BEIJING, July 17 (Xinhua) -- A Chinese team has developed the country's first versatile artificial intelligence (AI) model that is adept at analyzing a diverse range of pathological images. This advanced model is capable of examining more than 20 human organs, including the lungs, breast and liver.
This large language model (LLM) known as PathOrchestra represents a breakthrough in AI-assisted disease diagnosis, heralding a transformative shift from a singular model dedicated to a specific cancer to a versatile one capable of addressing a multitude of them.
The researchers from Air Force Medical University (AFMU), Tsinghua University and SenseTime leveraged China's largest domestic dataset comprising nearly 300,000 whole-slide digital pathology images, which equates to an impressive 300 terabytes of data.
Harnessing self-supervised learning, the model "cross-learned" to analyze over 20 different organs and has accomplished a slew of clinical tasks, including pan-cancer classification, lesion identification and detection, multi-cancer subtype differentiation and biomarker assessment.
The diversity in pathological images poses a formidable challenge for AI applications, and this complexity has earned it the title of the "jewel in the crown" in the realm of image processing, said Wang Zhe, a professor from the Basic Medical Science Academy under the AFMU.
PathOrchestra has achieved an accuracy rate exceeding 95 percent in nearly 50 clinical tasks, including lymphoma subtype diagnosis and bladder cancer screening, according to an AFMU news release on Tuesday.
This advancement can substantially reduce the workload of pathologists, and notably increased the efficiency of reviewing medical images, said the researchers.
PathOrchestra stands as an example among the burgeoning landscape of large models in China, representing the country's rapid and vibrant growth in the field of AI.
Of the more than 1,300 AI LLMs globally, 36 percent are from China, the second-largest proportion after the United States, according to a white paper on the global digital economy released recently by the China Academy of Information and Communications Technology at the Global Digital Economy Conference 2024. ■
