Intel will develop AI to accurately detect brain tumors at an early stage. The collaboration of engineers with scientists from the University of Pennsylvania is said in a company statement.
According to the American Association of Brain Tumors, this year brain tumors will be diagnosed in approximately 80 thousand people, more than 4 600 of them in children. To train and build a model for detecting such tumors at an early stage, researchers need access to large volumes of relevant medical data. However, it is important that the data remains confidential and protected.
29 international medical and research institutions led by Penn Medicine will participate in the development and training of the neural network: together they will upload the images they have at their disposal and check how accurately AI identifies brain tumors using confidentiality techniques.
Intel notes that the neural network was created according to the method of federal training – this will allow organizations to conduct training without exchanging patient data. It is planned that the development of AI will be completed before the end of the year, and the expected accuracy of determining brain tumors by the system is 99%.
AI shows great promise for the early detection of brain tumors, but it will need more data to reach its full potential. Using Intel software and hardware, we work with the University of Pennsylvania and a federation of 29 medical centers to promote the identification of brain tumors while protecting sensitive patient data.
Jason Martin, Chief Engineer, Intel Labs