会议号:986424228
时间:2021.04.20 下午13:00-14:00
题目:Cost-effectiveness analysis of cervical cancer screening by artificial intelligence-assisted liquid-based cytology diagnosis
基于人工智能诊断筛查宫颈癌的成本效果分析
Background: In November 2020, the World Health Organization (WHO) launched the Global Strategy to Accelerate the Elimination of Cervical Cancer’. Artificial intelligence (AI)-assisted liquid-based cytology (LBC) system based on deep learning algorithms for pathological images reading holds promise for prompt scale-up of cervical cancer screening. We aimed to evaluate the cost-effectiveness of AI-assisted LBC, compared with the manual LBC and human papillomavirus (HPV) DNA testing for cervical cancer screening in China.
Methods: A Markov model was developed for a cohort of 100,000 girls aged 30 years over a lifetime to simulate the natural history of cervical cancer progression. The incremental cost-effectiveness ratios (ICER) of 18 screening strategies (a combination of the three screening methods at six different frequencies, i.e., once per lifetime, twice per lifetime, once every 10 years, once every 5 years, once every 3 years, and once every year) were simulated from a health care system perspective. The willingness-to-pay threshold (US$30,828) was chosen as three times the Chinese per-capita gross domestic product in 2019. Sensitivity analyses were performed to examine the robustness of the results.
Findings: Compared with no screening, screening once or twice per lifetime by AI-assisted LBC were cost-saving, and the rest of the 16 combined screening strategies were cost-effective with an ICER of $156-19,334 per quality-adjusted life-year (QALY) gained. Screening once every 5 years using AI-assisted cytology would be the most cost-effective strategy with an ICER of $8,003 per QALY gained compared with the lower-cost non-dominated strategy on the cost-effectiveness frontier. Its probability of being cost-effective was 54.4% and outperformed other strategies. Sensitivity analyses showed that the most cost-effective strategy might become screening once every 5 years using manual LBC if screening sensitivity and specificity were lowered.
Interpretation: The most cost-effective strategy for cervical cancer screening is once every 5 years using AI-assisted LBC. This provides a new alternative for cervical cancer screening scale-up and elimination.
个人简介:
沈明望,现任西安交通大学公共卫生学院流行病与卫生统计学系副教授。2017年12月在西安交通大学数学与统计学院获得应用数学博士学位。2015年至2017年在美国德克萨斯大学奥斯汀分校生物统计系公派交流学习两年,于2016年4月获“徐宗本应用数学论文奖”(一等奖)。近3年主持项目6项,包括国家自然科学基金青年基金项目、中国博士后科学基金面上项目(一等资助)和新冠肺炎疫情防控专项特别资助、陕西省自然科学基础研究计划一般项目(青年)、中央高校基本科研业务费专项资金资助交叉类和应急类(新冠肺炎)。近5年以第一作者发表SCI论文14篇,包括BMC Medicine、Proceedings of the Royal Society B:Biological Sciences、Vaccine等。主要研究方向为生物数学、理论流行病学、人工智能在医学中的应用等交叉学科研究,具体包括艾滋病、肝炎、流感、麻疹、新型冠状病毒肺炎等传染病以及胃癌、宫颈癌、糖尿病等慢性病的数学建模研究和卫生经济学评价。