Objective To evaluate the applicability and guideline concordance of the Chinese Society of Clinical Oncology(CSCO)arti-ficial intelligence(AI)system in clinical decision-making for colorectal cancer(CRC)patients,and to explore its feasibility in real-world clinical applications.Methods A total of 972 CRC patients diagnosed and treated at the Second Affiliated Hospital,Zhejiang University School of Medicine,from January 2010 to December 2021,were included.Patient data were analyzed by the CSCO AI system to gener-ate treatment decisions,and decision concordance was assessed by a blinded independent central review(BICR)panel.The applicability and guideline concordance rates of the CSCO AI system were calculated for different treatment stages,and a logistic regression model was used to analyze factors influencing the system's decision discrepancies with actual treatments.Results The overall applicability rate of the CSCO AI system was 96.2%,and the overall guideline concordance rate was 94.9%.In the adjuvant and palliative treatment stages,the system's applicability rates were 95.8%and 96.7%,respectively,and the guideline concordance rates were 95.0%and 94.9%,respective-ly.Multivariate logistic regression analysis showed that age≥65 years and high-risk stage Ⅱ treatment were significant factors affecting guideline concordance in the adjuvant treatment stage(both P<0.05).Conclusions The CSCO AI system demonstrated high applicability and guideline concordance in the adjuvant and palliative treatment stages for CRC.The system's clinical decision-making potential is sig-nificant,and it can be further optimized for specific clinical scenarios and promoted for use across various medical institutions.

Objective To evaluate the applicability and guideline concordance of the Chinese Society of Clinical Oncology(CSCO)arti-ficial intelligence(AI)system in clinical decision-making for colorectal cancer(CRC)patients,and to explore its feasibility in real-world clinical applications.Methods A total of 972 CRC patients diagnosed and treated at the Second Affiliated Hospital,Zhejiang University School of Medicine,from January 2010 to December 2021,were included.Patient data were analyzed by the CSCO AI system to gener-ate treatment decisions,and decision concordance was assessed by a blinded independent central review(BICR)panel.The applicability and guideline concordance rates of the CSCO AI system were calculated for different treatment stages,and a logistic regression model was used to analyze factors influencing the system's decision discrepancies with actual treatments.Results The overall applicability rate of the CSCO AI system was 96.2%,and the overall guideline concordance rate was 94.9%.In the adjuvant and palliative treatment stages,the system's applicability rates were 95.8%and 96.7%,respectively,and the guideline concordance rates were 95.0%and 94.9%,respective-ly.Multivariate logistic regression analysis showed that age≥65 years and high-risk stage Ⅱ treatment were significant factors affecting guideline concordance in the adjuvant treatment stage(both P<0.05).Conclusions The CSCO AI system demonstrated high applicability and guideline concordance in the adjuvant and palliative treatment stages for CRC.The system's clinical decision-making potential is sig-nificant,and it can be further optimized for specific clinical scenarios and promoted for use across various medical institutions.