摘要 : Meaningful integration of artificial intelligence (AI) will transform the application of ‘‘big data’’ for patient care, diagnosis, and research. In this issue of Cancer Cell, Chen et al. describe a transparent system to integrate ... 展开 Meaningful integration of artificial intelligence (AI) will transform the application of ‘‘big data’’ for patient care, diagnosis, and research. In this issue of Cancer Cell, Chen et al. describe a transparent system to integrate histopathology and molecular data to predict outcomes and identify novel biomarkers in cancer. In the past decade, artificial intelligence (AI) approaches have achieved preeminence in data analysis from large, complex datasets such as digital images, e.g., radiology, digitized pathology slides (whole-slide images [WSIs]), and molecular data. Methods combining molecular and image data are not in common use, and as technology has improved, there is increased interest in extracting features from the data-rich material of archival pathology tissue slides. Early AI applications in histopathology included relatively simple supervised tasks such as counting of positive immunohistochemical stained cells to create accurate proliferation indices for cancer classification and grading (Kolles et al., 1995; Saha et al., 2017). More sophisticated models allow recognition of histologic patterns with known prognostic impact, e.g., Gleason score in prostate adenocarcinoma (Bulten et al., 2022). With larger datasets and improved learning algorithms, AI can Elsevier Inc. 收起