will have even greater importance in future pandemics.
The overview of more than 100 recently published studies on the COVID-19 pandemic(s) looks at triage, diagnosis and risk prediction, among other areas.
AI has been applied to determine treatment priorities in COVID-19 patients, or triage, and to better allocate limited resources. It has also been utilized to identify patients at higher risk of mortality and ICU/ventilator need.
For example, researchers at the Tongji Hospital, Wuhan, China, screened Electronic Health Records (EHRs) of 375 discharged patients and trained a model to predict their mortality risk, based on clinical measurements. The accuracy of the system was 93%. Its utilization would make it possible for physicians to immediately identify the critical cases and act accordingly. In addition, AI models have been applied to medical imaging, supporting radiologists in their expert decisions.
“The pandemic has shown that our health care systems are prone to fragility. The use of artificial intelligence to achieve precision medicine is the only possible way to guarantee universal health access by sparing our limited resources and rendering healthcare sustainable,” co-author Dr Antonella Santuccione Chadha, CEO of the Women’s Brain Project (WBP) explains.
The trend is clear. “The COVID-19 crisis is accelerating anticipated changes towards a stronger collaboration between computer science and medicine,” Enrico Santus, senior data scientist at Bayer and AI advisor of WBP adds.
The way in which the AI systems will be exploited is a delicate topic with ethical considerations, in particular regarding the decisional independence of medical staff. However, the review notes, humans are already affected by numerous cognitive biases.
“Translating patient data to successful therapies is the major ambition of implementing AI for health, especially in times of pandemic crisis, with the ultimate goal of achieving a successful bench-to-bedside model for better clinical decision-making,” the authors say.
Dr Santuccione Chadha gives the example of how precision medicine could be used for vaccination campaigns: “We would be able to protect the most susceptible individuals first by analyzing their biological characteristics and assessing their risk profile.”
Sensitizing AI to the diversity of patient populations
The Swiss-based Women’s Brain Project (WBP) has a particular focus on sex and gender differences in diseases. Since the early months of the COVID-19 pandemic, the WBP has been highlighting the clear sex differences in the impact of the virus. Men tend to suffer more severe symptomology and mortality rates, a phenomenon that needs to be taken into account in prevention, treatment and care.
An understanding of sex and gender differences is also crucial for the correct and safe evolution of AI-powered healthcare, which must be based on fair and broad data-collecting systems.
The absence of transparency and interpretability in AI models has long hidden the reality that the efficacy of these technologies is not equal across population groups, the review states.
In particular, COVID-19 incidence and outcomes vary according to a large number of individual factors, including age, sex, ethnicity, health status and drug utilization. Sensitizing AI technologies to the diversity of the patient population and ensuring data security is imperative to avoid biased decisions.
The three other main AI-related areas covered by the review are: drug repurposing and development; pharmacogenomics and vaccines; and mining of medical literature.
For more information, see the article online:
About the Women’s Brain Project: The Women’s Brain Project is an international non-profit organization based in Switzerland and acting on a global level. We are a group of experts from various disciplines who advocate for sex and gender sensitive precision medicine, from basic science to novel technologies.
Media contacts: Co-author and Women’s Brain Project CEO Antonella Santuccione Chadha, firstname.lastname@example.org; WBP Head of Communications Clare O’Dea, email@example.com