Telemedicine-Enhanced Lung Cancer Screening Using Mobile Computed Tomography Unit with Remote Artificial Intelligence Assistance in Underserved Communities: Initial Results of a Population Cohort Study in Western China
Introduction
Lung cancer is the most prevalent type of cancer and the leading cause of cancer-related deaths.1 The human and economic burden of this disease poses an urgent public health crisis, particularly in low- and middle-income countries (LMICs), where an increasing trend is expected.2 Data from the 2016 cancer statistics in China showed that lung cancer ranks first in both new cases (828,100) and deaths (657,000).3 Screening using low-dose computed tomography (LDCT) has been instrumental in diagnosing early-stage lung cancer and reducing disease-related mortality.4 Annual LDCT screening for high-risk populations is recommended in the China National Lung Cancer Screening (LCS) Guidelines.5 However, screening with LDCT among individuals remains low globally.6 In China, the participation rate in national LCS programs is only 33.0%, with significantly lower rates in less developed regions.7
Barriers to cancer screening include lack of provider availability and community access to screening. Owing to the size and weight of whole-body computed tomography (CT) scanners, installation at hospitals or centers requires the right capabilities. Consequently, the majority of community health service centers and clinics in primary care in China lack LDCT, restricting access to LCS services. Community access to screening is further limited by prohibitive costs and inaccessibility of screening sites owing to distance. For instance, in Western China, 47% of impoverished individuals live within 2 km of a medical clinic, compared with 65% in metropolitan areas.8 Furthermore, client demand for screening can pose a barrier in situations where individuals are unaware of the benefits of screening or do not perceive themselves at risk.9
Mobile screening units (MSUs) are an innovative method of increasing community access by offering screening at convenient locations, thereby decreasing the distance and travel time needed to access screening services, particularly in areas without the requisite infrastructure for cancer screening services.9 Mobile CT screening units present a cost-effective and feasible solution for access to the LCS in communities, allowing for effective LCS in underserved areas,10 such as rural America and Brazilian underserved communities.11,12 However, evidence on MSU implementation and outcomes remains limited, particularly in resource-limited settings such as LMICs.
Barriers around costs, distance, infrastructure, and personnel shortages pose greater challenges in these regions that restrict screening access and quality. Although mobile CT units may alleviate certain access issues in underserved communities, questions persist regarding optimal delivery models, integration of emerging technologies such as artificial intelligence (AI), and coordination across multiple sites and health care facilities. Few prior studies have examined the feasibility of implementing mobile LCS initiatives with remote specialist support through telemedicine and AI systems across large disadvantaged populations in China or comparable LMICs.
West China Hospital (WCH) of Sichuan University in Sichuan Province, China, developed a telemedicine-enhanced LCS program using mobile CT units with remote AI diagnosis assistance in cooperation with a medical consortium (integrated delivery system).13 The LCS program is part of the Health Benefit for the People Project (HBPP), designed to complement and extend existing community-based services by providing routine screening for eligible individuals.14 AI enables remote, rapid, and accurate image reading, assisting radiologists in improving diagnostic quality and efficiency. To screen individuals with abnormalities, WCH and medical consortium hospitals provide collaborative support in the management of abnormal results based on a shared information platform, multidisciplinary teams (MDT), and referral policy in the medical consortium.
This study aimed to evaluate the initial results of the implementation of a telemedicine-enhanced LCS program. This study sought to establish a model for mobile CT units with remote AI diagnostic support in the medical consortium and improve the accessibility of LCS in underserved, resource-limited areas of Western China. Moreover, this analysis could inform efforts to improve life-saving LCS using telemedicine technology to reduce disparities across China and comparable LMICs.
Methods
HBPP-LCS PROGRAM
WCH is one of the top and largest hospitals in China. The HBPP initiated by WCH began in 2020 by providing free medical examinations and health follow-ups to achieve early prevention, screening, and intervention and comprehensively improve public health. This project was a population-based multicenter cohort study covering 11 regions with hospitals of medical consortium members located in Sichuan Province. From April 2020 to June 2022, ∼40,000 participants were included, with follow-up for 1–2 years. Detailed information regarding the cohort study has been described elsewhere.14
The telemedicine-enhanced LCS program was part of this project. The WCH medical consortium provided screening to residents’ locations, ensuring an accessible, affordable, and equivalent LCS service to reduce cancer outcome disparities among underserved communities. The mobile screening service is helpful for target populations who are unable to access medical institutions to obtain appropriate screening. This service may contribute to improving compliance with the screening program by promoting trust in large hospitals. Based on data from the cohort study, this study conducted a retrospective analysis of activities between July 2020 and June 2021. This study was approved by the ethics committee of West China Hospital, Sichuan University.
PARTICIPANTS AND SITES
Sichuan Province, located in Western China, is the fifth largest province (481,400 km2), with the fifth largest population of more than 83.7 million. Western China is a developing region. Its economy is less developed than that of other regions, and its terrain is mostly mountainous with plateaus, making transportation difficult. Health resource allocation in Western China is inequitable in the population and geographic dimensions.15 Due to the high smoking rate and traditional lifestyle, lung cancer ranked first of all cancer diagnoses and deaths in the province.16 The LCS program included the following three sites: Mianzhu County (Site 1), Longquan District (Site 2), and Pidu District (Site 3). The locations of the screened sites are shown in Figure 1. Owing to local socioeconomic conditions, these sites have lower levels of medical resource allocation than the provincial average, such as the number of physicians per thousand individuals (Mianzhu, 2.50; Longquan, 2.14; Pidu, 1.73; provincial average, 2.51).17
Fig. 1. Locations of screening sites in Sichuan Province.
In the corresponding community, individuals aged 40–80 years were eligible for LCS.18,19 Eligible participants received informed written consent for the collection of medical data, outcomes, and a survey. Each participant was required to complete a questionnaire, which included demographic information; behavioral habits, such as smoking; disease history, such as chronic obstructive pulmonary disease; and family history of cancer. The exclusion criteria were a previous diagnosis of lung cancer, pulmonary surgery, chest CT scans within the past 6 months, and unwillingness to undergo mobile CT examination. The screening program at Site 1 was conducted from July 2020 to September 2020, at Site 2 from October 2020 to November 2020, and at Site 3 from April to June 2021.
MOBILE CT UNIT WITH REMOTE AI ASSISTANCE
Descriptions of the mobile CT units are shown in Figure 2. Chest CT images were acquired using mobile LDCT scanners (Neusoft Corporation Co., Ltd., China). The scanning parameters followed a standard protocol, with a tube voltage of 120 kV, radiation dose below 1 mSv, and pixel matrix of 512 × 512. Subsequently, these images were reconstructed into high-resolution formats (window width 1,800, level 400, and slice thickness 1 mm) before being transmitted to WCH through the picture archiving and communication system network. After transmission, the CT images were identified using an AI system. The AI system can detect nodules and stratify the risk automatically and has been proven to enable specialists to increase the accuracy and consistency of the workflow.20
Fig. 2. Mobile CT unit. (A) At the screening site. (B) Lateral side view, entrance on the right. (C) Anterior view. (D) Onsite checks in the community. (E) Interior view of the operating room with an adjustable scanning bed. (F) Interior view of the command room. CT, computed tomography.
PROCEDURES
Selection and recruitment for LDCT screening were performed by trained professionals from WCH, mainly doctors and nurses. They asked participants to complete the questionnaire. Potentially eligible high-risk individuals who matched the Chinese screening guidelines were identified and assessed.
Mobile CT units were dispatched to communities at the screening sites. At each site, mobile CT units remained for 2 months to allow residents to receive LDCT scans. Each unit was equipped with a driver and two CT technicians. The person driving the mobile unit performed the intake of participants and explained the process, and CT technologists recorded the information, performed the scan, and transmitted CT images.
Digital CT images obtained were sent in real time to the radiology department of WCH, where radiologists analyzed the images with the assistance of AI (nodules were detected and classified by the AI system) and filled in structured reports. The images and reports were referred in real time to local hospitals in the medical consortium.
Local hospitals informed the participants to obtain the printed paper reports. A MDT of thoracic surgeons, pulmonologists, and coordinators in local hospitals interpreted the screening results and provided advice. Participants with negligible nodules (<4 mm) or no abnormalities were advised to undergo LDCT annually; those with low- and medium-sized nodules were followed up by the MDT in local hospitals; and those with high-risk modules were recommended to receive treatment at the local hospital. Participants with severe conditions were referred to WCH through the medical consortium’s green channel. The schematic flow of procedures is shown in Figure 3.
Fig. 3. Schematic flow.
DATA COLLECTION AND ANALYSIS
A detailed epidemiological questionnaire was administered to the LDCT participants, capturing data on age, sex, educational level, insurance type, employment, smoking status, history of occupational exposure, and family cancer history. The survey data underwent strict quality control and were stored in an electronic data capture (EDC) system. When analyzing the survey data, all participants’ details were de-identified to ensure that the identity of any person could not be ascertained. Outcome variables, such as high-risk pulmonary nodules or lung cancer diagnosis, and follow-up status were also measured. Data analysis was conducted by descriptive analysis.
Results
CHARACTERISTICS OF PARTICIPANTS
During the study period, 28,728 individuals from 3 sites were registered for the cohort study, and 19,517 (67.94%) underwent mobile-unit screening with LDCT. Among these individuals, the average age was 57.88 ± 9.23 years. Most participants (60.37%) were female. Approximately 80% of the population had an educational level below middle school. More than 90% had social medical insurance coverage, whereas the proportion of the uninsured was under 3%. The overall unemployment rate exceeded 40%. Nearly 20% of the participants were smokers, with over 20 packs per year, and one-fifth reported secondhand smoke exposure. One in 10 had hazardous occupational exposure, and a few had a family history of lung cancer.
The baseline details are presented in Table 1.
| CHARACTERISTIC | TOTAL (N = 19,517) | SITE 1 (n = 3,626) | SITE 2 (n = 7,590) | SITE 3 (n = 8,301) |
|---|---|---|---|---|
| N (%) | n (%) | n (%) | n (%) | |
| Age (years) | ||||
| 40–49 | 3,805 (19.50) | 528 (14.56) | 1,816 (23.93) | 1,461 (17.60) |
| 50–59 | 8,067 (41.33) | 1,271 (35.05) | 3,162 (41.66) | 3,634 (43.78) |
| 60–69 | 5,246 (26.88) | 1,146 (31.61) | 1,938 (25.53) | 2,162 (26.05) |
| ≥70 | 2,399 (12.29) | 681 (18.78) | 674 (8.88) | 1,044 (12.58) |
| Sex | ||||
| Male | 6,735 (34.51) | 1,297 (35.77) | 2,454 (32.33) | 2,984 (35.95) |
| Female | 11,782 (60.37) | 2,329 (64.23) | 5,136 (67.67) | 4,317 (52.01) |
| Educational level | ||||
| Primary school or below | 7,613 (39.01) | 1,832 (50.52) | 2,232 (29.41) | 3,549 (42.75) |
| Middle school | 8,039 (41.19) | 1,331 (36.71) | 3,071 (40.46) | 3,637 (43.81) |
| High school or vocational school | 2,846 (14.58) | 375 (10.34) | 1,546 (20.37) | 925 (11.14) |
| University/college undergraduate or above | 1,019 (5.22) | 88 (2.43) | 741 (9.76) | 190 (2.29) |
| Insurance type | ||||
| Social medical insurancea | 18,524 (94.91) | 3,403 (93.85) | 7,292 (96.07) | 7,829 (94.31) |
| Commercial insurance | 719 (3.68) | 44 (1.21) | 409 (5.39) | 266 (3.20) |
| Uninsured | 409 (2.10) | 80 (2.21) | 142 (1.87) | 187 (2.25) |
| Employment | ||||
| Employed | 5,722 (29.32) | 805 (22.20) | 2,153 (28.37) | 2,764 (33.30) |
| Unemployed | 9,393 (48.13) | 1,638 (45.17) | 3,088 (40.69) | 4,667 (56.22) |
| Retired | 4,402 (22.55) | 1,183 (32.63) | 2,349 (30.95) | 870 (10.48) |
| Smokingb | ||||
| 20 or more packs per year | 3,231 (16.55) | 591 (16.30) | 959 (12.64) | 1,681 (20.25) |
| Exposure to secondhand smoke | 7,337 (37.59) | 838 (23.11) | 4,812 (63.40) | 1,687 (20.32) |
| No smoking in the past 6 months or quit over 6 months ago | 16,125 (82.62) | 3,014 (83.12) | 6,558 (86.40) | 6,553 (78.94) |
| History of occupational exposure (dust, fumes, radon, asbestos, or arsenic) | ||||
| Yes | 2,004 (10.27) | 264 (7.28) | 1,014 (13.36) | 726 (8.75) |
| No | 17,513 (89.73) | 3,362 (92.72) | 6,576 (86.64) | 7,575 (91.25) |
| History of cancer among close relatives | ||||
| Lung cancer | 483 (2.47) | 64 (1.77) | 267 (3.52) | 152 (1.83) |
| Other malignant tumors | 1,257 (6.44) | 243 (6.70) | 676 (8.91) | 338 (4.07) |
| No cancer history | 17,777 (91.08) | 3,319 (91.53) | 6,647 (87.58) | 7,811 (94.10) |
OUTCOMES
Overall, 524 participants (2.68%) had high-risk pulmonary nodules. After a 1-year follow-up, 107 (0.55%) patients were diagnosed with lung cancer. The detection rates varied across sites, ranging from 1.61% to 4.52% for high-risk nodules and from 0.33% to 0.83% for lung cancer. According to the classification of lung cancer diagnoses, most cancers (74.77%) were adenocarcinomas. Among individuals with high-risk pulmonary nodules, the majority (62.98%) received timely treatment at the hospital. Detailed CT scan and follow-up outcomes are shown in Table 2.
| TOTAL (N = 19,517) | SITE 1 (n = 3626) | SITE 2 (n = 7590) | SITE 3 (n = 8301) | |
|---|---|---|---|---|
| N (%) | n (%) | n (%) | n (%) | |
| CT scan outcomes | ||||
| High-risk pulmonary nodules | 524 (2.68) | 164 (4.52) | 226 (2.98) | 134 (1.61) |
| Follow-up outcomes | ||||
| Diagnosed lung cancer | 107 (0.55) | 30 (0.83) | 50 (0.66) | 27 (0.33) |
| Adenocarcinoma | 80 (0.41) | 27 (0.74) | 39 (0.51) | 14 (0.17) |
| Squamous carcinoma | 5 (0.03) | 2 (0.06) | 3 (0.04) | 0 (0.00) |
| Sarcomatoid carcinoma | 1 (0.01) | 1 (0.03) | 0 (0.00) | 0 (0.00) |
| Small-cell carcinoma | 1 (0.01) | 0 (0.00) | 1 (0.01) | 0 (0.00) |
| Uncleara | 20 (0.10) | 0 (0.00) | 7 (0.09) | 13 (0.16) |
| Treatment in hospital | 330 (1.69) | 60 (1.65) | 185 (2.44) | 85 (1.02) |
| Surgery | 134 (0.69) | 33 (0.91) | 64 (0.84) | 37 (0.45) |
| Chemotherapy | 17 (0.09) | 2 (0.06) | 11 (0.14) | 4 (0.05) |
| Radiotherapy | 4 (0.02) | 0 (0.00) | 3 (0.04) | 1 (0.01) |
| Immunotherapy | 1 (0.01) | 0 (0.00) | 0 (0.00) | 1 (0.01) |
| Targeted therapy | 5 (0.03) | 0 (0.00) | 4 (0.05) | 1 (0.01) |
| Regular checkupb | 167 (0.86) | 25 (0.69) | 101 (1.33) | 41 (0.49) |
| Deathc | 2 (0.01) | 0 (0.00) | 2 (0.03) | 0 (0.00) |
Discussion
PRINCIPAL FINDINGS
This article presents the initial results of a large population participating in an LDCT LCS service conducted in Western China. The screening was carried out using mobile CT scanners and remote AI diagnostic assistance. We targeted high-risk individuals in underserved communities using an invitation to receive convenient and immediate access to CT screening. We offered diagnostic results and collaborative support in the management of abnormal results. Our findings demonstrated the potential of telemedicine applications to decrease barriers to accessing services in the participants’ locations and transmit diagnostic images and related reports, virtually connecting patients and medical specialists in a screening program. Our telemedicine screening program had high rates of early detection of abnormal results and promoted early treatment.
PUBLIC PARTICIPATION
Our study achieved a high participation rate of 67.94%, likely due to enhanced accessibility and community engagement strategies. In addition, our collaboration with local governments and medical consortium hospitals, along with recruitment to community centers and venues, likely improved awareness and trust. Similar participation rates were seen in other mobile unit LCS studies in underserved populations (55.8–75%). These findings suggest the potential for opportunistic recruitment in the community by mobile units.12,21 Active outreach using MSU has been shown to improve participation in the community, as individuals tend to be excited about accessing new technology and curious about engaging with radiographic services.22 Moreover, the mobile unit could be an advertising and educational tool, promoting awareness of lung cancer risks and benefits of screenings.12
IMPROVED ACCESS
Mobile CT units can help overcome resource-limited obstacles by providing services locally in participants’ communities, thereby improving access and participation. In our study, the mobile unit increased screening uptake by removing transportation barriers and decreasing travel distance, which was consistent with the findings of other mobile screening initiatives.11,12,21,23 The convenience of screening services available nearby facilitated early detection of lung cancer. Moreover, the use of telemedicine facilitated remote specialist image reading, overcoming local radiologist shortages and enabling prompt scan interpretation. The use of telehealth technology to connect underserved communities with specialist expertise could enhance the quality, efficiency, and timeliness of lung screening. Our model demonstrated the potential of mobile telemedicine in reducing disparities and improving equity in the early detection of cancer.
DETECTION AND MANAGEMENT
This study identified early abnormal results in a significant number of participants, with 2.68% found to have high-risk pulmonary nodules and 0.55% diagnosed with lung cancer. The lung cancer detection rate in our study was lower than that reported in Manchester, United Kingdom (3%, n = 1,384),21 and North Carolina, USA (2.2%, n = 550).11 This could be due to differences in the demographic and health profiles of the populations studied and varying levels of exposure to risk factors of the participants. Our approach not only facilitated early detection but also ensured the timely management of abnormal results. Most patients with high-risk pulmonary nodules received timely treatment at the hospital. The collaboration between WCH and local hospitals in the medical consortium allowed for efficient follow-up and treatment, emphasizing the importance of integrated health care systems in managing complex diseases, such as lung cancer.
IMPLEMENTATION CHALLENGES
Although the potential benefits of telemedicine-enhanced screening are vast, our study faced some challenges. First, financial sustainability is important for this program. The costs of running a mobile screening program, such as equipment maintenance, present a challenge for continuing to provide radiographs. In China, LDCT LCS is not currently reimbursed under medical insurance. Establishing a government insurance program or providing financial support to afford public LCS is recommended to ensure long-term program sustainability. Other possible solutions include seeking support from aligned local health systems and stakeholders and establishing relationships with foundations and companies interested in mobile screening programs.
Second, other mobile screening programs have reported unreliable internet connectivity and climate control for sensitive computing equipment.23 The equipment’s dimensions and sensibility pose limits to extending the usage to any context; therefore, remote areas with bad traffic conditions could be problematic, and a proper logistic arrangement should be established.12 These operational and technical barriers should be considered when developing mobile programs.
IMPLICATIONS FOR PRACTICE
Our findings demonstrated the feasibility of implementing telemedicine-enabled mobile CT LCS and diagnostic programs in underserved communities. This model could be expanded to screen high-risk populations in other disadvantaged regions. This study has several implications for future practice.
First, mobile CT screening units could be scaled up to extend their services to additional underserved communities across regions. Collaboration with local governments and health facilities is key to planning and outreach. In the future, dedicated patient navigators at each site can be assigned to assist with recruitment, appointment, result communication, and care coordination.
Second, the integration of AI-assisted diagnostic systems can enhance efficiency and quality when dealing with large volumes of screening scans from mobile units across communities. Machine learning algorithms can perform an initial image review to flag abnormal scans for radiologist reading, thereby increasing the yield of actionable findings. Furthermore, AI triage systems can prioritize high-risk cases requiring urgent specialist review, ensuring timely follow-up of critical findings.
Third, the implementation of this screening initiative through existing regional medical consortiums can promote integrated care. Conducting outreach within established health care networks allows seamless coordination between mobile units, primary care providers, specialists, and cancer centers. These partnerships enable the transparent sharing of patient records, imaging, and biopsy results across facilities. Leveraging current medical consortiums to deploy a screening model can strengthen linkages across the care continuum.
STRENGTHS AND LIMITATIONS
This study had several strengths. It was the first to examine a telemedicine LCS method developed in Western China. The use of mobile units with remote AI assistance is the first approach to reach communities, with particular attention paid to resource-limited areas that require screening equipment and diagnostic assistance. The large sample size across multiple rural and urban sites allowed the characterization of diverse underserved groups. The pragmatic setting of existing communities enabled the evaluation of real-world implementation challenges.
However, some limitations must be addressed in future studies. First, as this was a preliminary feasibility assessment with a limited observational period, incomplete follow-up tracking of outcomes such as cancer diagnosis rates could contribute to biased results. Besides, there may be inconsistencies in screening procedures across sites, such as variability in quality control protocols for screening, which could bias the results. Further long-term studies are needed to track follow-up outcomes, participant satisfaction, and multisites implementations according to RE-AIM (Reach, Effectiveness, Adoption, Implementation, and Maintenance) framework. Second, we were unable to characterize the cost-effectiveness of the telemedicine-enhanced screening model considering the lack of patient-level clinical and outcome data. Further health and economic evaluations comparing the cost-effectiveness of mobile CT screening and fixed CT scanning in hospitals are warranted. Such studies could inform policymakers regarding resource allocation decisions regarding decentralized mobile screening models.
While this study focused on underserved communities in China, the screening model could potentially be adapted to other LMICs facing comparable resource limitations or infrastructure barriers to cancer screening in rural and remote areas. Further research examining the portability of decentralized mobile screening initiatives across diverse global settings is warranted.
Conclusions
This study was the first to examine a telemedicine initiative utilizing mobile units with remote AI assistance in LCS to reach underserved communities in Western China. This approach could address resource limitations, increase screening uptake, and facilitate the early detection of lung cancer. This telemedicine-enhanced LCS model holds promise for improving access to crucial health care services in a large sample population and could guide future policies and practices in similar contexts such as LMICs. Further research is required to explore the long-term impact and cost-effectiveness of this initiative.
Acknowledgments
We thank Wenxia Huang, Hongli Bai, Jinghong Xian, Dan Hu, and Yong Li from the West China Hospital of Sichuan University for their assistance in providing photographs, information, data processing, and other support. We are also grateful for the support of all the participants in this study.
Authors’ Contributions
W.T.: Conceptualization (lead), data curation (supporting), formal analysis (equal), funding acquisition (equal), writing—original draft (equal). X.Y.: Data curation (lead), formal analysis (equal), writing—original draft (equal). J.S. and R.L.: Writing—review and editing (equal). W.L.: Supervision (lead), funding acquisition (equal), writing—review and editing (equal). All authors contributed to the article and approved the submitted version.
Disclosure Statement
No competing financial interests exist.
Funding Information
This work was supported by the National Natural Science Foundation of China (No. 92159302), the Science and Technology Project of Sichuan (2022ZDZX0018), 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University (ZYGD22009), and China Postdoctoral Science Foundation (No. 2023M732419).
REFERENCES
- 1. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71(3):209–249; doi:
10.3322/caac.21660 Crossref, Medline, Google Scholar - 2. The challenges of implementing low-dose computed tomography for lung cancer screening in low- and middle-income countries. Nat Cancer 2020;1(12):1140–1152; doi:
10.1038/s43018-020-00142-z Crossref, Medline, Google Scholar - 3. Lung cancer incidence and mortality in China: Updated statistics and an overview of temporal trends from 2000 to 2016. J Natl Cancer Center 2022;2(3):139–147; doi:
10.1016/j.jncc.2022.07.004 Crossref, Google Scholar - 4. Screening for lung cancer with low-dose computed tomography: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA 2021;325(10):971–987; doi:
10.1001/jama.2021.0377 Crossref, Medline, Google Scholar - 5. . Landscape on CT screening for lung cancer in Asia. Lung Cancer (Auckl) 2019;10:107–124; doi:
10.2147/LCTT.S192643 Crossref, Medline, Google Scholar - 6. Poor uptake of lung cancer screening: Opportunities for improvement. J Am Coll Radiol 2019;16(4 Pt A):446–450; doi:
10.1016/j.jacr.2018.12.018 Crossref, Medline, Google Scholar - 7. Uptake of lung cancer screening with low-dose computed tomography in China: A multi-centre population-based study. EClinicalMedicine 2022;52:101594; doi:
10.1016/j.eclinm.2022.101594 Crossref, Medline, Google Scholar - 8. Challenges to effective cancer control in China, India, and Russia. Lancet Oncol 2014;15(5):489–538; doi:
10.1016/S1470-2045(14)70029-4 Crossref, Medline, Google Scholar - 9. Mobile screening units for the early detection of cancer: A systematic review. Cancer Epidemiol Biomarkers Prev 2017;26(12):1679–1694; doi:
10.1158/1055-9965.EPI-17-0454 Crossref, Medline, Google Scholar - 10.
Screening in Underserved Communities: A Mobile Unit Brings Lung Cancer Screening to Remote Areas of West Virginia . 2022. Available from: https://www.acr.org/Practice-Management-Quality-Informatics/Imaging-3/Case-Studies/Lung-Cancer-Screening/Screening-in-Underserved-Communities [Last accessed:November 1, 2023 ]. Google Scholar - 11. Initial results from mobile low-dose computerized tomographic lung cancer screening unit: Improved outcomes for underserved populations. Oncologist 2020;25(5):e777–e781; doi:
10.1634/theoncologist.2019-0802 Crossref, Medline, Google Scholar - 12. Implementation of an integrated lung cancer prevention and screening program using a mobile computed tomography (CT) Unit in Brazil. Cancer Control 2022;29:10732748221121385; doi:
10.1177/10732748221121385 Crossref, Google Scholar - 13. The pathway of China’s Integrated Delivery System: Based on the analysis of the Medical Consortium Policies. Curr Med Sci 2022;42(6):1164–1171; doi:
10.1007/s11596-022-2654-9 Crossref, Medline, Google Scholar - 14. The design and implementation of natural population cohort study Biobank: A multiple-center project cooperation with medical consortia in Southwest China. Front Public Health 2022;10:996169; doi:
10.3389/fpubh.2022.996169 Crossref, Medline, Google Scholar - 15. Health resource allocation in Western China from 2014 to 2018. Arch Public Health 2023;81(1):30; doi:
10.1186/s13690-023-01046-x Crossref, Medline, Google Scholar - 16. Performance of low-dose computed tomography on lung cancer screening in high-risk populations: The experience over five screening rounds in Sichuan, China. Cancer Epidemiol 2020;69:101801; doi:
10.1016/j.canep.2020.101801 Crossref, Medline, Google Scholar - 17. Sichuan Health Statistical Yearbook (2021). Southwest Jiaotong University Press: Chengdu; 2022. Google Scholar
- 18. [China National Lung Cancer Screening Guideline with low-dose computed tomography (2018 Version)]. Zhongguo Fei Ai Za Zhi 2018;21(2):67–75; doi:
10.3779/j.issn.1009-3419.2018.02.01 Crossref, Medline, Google Scholar - 19. . Chinese expert consensus on diagnosis of early lung cancer (2023 Edition). Chinese J Tuberculosis Respir Dis 2023;46(1):1–18; doi:
10.3760/cma.j.cn112147-20220712-00592 Crossref, Google Scholar - 20. Deep learning empowers lung cancer screening based on mobile low-dose computed tomography in resource-constrained sites. Front Biosci (Landmark Ed) 2022;27(7):212; doi:
10.31083/j.fbl2707212 Crossref, Medline, Google Scholar - 21. Implementing lung cancer screening: baseline results from a community-based ‘Lung Health Check’ pilot in deprived areas of Manchester. Thorax 2019;74(4):405–409; doi:
10.1136/thoraxjnl-2017-211377 Crossref, Medline, Google Scholar - 22. Mobile lung screening: Should we all get on the bus? Ann Thorac Surg 2020;110(4):1147–1152; doi:
10.1016/j.athoracsur.2020.03.093 Crossref, Medline, Google Scholar - 23. Implementing a mobile diagnostic unit to increase access to imaging and laboratory services in western Kenya. BMJ Glob Health 2018;3(5):e000947; doi:
10.1136/bmjgh-2018-000947 Crossref, Medline, Google Scholar