Applications of artificial intelligence in COVID-19 clinical response measures

In a recent study published in PLOS Digital Health, researchers reviewed existing literature on the use of artificial intelligence (AI) in health care to characterize the AI applications used in the clinical applications during the coronavirus disease 2019 (COVID-19) pandemic, investigate the location, timing, and extent of AI use in healthcare, and examine the United States (U.S.) regulatory approval processes.

Background

Despite the large number of approvals granted by the U.S. Food and Drug Administration (FDA) to AI applications in healthcare in the last six years, the adoption of AI applications in different areas of healthcare has been limited. Furthermore, there is limited information on the development and use of AI applications during the COVID-19 pandemic, unlike the significant and rapid growth in telehealth and vaccine technologies.

While previous reviews have reviewed the potential uses, challenges, and impacts of AI applications for COVID-19 clinical response, many of the reviews found methodological flaws and potential biases in the use of AI applications in clinical practice. A scarcity of reviews provides a comprehensive report on the development, testing, and applications of AI in COVID-19 clinical responses.

About the study

In the present scoping review, the researchers searched academic and grey literature for studies that would provide answers to the following four questions: 1) what are the AI applications used in COVID-19 clinical responses; 2) what are the locations, timelines, and extent of the use of these applications; 3) how do these applications differ from the pre-pandemic healthcare technology, and how stringent is the U.S. FDA approval criteria for these applications; and 4) what is the publicly available evidence recommending the use of AI applications in healthcare?

The study began with consulting healthcare stakeholders such as clinicians, patient advocates, insurers and health system representatives, researchers, public policymakers, industry representatives, and public health officials for recommendations on study design and documents to be included in the review.

Various databases were then searched for literature available after January 2020 on the use of AI in COVID-19 clinical responses, and the AI applications identified from the literature review were examined in detail for additional information on the developer and use.

Applications were included in the review if they met three criteria. First, the application had a patient health-related function as some part of the patient evolution, diagnosis, decision-making, or treatment process. Applications that were part of drug development or biomedical research were not included.

Lab Diagnostics & Automation eBook

Compilation of the top interviews, articles, and news in the last year.

Download a copy today

Secondly, the AI application was directly involved in the COVID-19 clinical response. Applications available through government health websites and medical testing and symptom check sites for even limited periods qualified for the review. Lastly, the application used artificial intelligence, machine learning, or deep learning algorithms.

The extent of AI application use was determined by the number of patients cared for with the use of the AI application in the COVID-19 clinical response. The Organization for Economic Cooperation and Development’s classification of high-, middle-, or low-income countries was used to determine the location of AI application use.

Results

The results reported the use of 66 AI applications in the COVID-19 clinical response, which were grouped into six functional categories. The lung evaluation AI applications assessed computed tomography (CT) scans or X-rays or both, for the presence of pneumonia, pneumothorax, or other lung abnormalities due to COVID-19. Symptom checker AI applications used demographic data, risk factors, and symptoms provided by the patient to calculate COVID-19 risks and provide healthcare recommendations.

Patient deterioration AI applications monitored the vital signs and health conditions of COVID-19 patients and provided information to make healthcare decisions. These were used by clinicians in hospitals, assisted living facilities, and to monitor patients isolated in their homes.

Several applications predicted the likelihood of COVID-19 infections from different sources of information, such as volatile organic compounds in breath, geographically grouped data, patient demographics, blood test results, and luminescent signals from antigen test strips.

Some applications used demographic and medical record data to predict the risk of severe COVID-19 outcomes and were used by clinicians in hospitals, telehealth facilities, and outpatient clinics. Other applications carried out various functions such as image acquisition, response prediction to treatment, immune response detection, etc.

These applications used neural networks, advanced tree-based methods, and supervised and unsupervised machine learning methods. A large number of the applications were deployed between January and June 2020, in the initial phases of the pandemic, and were largely used in high-income countries such as the U.S., and China, with very few being used in middle to low-income countries.

No clinical trials evaluating the use of AI applications in COVID-19 response were found, and the few publications supporting the use of some of the applications were independent evaluations.

Conclusions

In conclusion, the scoping review reported the location, function, potential benefits, scope, type, and input data of AI applications used in clinical practice during the COVID-19 pandemic. While there is a plethora of academic literature discussing AI models, there is a dearth of academic reports on the use of these AI models in clinical practice.

The limited evidence in the literature makes it difficult to determine the benefits of AI use in the pandemic response efforts. Further studies on the real-world applications of AI in healthcare are required.

Written by

Dr. Chinta Sidharthan

Chinta Sidharthan is a writer based in Bangalore, India. Her academic background is in evolutionary biology and genetics, and she has extensive experience in scientific research, teaching, science writing, and herpetology. Chinta holds a Ph.D. in evolutionary biology from the Indian Institute of Science and is passionate about science education, writing, animals, wildlife, and conservation. For her doctoral research, she explored the origins and diversification of blindsnakes in India, as a part of which she did extensive fieldwork in the jungles of southern India. She has received the Canadian Governor General’s bronze medal and Bangalore University gold medal for academic excellence and published her research in high-impact journals.