Researchers from the Technical University of Munich (TUM) and the University Hospital Bonn (UKB) have developed a federated AI-based algorithm that has the capacity to independently learn across various medical institutions. The key feature of this algorithm is that it does not require time-consuming findings in the MRI images by radiologists. In other words, it is self-learning while also being compliant with data protection. The algorithm analysed MRI scans of almost 500 patients for the diagnosis of diseases such as multiple sclerosis and other forms of brain tumours that the algorithm was not subjected to. This is a breakthrough for scientists who intend to develop AI-powered algorithms that are capable of learning while also protecting privacy.
Read More from itnOnline