Artificial Intelligence in Health and Medicine: Progress, Challenges, and Recommendations

Xi Li¹; Jue Liu^1,#

1. Department of Epidemiology and Biostatistics, School of Public Health, Peking University, Beijing, China.

^# Corresponding author: Jue Liu, jueliu@bjmu.edu.cn.

Artificial intelligence (AI) has broadly reshaped health and medicine, benefiting clinicians, patients, and health systems. However, technical, regulatory, and ethical challenges exist in the application of medical AI, ranging from data scarcity to fairness. We provide our perspective on how to address the major challenges facing widespread clinical adoption from both technical (e.g., building high-quality datasets, using larger and more diverse datasets for training, creating problem formulations that go beyond supervised learning, and combining human skills with AI tools) and ethical (e.g., using highly secure data platforms and strengthening governmental legislation) perspectives.

卫生健康人工智能：进展、挑战与建议

李希¹；刘珏^1,#

1. 流行病与卫生统计学系，公共卫生学院，北京大学，北京，中国。

^# 通讯作者：刘珏，jueliu@bjmu.edu.cn。

人工智能（AI）已广泛重塑医疗健康领域，为临床医生、患者及卫生系统带来诸多益处。然而，医疗AI的应用仍存在技术、监管和伦理层面的挑战，包括数据稀缺性、公平性等问题。本文从技术（如构建高质量数据集、采用更大规模且多样化的训练数据、突破监督学习范式的问题建模、实现人类专业技能与AI工具的人机协同）与伦理（如采用高安全性数据平台、加强政府立法）角度，就如何应对AI在卫生健康领域广泛落地应用所面临的核心挑战提出建议。

For more information: https://weekly.chinacdc.cn/en/article/doi/10.46234/ccdcw2025.218

Evaluating Large Language Models’ Potential in Field Epidemiology Investigation Based on Chinese Context — Zhejiang Province, China, 2025

Tao Zhang^1,&, Qifeng Zhao^2,&, Yaxin Dai³, Mengna Wu¹, Yujia Zhai¹, Le Xu¹, Xue Gu¹, Junfen Lin¹ , Chen Wu^1,#

1. Department of Public Health Surveillance and Advisory, Zhejiang Provincial Center for Disease Control and Prevention, Hangzhou City, Zhejiang Province, China;

2. Department of Communicable Disease Control and Prevention, Shaoxing Center for Disease Control and Prevention, Shaoxing City, Zhejiang Province, China;

3. Department of Communicable Disease Control and Prevention, Zhoushan Center for Disease Control and Prevention, Zhoushan City, Zhejiang Province, China.

^& Joint first authors.

^# Corresponding author: Chen Wu, chenwu@cdc.zj.cn.

Large language models (LLMs) have demonstrated potential applications across diverse fields, yet their effectiveness in supporting field epidemiology investigations remains uncertain. We assessed six prominent LLMs (ChatGPT-o4-mini-high, ChatGPT-4o, DeepSeek-R1, DeepSeek-V3, Qwen3-235B-A22B, and Qwen2.5-max) using multiple-choice and case-based questions from the 2025 Zhejiang Field Epidemiology Training Program entrance examination. Model responses were evaluated against standard answers and benchmarked against performance scores from junior epidemiologists. For multiple-choice questions, only DeepSeek-V3 (75%) exceeded the 75th percentile performance level of junior epidemiologists (67.5%). In case-based assessments, most LLMs achieved or surpassed the 75th percentile of junior epidemiologists, demonstrating particular strength in data analysis tasks. Although LLMs demonstrate promise as supportive tools in field epidemiology investigations, they cannot yet replace human expertise. Significant challenges persist regarding the accuracy and timeliness of model outputs, alongside critical concerns about data security and privacy protection that must be addressed before widespread implementation.

大语言模型在现场流行病学调查中的应用潜力评估：基于中文语境的分析 — 浙江省，中国，2025年

章涛^1,&, 赵棋锋^2,&, 戴亚欣³, 吴梦娜¹, 翟羽佳¹, 徐乐¹, 古雪¹,  林君芬¹, 吴晨^1,#

1. 公共卫生监测与业务指导所，浙江省疾病预防控制中心，杭州市，浙江省，中国；

2. 传染病预防控制科，绍兴市疾病预防控制中心，绍兴市，浙江省，中国；

3. 传染病预防控制科，舟山市疾病预防控制中心，舟山市，浙江省，中国。

^& 共同第一作者

^# 通讯作者: 吴晨，chenwu@cdc.zj.cn。

大语言模型（Large Language Models, LLMs）已在多个领域展现出较强的应用潜力，但其在现场流行病学调查中的辅助作用尚不清楚。本研究利用2025年浙江省现场流行病学培训项目入学考试中的单项选择题与案例分析题，对6个主流大语言模型（ChatGPT-o4-mini-high、ChatGPT-4o、DeepSeek-R1、DeepSeek-V3、Qwen3-235B-A22B、Qwen2.5-max）进行评估，将模型作答结果与标准答案进行比对并量化评分。以参加考试的低年资现场流行病学专家的考试成绩作为基准参照，评估大语言模型的表现。在单项选择题部分，仅有DeepSeek-V3模型（正确率75%）超过了低年资现场流行病学专家成绩的第75百分位（正确率67.5%）。在案例分析题中，多数大语言模型的成绩达到或超过低年资现场流行病学专家成绩的第75百分位，其中在数据分析任务中大语言模型表现突出。尽管大语言模型在现场流行病学调查中展现出作为辅助工具的潜力，但目前尚无法替代人类专业知识。模型输出的准确性与时效性仍面临显著挑战，且数据安全和隐私保护方面的关键问题也需在大规模应用前得到妥善解决。

For more information: https://weekly.chinacdc.cn/en/article/doi/10.46234/ccdcw2025.220

Spatial Distribution and Clustering Patterns of Cognitive Impairment Among the Older Population — 31 PLADs, China, 2024

Ying Liu^1,&; Yushan Zhang^2,&; Ji Shen²; Chi Zhang³; Yingchen Sang¹; Youjiao Wang¹; Houguang Zhou⁴; Lei Wang⁵; Jie Zhang²; Ying Yuan²; Shiwei Liu^1,#; Hong Shi^2,#

1. Tobacco Control Office, Chinese Center for Disease Control and Prevention, Beijing, China;

2. Department of Geriatrics, Beijing Hospital, National Center of Gerontology; Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, China;

3. The Key Laboratory of Geriatrics, Beijing Institute of Geriatrics, Beijing Hospital, National Center of Gerontology, National Health Commission; Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, China;

4. Department of Geriatrics, Huashan Hospital, National Clinical Research Center for Aging and Medicine, Fudan University, Shanghai, China;

5. Department of Geriatrics, Medical Center on Aging of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.

^& Joint first authors.

^# Corresponding author: Shiwei Liu, liusw@chinacdc.cn；Hong Shi, shihong2584@bjhmoh.cn .

Cognitive impairment has emerged as a major public health challenge threatening the physical and mental well-being of older adults. This study aimed to elucidate the spatial distribution patterns and clustering characteristics of cognitive impairment among Chinese adults aged 65 years and older, thereby establishing an evidence base for the development of geographically targeted prevention and intervention strategies. We analyzed data from the 2024 China National Survey on Aging and Health (CAHS), employing Rao-Scott chi-square tests to assess the differences in prevalence across demographic subgroups. Global and local spatial autocorrelation analyses were conducted to characterize the geographic distribution and clustering patterns of cognitive impairment. Our findings revealed that the prevalence of subjective cognitive decline and mild cognitive impairment among Chinese adults aged ≥65 years in 2024 was 38.8% and 28.4%, respectively. Subjective cognitive decline was most prevalent in the Western region (45.1%), whereas mild cognitive impairment peaked in the Central region (31.0%). Global spatial autocorrelation analysis demonstrated significant spatial clustering for both subjective cognitive decline (Moran’s I=0.162, Z=2.242, P=0.025) and mild cognitive impairment (Moran’s I=-0.242, Z=-2.431, P=0.015). These results underscore the substantial burden of cognitive impairment among China’s aging population and highlight pronounced regional disparities. Prevention strategies should prioritize geographic heterogeneity, emphasizing resource allocation to high-prevalence regions with limited healthcare infrastructure, while leveraging the comparative advantages of low-prevalence areas to facilitate the nationwide implementation of evidence-based, precision-oriented, and resource-efficient cognitive health interventions.

老年人认知障碍的空间分布及聚类模式 — 31省份，中国，2024年

刘影^1，&；张玉珊^2；&；沈姞²；张驰³；桑莹辰¹；王友娇¹；周厚广⁴；王蕾⁵；张洁²；袁莹²；刘世炜^1，#；施红^2，#

1. 控烟办，中国疾病预防控制中心，北京，中国；

2. 老年病科，北京医院，国家老年医学中心，中国医学科学院老年医学研究院，北京，中国；

3. 老年病科，北京医院，国家老年医学中心，国家卫生健康委北京老年医学研究所，国家卫生健康委老年医学重点实验室，中国医学科学院老年医学研究院，北京，中国；

4. 老年病科，复旦大学附属华山医院，国家老年疾病临床医学研究中心，上海，中国；

5. 老年病科，上海交通大学附属瑞金医院，上海市，中国。

^& 共同第一作者

^# 通讯作者：刘世炜，liusw@chinacdc.cn；施红，shihong2584@bjhmoh.cn。

认知障碍已成为严重威胁老年人身心健康的重要公共卫生问题之一。本研究旨在了解中国65岁及以上老年人群认知障碍的空间分布及聚集模式，为制定有针对性的区域防控策略提供科学依据。本研究采用2024年全国老年健康调查（CAHS）的数据。运用Rao - Scott卡方检验比较不同亚组的认知障碍患病差异，通过全局和局部空间自相关分析，探究认知障碍的空间分布特征和聚集模式。结果显示，2024年中国65岁及以上老年人群主观认知下降和轻度认知障碍的患病率分别为38.8%和28.4%。主观认知下降患病率在西部地区最高（45.1%），而轻度认知障碍患病率在中部地区最高（31.0%）。全局空间自相关分析显示，主观认知下降（莫兰指数I=0.162，Z=2.242，P=0.025）和轻度认知障碍（莫兰指数I=-0.242，Z=-2.431，P=0.015）的分布存在空间聚集性。当前中国老年人群的认知障碍负担较重，且存在显著的区域差异，防治措施应注重区域差异，优先支持患病率高且资源有限的地区，发挥低患病率地区的优势，灵活应对空间异质性，推动全国范围内科学、精准、高效的认知障碍防治策略的实施。

For more information: https://weekly.chinacdc.cn/en/article/doi/10.46234/ccdcw2025.219

The Cluster of Mpox (Clade Ib) Infections — Yiwu City, Zhejiang Province, China, July–August 2025

Zhiping Long^1,&, Liebo Zhu^2,&, Jian Cai³, Hanran Ji⁴, Qin Ye², Xuguang Shi³, Guangming Zhang¹, Shuying Zhu¹, Xuelian Zhang¹, Zhifeng Pang¹, Kaizhi Bai¹, Jun Jiang^2,#, Jiming Sun^3,#

1. Jinhua Center for Disease Control and Prevention, Jinhua City, Zhejiang Province, China.

2. Yiwu Center for Disease Control and Prevention, Jinhua City, Zhejiang Province, China.

3. Zhejiang Provincial Center for Disease Control and Prevention, Hangzhou City, Zhejiang Province, China.

4. Center for Global Public Health, Chinese Center for Disease Control and Prevention, Beijing, China.

^& Joint first authors.

^# Correspondence: Jiming Sun, jmsun@cdc.zj.cn; Jun Jiang, junjiang3176@163.com.

On August 6, 2025, the Yiwu CDC received notification of a suspected mpox case. Subsequent laboratory testing confirmed mpox virus (MPXV, Clade Ib) infection, prompting the initiation of a comprehensive multi-level epidemiologic investigation. Between August 6–10, 2025, provincial, municipal, and county expert teams conducted systematic case finding, interview-based exposure mapping, and linkage analysis of public security and border records to reconstruct contact trajectories. The investigation included comprehensive contact tracing, environmental sampling, and real-time PCR testing of lesion and oropharyngeal specimens for MPXV detection. Six laboratory-confirmed cases were identified with symptom onset dates spanning July 23 through August 6. Five patients received treatment in Yiwu, while one was managed in Changzhou. The cases were predominantly male (5/6), with ages ranging from 22–43 years (median 30 years). Four of the six cases were foreign nationals. Investigators identified and monitored 52 core close contacts and 38 general contacts under 21-day health surveillance protocols. Environmental sampling (n=43) conducted at five case residences and personal items yielded 27 positive results (62.8%) for MPXV. This outbreak represents an imported mpox cluster with subsequent person-to-person transmission occurring primarily through intimate contact. We documented substantial household environmental contamination, emphasizing the critical importance of comprehensive decontamination measures. Rapid case detection, systematic contact management, and terminal disinfection protocols effectively contained further viral spread.

猴痘（Ib进化分支）聚集性病例报告 — 义乌市，浙江省，中国，2025年7–8月

龙智平^1,&，朱列波^2,&，蔡剑³，纪瀚然⁴，叶蓁²，施旭光³，章光明¹，朱淑英¹，张雪莲¹，庞志峰¹，白开智¹，蒋君^2,#，孙继民^3,#

1.疾病预防控制中心，金华市，浙江省，中国；

2.义乌市疾病预防控制中心，金华市，浙江省，中国；

3.浙江省疾病预防控制中心，杭州市，浙江省，中国；

4. 全球公共卫生中心，中国疾病预防控制中心，北京，中国。

^& 共同第一作者。

^# 通讯作者: 孙继民，jmsun@cdc.zj.cn；蒋君，junjiang3176@163.com。

2025年8月6日，义乌市疾控中心接1例猴痘疑似病例报告，实验室确诊为猴痘病毒（Ib分支）感染后，启动多层次流调。8月6-10日，省、市、县专家组开展病例排查、暴露图谱绘制及接触轨迹还原，同步追踪密接、环境采样，并以实时PCR检测标本。共确诊6例病例，症状出现于7月23日-8月6日，5例在义乌治疗、1例在常州，以男性为主（5/6），年龄22-43岁（中位30岁），含4例外籍人员。共识别并管控52名核心密切接触者和38名一般接触者，均按规范实施21天健康监测；43份环境标本中27份病毒阳性（62.8%）。本次疫情为输入性猴痘聚集性疫情，后续人际传播主要通过密切接触发生，病例住所污染严重。快速发现病例、系统管理接触者及终末消毒，有效遏制了病毒传播。

For more information: https://weekly.chinacdc.cn/en/article/doi/10.46234/ccdcw2025.221