Synthetic Of Medicines

this graph was created in researchgate :

Generator and Discriminator losses for 50,000 epochs: (a) GAN, (b) CGAN and (c) WGAN. WGAN resulted in a more meaningful loss, as the synthetic data converged with the real data, reducing its loss, and the discriminator not being able to confidently classify data.
Digital health applications can improve quality and effectiveness of healthcare, by offering a number of new tools to users, which are often considered a medical device. Assuring their safe operation requires, amongst others, clinical validation, needing large datasets to test them in realistic clinical scenarios. Access to datasets is challenging.
Medical device (MD) developers currently use the Technology Readiness Levels (TRLs) to determine the status of technology. However, the broad definition of each level makes it difficult to ascertain readiness of MDs. The TRL is also indifferent to the strict regulatory requirements for MDs, requirements for human systems integration, and market competition. This paper proposes a technology readiness framework for communicating and planning the development of Class III MDs called the Medical Device Readiness Levels (MDRLs). Five exit criteria were staged in a meaningful sequence within the framework: safety, clinical effectiveness, usability, comfort, and affective response. It also incorporates the stakeholders’ perspectives, mindful of the users’ varying needs in manipulating the MD. The usefulness of the framework was affirmed by professionals involved in MD development.