Health Economic Evaluation of Artificial Intelligence-Assisted Film Reading for Early-Stage Lung Cancer Screening

Xu Huanhuan; Xiao Yue; Shi Liwei; Wu Di; Qiu Yingpeng; Fu Wenqi

Abstract：Objective:To evaluate the cost-effectiveness of Artificial Intelligence(AI)-assisted physician image interpretation in the screening of early-stage lung cancer(stage Ⅰ)from the perspective of the healthcare system,so as to provide evidence for screening strategy optimization.Methods:Based on community populations,a decision tree model was constructed to simulate the cost-effectiveness of two screening strategies:Al-assisted physician image interpretation and independent physician image interpretation,and the Incremental Cost-Effectiveness Ratio(ICER)was calculated.Results:In the basic analysis,the per capita costs of the AI-assisted group and the physician group were 1 483 yuan and 1 489 yuan,respectively,and the effectiveness was 17.02 Quality-Adjusted Life Years(QALYs)and 16.99 QALYs,respectively.Compared with the physician group,the AI-assisted group saved 6 yuan per case and obtained an additional 0.03 QALY per case.The ICER was negative,indicating that the AI-assisted group had an absolute advantage.Threshold analysis showed that when the inspection cost of Al-assisted physician image interpretation increased to 428 yuan per case,the average cost per case was the same between these two groups,and the ICER was 0.When the inspection cost of Al-assisted physician image interpretation was above 428 yuan,the ICER was positive,still demonstrating economic efficiency.Conclusion:AI-assisted image interpretation is cost-effective in the screening of early-stage lung cancer and can facilitate the"early detection,early diagnosis,and early treatment"of lung cancer based on improving the efficiency and accuracy of screening,so as to provide scientific support for health system resource optimization.