While AI is rapidly being incorporated into oncologic research, work remains to be done to translate these studies into real-world, clinically meaningful applications. One of the biggest barriers is in external validation and proving the generalizability of DL applications. Given the complexity of neural networks and the extremely large number of parameters (often in the millions), there is a high tendency for neural networks to create overfitted models that do not generalize across different populations. Additionally, because there is a significant amount of heterogeneity of medical data across institutions, multiple external validation sets may be required to prove the performance of an application. If we can address these challenges, AI has the potential to transform oncology, harnessing the power of big data to drive cancer care into the 21st century and beyond.
Jonathan Yao has received the Ph.D from Washington University in St. Louis in 1990. He has been a visiting professor of Washington University in St. Louis, and a professor at Wuhan University. Jonathan supervised more than one hundred MS students and over one dozen Ph.D. students, currently worked as the CEO of LinaTech LLC. Over the years, he has obtained over 50 patents, published more than 40 papers, lead the development of 3D precision radiotherapy dose calculation algorithm, IMRT algorithm, based on AI and GPU-accelerated Monte-Carlo photon dose calculation, multi-leaf collimator and MV Image guidance system, which has been used in over 600 clinics and hospitals.