Socio-technical Barriers to AI Diagnostic Tool Integration in South Asian Primary Healthcare: A Scoping Review
Article Sidebar
Main Article Content
Abstract
Background: Artificial intelligence (AI)-based diagnostic tools and digital decision support systems are increasingly deployed in South Asian primary healthcare (PHC). However, real-world integration is shaped by socio-technical constraints affecting feasibility, safety, acceptance, and continuity of care.
Objective: To map published evidence on AI-based diagnostic tools in South Asian PHC settings and synthesize reported socio-technical barriers to their integration.
Methods: A scoping review was conducted using PubMed as the sole database. Keyword and MeSH-based search strategies combined AI, PHC, implementation/barrier, and South Asia geography terms. Two reviewers independently screened titles/abstracts and full texts using Rayyan. Data were charted using a standardized form and barriers synthesized thematically into five domains: (1) technical/infrastructure, (2) human factors, (3) ethics/data governance, (4) health system/organizational, and (5) socioeconomic/access.
Results: Of 1,759 records screened, 14 evidence sources were included primarily from India (n=11), Bangladesh (n=2), and Pakistan (n=1). AI applications included diabetic retinopathy screening, cervical and breast cancer screening, tuberculosis/chest X-ray computer-aided detection, COVID-19 surveillance, and CDSS/DDSS. Barriers were multi-level: unreliable electricity and connectivity; image quality degradation; prediction instability from WhatsApp-compressed CXRs; workflow burden; ethical risks including inadequate consent and unclear accountability; weak referral pathways with low follow-up adherence (~13–17%); and socioeconomic barriers including transport costs and competing priorities.
Conclusion: Integration of AI diagnostic tools in South Asian PHC is constrained by interacting socio-technical barriers. Implementation strategies should address infrastructure readiness, workflow redesign, governance/accountability, and referral completion to translate AI screening into patient benefit.
Article Details
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
The Journal of Public Health and Family Medicine ( JPHFM) publishes all articles under the Attribution 4.0 Creative Commons International License (CC BY 4.0) as its default publishing license. This licensing approach is intended to maximize lawful reuse, dissemination, and scholarly impact while ensuring that authors receive appropriate credit for their work and that the integrity of the scholarly record is maintained.
Permitted Uses Under CC BY 4.0
Under the CC BY 4.0 license, readers and users may copy and redistribute the material in any medium or format and may remix, transform, and build upon the material for any purpose, including commercial use. These uses do not require prior permission from the publisher or the author(s), provided that appropriate attribution is given to the author(s) and the original source, a link to the license is included, and any changes made are clearly indicated.
Author Rights
Authors retain full copyright of their work published in JPHFM. By submitting a manuscript and proceeding to publication, authors grant JHWCR a non-exclusive license to publish and disseminate the article under CC BY 4.0. This arrangement ensures that authors remain free to reuse, share, and distribute their work broadly while allowing the journal to publish, archive, and index the Version of Record. Authors may deposit the submitted version (pre-print), the accepted version (post-print), and the published Version of Record in repositories, personal websites, and institutional platforms in accordance with the journal's Self-Archiving provisions.
License Display and Article-Level Licensing
For transparency, discoverability, and consistent indexing, JPHFM ensures that the CC BY 4.0 license is clearly displayed at the article level. The license statement and link are presented on each article landing page and, where applicable, within the full-text formats provided by the journal, including PDF and HTML versions.
Third-Party Content
Authors are responsible for ensuring that they have appropriate permission to reuse, reproduce, or adapt any third-party content included in their manuscript, including figures, tables, images, instruments, extensive quotations, or other copyrighted materials. If third-party content is not covered by CC BY 4.0, this must be made clear to readers to prevent unintended reuse beyond the permissions granted.
Funder Compliance
The CC BY 4.0 license is widely recognized as compatible with common open access requirements and supports compliance with a broad range of funder and institutional policies that require immediate open access and flexible reuse rights.
How to Cite
References
1.Jindal D, Sharma H, Gupta Y, Ajay VS, Roy A, Sharma R, Ali M, Jarhyan P, Gupta P, Srinivasapura Venkateshmurthy N, Ali MK. Improving care for hypertension and diabetes in india by addition of clinical decision support system and task shifting in the national NCD program: I-TREC model of care. BMC health services research. 2022 May 23;22(1):688.
2.Zaman M, Hridhee RA, Bhuiyan RA, Gomes CA, Rahman MM, Islam SM, Sarker F, Mamun KA. Efficacy of using a digital health intervention model using community health workers for primary health services in Bangladesh: a repeated cross-sectional observational study. BMC Public Health. 2025 Dec;25(1):1-4.
3.Tahir HN, Habib S, Jabeen R, Dhanwani A, Jawed S, Guerrero-Torres L, Khurram SS, Aga IZ, Karim M, Kazi AM, Zaidi S. Protocol of digital decision support system (DDSS) embedded within a tele-primary healthcare platform in Pakistan: an assessment of usability, acceptability, effectiveness and perceived quality of care in a resource-constrained setting. BMJ open. 2024 Sep 1;14(9):e082167.
4.Chauhan A, Vale L, Kankaria A, Gupta V, Singh M, Kaur G, Kumar S, Yadav M, N N, Tigari B, Bhadada S. Reach and implementation of human and AI-assisted diabetic retinopathy screening models in primary healthcare settings in India. Scientific reports. 2025 Nov 21;15(1):41355.
5.Chauhan A, Vale L, Kankaria A, Gupta V, Kaur G, N N, Sood N, Tigari B, Bhadada S, Duggal M. Assessing adoption of human and AI-enabled diabetic retinopathy screening in primary healthcare settings: findings from a pragmatic trial. Archives of Public Health. 2025 Nov 10;83(1):268.
6.Duggal M, Chauhan A, Gupta V, Kankaria A, Budhija D, Verma P, Miglani V, Syal P, Kaur G, Kumar L, Mutyala N. Real-World Evaluation of AI-Driven Diabetic Retinopathy Screening in Public Health Settings: Validation and Implementation Study. JMIR medical informatics. 2025 Sep 9;13(1):e67529.
7.Nderitu P, Do Rio JM, Rasheed R, Raman R, Rajalakshmi R, Bergeles C, Sivaprasad S. Deep learning for gradability classification of handheld, non-mydriatic retinal images. Scientific reports. 2021 May 4;11(1):9469.
8.Chauhan A, Sarkar D, Verma GS, Rastogi H, Adapa K, Duggal M. Evaluating trustworthiness in AI-Based diabetic retinopathy screening: addressing transparency, consent, and privacy challenges. BMC Medical Ethics. 2025 Oct 17;26(1):140.
9.Poli UR, Gudlavalleti AG, Bharadwaj Y J, Pant HB, Agiwal V, Murthy GV. Development and clinical validation of visual inspection with acetic acid application-artificial intelligence tool using cervical images in screen-and-treat visual screening for cervical cancer in south India: a pilot study. JCO Global Oncology. 2024 Dec;10:e2400146.
10.Davalagi SB, Palicheralu BS, Murthy SS, Hurlihal S. Acceptance of artificial intelligence (AI)-based screening for breast health in urban slums of central Karnataka, India–SWOC analysis. Journal of Family Medicine and Primary Care. 2022 Oct 1;11(10):6023-8.
11.Nath A, Hashim Z, Shukla S, Poduvattil PA, Neyaz Z, Mishra R, Singh M, Misra N, Shukla A. A multicentre study to evaluate the diagnostic performance of a novel CAD software, DecXpert, for radiological diagnosis of tuberculosis in the northern Indian population. Scientific reports. 2024 Sep 5;14(1):20711.
12.Kumar S, Rayal S, Bommaraju R, Varasala NP, Papineni S, Deo S. Understanding providers’ attitude toward AI in India’s informal health care sector: survey study. JMIR Formative Research. 2025 Feb 10;9(1):e54156.
13.Antony M, Kakileti ST, Shah R, Sahoo S, Bhattacharyya C, Manjunath G. Challenges of AI driven diagnosis of chest X-rays transmitted through smart phones: a case study in COVID-19. Scientific Reports. 2023 Oct 23;13(1):18102.
14.Sarker F, Chowdhury MH, Ratul IJ, Islam S, Mamun KA. An interactive national digital surveillance system to fight against COVID-19 in Bangladesh. Frontiers in Digital Health. 2023 May 11;5:1059446.