2026-10-13T08:21:58Z https://www.scidb.cn/oai

10.57760/sciencedb.43190 2026-07-28T14:37:16Z

2026-07-28 GIST AI-organoid research: current challenges and future directions doi:10.57760/sciencedb.43190 en This figure focuses on GISTAI-organoid research, and systematically presents the core challenges currently faced at the data level (lack of multimodal data, missing labels, insufficient dimensions and domain deviation), algorithm level (poor generalization ability and weak interpretation) and clinical translation level (insufficient evidence, drug susceptibility testing cycle of 2 to 4 weeks, and unclear regulatory approval path). Meanwhile, it proposes the future directions such as multi-center collaboration to build high-quality datasets, development of interpretable AI (such as attention mechanism visualization), building a joint platform of "biopsy → organoid culture →AI real-time analysis → clinical decision making", conducting real-world research to accumulate evidence of endpoint indicators, and promoting scientific supervision and product registration. The aim is to promote the effective transformation of research results into clinical practice, so as to improve the overall efficiency of diagnosis and treatment and the long-term benefits of patients. Gastrointestinal stromal tumors; Artificial intelligence; Organoids; Drug susceptibility testing; Clinical translation; Scientific supervision You Junzhao Tang Hongtao Jiang Shun Quan Zhengchao Zhang Yimou Hou Nengyi Zhou Chuan Pang Minghui PUBLIC https://creativecommons.org/licenses/by-nc-nd/4.0/ dataset Science Data Bank