Clinical Trial Recruitment
Finding suitable patients for clinical trials is a time consuming, expensive and inefficent process significantly slowing down drug development. Slow drug development and lenghty regulatory processes block the access to new drugs for patients and lead to increasing healthcare costs. In addition, the current process of recruitment often results in small test populations. This can result in side effects with low incidence not being identified.
Using machine learning, clinical trial sponsors and contract research organizations can use their existing databases to identify eligible patients. By combining multiple sources of data (e.g. different countries), the number of eligible patients grows and suitable people can be identified faster. Further, extended study populations ensure ensure that rare side effects are more likely to be detected.
Modulos AutoML enables training on sensitive, patient data in a fully privacy-preserving manner as it can be deployed locally without the need of a public cloud. Further, all models trained by Modulos AutoML belong to the user. As such they can be shared among contract research organization centers or can be used to generate predictions using other privacy-preserving technologies.