Fujitsu has made waves in the tech world with the announcement of their groundbreaking explainable AI technology. This innovation harnesses data in various formats, such as text, images, and numerical data, to construct knowledge graphs. These graphs enable users to extract meaningful insights from vast datasets with remarkable precision, particularly in medical applications. One of the key strengths of Fujitsu’s technology lies in its ability to process complex medical data.
Through rigorous testing in fields like lung cancer type classification and breast cancer patient survival prediction, Fujitsu has demonstrated the effectiveness of their approach. For instance, their technology accurately identifies different lung cancer types by analyzing visual cues, shedding light on the underlying factors that drive pathological classification.
Moreover, Fujitsu has developed a pioneering algorithm that can extract and train on the unique features of images depicting objects in vastly different ways, leading to highly accurate assessments. This capability holds significant promise for training AI to support precise evaluations from pathological images, even when training data is limited. By the end of fiscal 2024, Fujitsu aims to make these technologies accessible through the Fujitsu Research Portal, allowing users to quickly test and leverage their advanced solutions.