A look into the future of artificial intelligence in healthcare

In many areas of healthcare, artificial intelligence approaches promise great opportunities and support for patients and doctors. However, before these approaches can be used in clinical practice or financed by health insurance companies, extensive clinical studies must be carried out. The aim of these studies - as with all medical products - is to obtain valid data on the safety and efficacy of the applications in a statistically relevant number of patients. As a result, the development and authorisation process for digitally supported medical products can sometimes take more than ten years and is associated with high (financing) risks.

Number of registered studies with algorithm-based innovations on the rise

Scientists at Witten/Herdecke University (UW/H) have now analysed the world's largest database for the registration of clinical trials with regard to the development and research status of algorithm-based innovations. This showed that the number of registered studies with algorithmic approaches has increased year on year since 2015, with a particularly significant increase in 2019 and 2020. It also became clear that algorithmic approaches are now being developed for a wide range of medical issues and are being brought into the approval process. "In addition to medical imaging, studies on digital approaches are being initiated primarily in cardiology, psychiatry, anaesthesia/intensive care medicine and neurology," says Dr Claus Zippel, lead author of the study, naming the specialist areas particularly addressed.

The scientists also evaluated parameters relating to the organisation and financing of the study. It became clear that although the majority of studies on algorithm-based approaches continue to be initiated by academic institutions, funding is increasingly being provided by industry. "The development of digital health products is sometimes associated with high risks for investors due to regulatory hurdles. The increasing funding from industry therefore suggests that the digital transformation in this area is picking up speed," says Prof Dr Sabine Bohnet-Joschko, commenting on the results. The author of the study is also head of the ATLAS project, which is funded by the Ministry of Economic Affairs of North Rhine-Westphalia and focuses on the digital transformation in the health care system, and in which the results presented are now to be further analysed.

Original publication (Open Access)
Zippel C, Bohnet-Joschko S. Rise of Clinical Studies in the Field of Machine Learning: A Review of Data Registered in ClinicalTrials.gov, International Journal of Environmental Research and Public Health 2021; 18(10): 5072. doi: 10.3390/ijerph18105072.