Implementing AI-enabled chest X-ray for community-based integrated screening for tuberculosis, chronic respiratory diseases, and cardiovascular diseases in Nigeria

Implementing AI-enabled chest X-ray for community-based integrated screening for tuberculosis, chronic respiratory diseases, and cardiovascular diseases in Nigeria | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Implementing AI-enabled chest X-ray for community-based integrated screening for tuberculosis, chronic respiratory diseases, and cardiovascular diseases in Nigeria Chidimma Okoye, Jude Ilozumba, John Okoh, Amana Effiong, Young Oluokun, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9180615/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 13 You are reading this latest preprint version Abstract Background Low- and middle-income countries face a growing dual burden of communicable and non-communicable diseases, while health services remain largely organised around vertical programmes. In Nigeria, tuberculosis (TB) services are relatively established at the primary healthcare level, whereas access to cardiovascular disease (CVD) and chronic respiratory disease (CRD) care remains limited in rural settings. Artificial intelligence (AI)–enabled chest X-ray offers an opportunity to integrate TB screening with the identification of other cardiopulmonary abnormalities at the community level. This study describes the implementation and outcomes of a community-based, AI-enabled integrated screening intervention in rural Nigeria. Methods We conducted a non-randomised implementation study across five Local Government Areas in Ebonyi and Nasarawa States between January 2023 and December 2024. Community outreach activities used portable digital chest X-ray integrated with AI software to screen individuals aged six years and above. Presumptive TB cases underwent GeneXpert testing, while non-TB radiographic abnormalities were referred for further evaluation. Descriptive analyses summarised screening yield, diagnostic outcomes, and linkage to care. Results A total of 9,585 individuals were screened through 93 community-based outreach activities. Overall, 3,166 (33%) chest radiographs were flagged as abnormal by AI. In total, 1,336 individuals were classified as presumptive TB cases, of whom 1,123 (84%) produced sputum samples for GeneXpert testing. 204 individuals were bacteriologically confirmed with TB, and 194 (95%) were initiated on treatment. An additional 199 individuals were clinically diagnosed with TB following radiologist review. Among abnormal radiographs, 2,367 (75%) showed features suggestive of CVDs or CRDs. All clients with such conditions were referred to tertiary facilities; however, only 12% completed the referral. Implementation adaptations, including improved imaging protocols and community-based follow-up strategies, supported TB linkage but had a limited impact on non-TB referral completion. Conclusions AI-enabled community chest X-ray screening is feasible for TB case finding in rural Nigeria and achieves high linkage to TB treatment. However, limited decentralisation of non-communicable disease services constrains care continuity for CVDs and CRDs. Integrated screening programmes should be paired with strengthened primary healthcare capacity, complementary tools such as blood pressure measurement, and context-specific community engagement strategies. Figures Figure 1 Background Low- and middle-income countries (LMICs) are experiencing a growing double burden of disease, in which communicable diseases coexist with an increasing prevalence of non-communicable diseases (NCDs) within the same population( 1 ). This reflects overlapping demographic and epidemiological transitions, including population growth, increased life expectancy, urbanisation, and lifestyle changes( 2 ). In parallel, many LMICs face a double burden of malnutrition, with undernutrition and overnutrition occurring concurrently, further increasing vulnerability to both infectious and chronic conditions ( 3 , 4 ). These converging trends place sustained pressure on health systems that remain largely organised around vertical, disease-specific programmes. Nigeria exemplifies this challenge, bearing the highest tuberculosis (TB) burden in Africa, alongside persistent gaps in the prevention and management of drug-resistant TB and TB/HIV co-infection( 5 ). TB remains among the leading causes of death nationally. The World Health Organization estimated approximately 56,000 TB deaths among HIV-negative individuals in Nigeria in 2024, with an additional 5,800 deaths among people living with HIV. ( 5 )” At the same time, the burden of NCDs has increased steadily. In 2019, NCDs accounted for approximately 29% of all deaths in Nigeria( 6 ). Cardiovascular diseases (CVDs) contributed about 11% of all deaths, while chronic respiratory diseases (CRDs) accounted for approximately 2%, alongside a substantial loss of disability-adjusted life years ( 6 ). Hypertension, the leading modifiable risk factor for cardiovascular disease, affects over 30% of Nigerian adults and contributes substantially to myocardial infarction and stroke( 7 ). National evidence also indicates a high prevalence of additional CVD risk factors, including overweight and obesity, diabetes mellitus, physical inactivity, tobacco use, exposure to household air pollution, and unhealthy dietary patterns( 6 , 8 ). Hospital-based studies further demonstrate a sustained rise in CVD-related admissions and mortality. A 15-year review from Lagos reported that CVDs accounted for 20.8% of all medical admissions and 30.4% of in-hospital mortality, underscoring the growing clinical and economic burden of chronic diseases in Nigeria ( 9 ). Nigeria’s primary healthcare system is relatively better resourced for the surveillance and management of endemic communicable diseases such as malaria, TB, and HIV, largely through established vertical programmes and sustained donor support ( 5 , 10 ). While TB case detection and treatment coverage improved between 2019 and 2023 ( 5 ), Access to NCD prevention, diagnosis, and long-term care remains limited, particularly at the primary healthcare level. For example, a recent assessment of primary healthcare facility readiness in Nigeria found that many facilities lacked essential diagnostic tools, trained personnel, and medicines required to implement basic WHO Package of Essential NCD Interventions (PEN), limiting routine screening and management of conditions such as hypertension and diabetes ( 10 – 12 ). Services for CVDs and CRDs are often unavailable or fragmented at facilities closest to communities, resulting in delayed diagnoses, high out-of-pocket expenditure, and poor continuity of care. These gaps undermine progress toward universal health coverage and disproportionately affect rural and hard-to-reach populations ( 11 ). Integrated, community-based approaches that combine screening for communicable and non-communicable diseases are therefore increasingly important. Systematic chest X-ray–based screening has re-emerged as an effective strategy for TB case finding in high-burden settings ( 13 , 14 ). AI–enabled chest X-ray interpretation systems have demonstrated high accuracy for TB screening and increasing utility in identifying non-TB abnormalities, while reducing reliance on scarce specialist expertise ( 15 – 17 ). In 2021, the World Health Organization endorsed the use of AI-based chest X-ray interpretation for TB screening, recognising its potential to expand access, improve efficiency, and strengthen case detection in resource-limited settings ( 18 ). Evidence also shows that chest radiography can identify abnormalities suggestive of CRDs and CVDs, including cardiomegaly and chronic lung changes ( 19 ). The INTEGRATE TB project was designed to apply community-based, AI-enabled chest X-ray screening for tuberculosis, while identifying radiographic abnormalities suggestive of cardiovascular and chronic respiratory diseases in remote rural communities in Nigeria. By using TB screening as an entry point for broader lung and cardiovascular health assessment, the project aimed to expand access to diagnostic services for underserved populations and to document implementation experiences within routine programme settings. This paper describes the implementation approach, summarises findings across the screening and referral cascade, and highlights key operational considerations, guided by the Template for Intervention Description and Replication (TIDieR) framework ( 20 ). Methodology Study design and setting This study was a non-randomised implementation study using routinely collected programme data to evaluate a community-based screening intervention. It was conducted across five Local Government Areas (LGAs) in Nigeria: three LGAs in Ebonyi State (South-East) and two LGAs in Nasarawa State (North-Central), between January 2023 and December 2024. Communities were selected using the State Primary Health Care Development Agency annual Updated Directory of Settlements, 2022. Ministry of Health; 2022 and restricted to those classified as hard-to-reach, based on geographic remoteness, limited transport connectivity, and poor access to health facilities, consistent with WHO systematic screening guidance ( 21 , 22 ). Intervention description We evaluated the screening cascade of a community-based TB screening programme that used AI-enabled chest X-ray to simultaneously screen for tuberculosis and identify radiographic abnormalities suggestive of CVDs and CRDs. The intervention used portable digital chest X-ray (PDX) machines equipped with artificial intelligence (AI) software to enable real-time interpretation during outreach activities, in line with WHO recommendations on systematic TB screening and use of chest radiography ( 21 – 24 ). Tuberculosis screening served as the primary entry point, with non-TB abnormalities identified and referred through established pathways. The intervention aimed to improve early detection, strengthen linkage to care, and generate operational evidence on the feasibility of AI-enabled, multi-disease screening in underserved settings ( 22 , 25 ). Community engagement and delivery Prior to implementation, structured engagement was conducted with community leaders, women’s and youth groups, religious institutions, and local government health teams to introduce the intervention and adapt implementation plans to local contexts, consistent with WHO recommendations for community-based TB screening ( 18 , 21 ). Outreach schedules were aligned with community calendars, including market days and cultural events. Community volunteers were recruited from respected community members to support mobilisation, health education, and participant flow. Health education sessions were delivered in local languages using visual aids and public address systems, following best practices for community TB care and prevention ( 18 , 21 ). Outreach implementation and eligibility Outreach teams comprised a radiographer, a linkage coordinator, the Local Government TB and Leprosy Supervisor (LGTBLS), community volunteers, an administrative assistant responsible for real-time data entry, and CCFN supervisory staff. Outreach activities were conducted in accessible public spaces such as markets, town halls, and school compounds. Individuals aged six years and above were eligible for screening, excluding pregnant women. Verbal informed consent was obtained from adult participants, and parental consent was obtained for minors. Following consent, participants underwent symptom screening and chest radiography, with AI-generated results available within an average of 3 minutes. Screening tools and procedures Portable digital chest X-ray machines (Mini X-ray, Northbrook, IL, USA) were integrated with qXR Version 3, a CE-certified and WHO recommended AI software developed by Qure.ai (Mumbai, India). The software automatically analysed chest radiographs, generating a TB abnormality score ranging from 0.01 to 0.99 and also identifies non-TB abnormalities. Chest radiographs were analysed using qXR Version 3 (Qure.ai), which generates a TB abnormality score ranging from 0.01 to 0.99 and provides localisation of both TB-related and non-TB radiographic findings. Abnormal chest radiograph: A chest radiograph was classified as abnormal if the AI software localized any radiographic abnormalities, including TB-related abnormalities (e.g., cavities, nodules, consolidation) or non-TB abnormalities (e.g., cardiomegaly, fibrosis, pleural effusion, or structural changes), based on predefined classification outputs of the qXR algorithm. Presumptive TB: Defined as the presence of TB-related symptoms (using the WHO four-symptom screening tool: cough, fever, weight loss, or night sweats) and/or an AI-generated qXR score ≥ 0.5, in line with the manufacturer's default threshold and prior validation studies of AI-based chest X-ray triage Non-TB abnormalities (CVDs and CRDs): Radiographic abnormalities were categorised as suggestive of CVDs or CRDs based on AI-generated outputs. A decision-support algorithm, developed in collaboration with cardiopulmonary specialists, guided the categorisation and referral of non-TB abnormalities based on structured outputs generated by the qXR artificial intelligence software. CVD-related findings included features such as cardiomegaly. CRD-related findings included pulmonary and pleural abnormalities such as fibrosis, consolidation, reticulonodular patterns, pleural effusion, and other chronic lung changes. These classifications were used to guide referral for further clinical evaluation and were not considered definitive diagnoses. Diagnostic and referral pathways Presumptive TB cases provided sputum samples on site, which were transported to nearby GeneXpert laboratories following WHO-recommended biosafety and triple-packaging protocols( 22 ). Bacteriologically confirmed TB cases were linked to treatment at the nearest DOTS facility or a facility preferred by the participant, in line with Nigeria’s national TB guidelines. For individuals with TB-suggestive radiographic abnormalities identified by AI but with negative GeneXpert results or inability to produce sputum, chest radiographs and clinical histories were reviewed remotely by radiologists via a secure cloud-based platform. Based on this review, a clinical decision was made to initiate TB treatment or pursue alternative management in accordance with national TB guidelines and WHO diagnostic recommendations ( 18 , 22 ) Participants with non-TB abnormalities suggestive of CVDs or CRDs were referred to partnering tertiary hospitals: Federal Teaching Hospital Abakaliki (Ebonyi State) and Dalhatu Araf Specialist Hospital (Nasarawa State). Pre-implementation advocacy established expedited referral pathways, allowing referred clients to bypass routine general outpatient procedures and access specialist care directly, informed by radiographic criteria for chronic lung and cardiac disease. While no financial support for treatment was provided, logistical assistance was offered to reduce access barriers. Fidelity, supervision, and quality assurance Standard operating procedures and job aids guided all field activities. Field teams received initial and refresher training, supported by routine supervision. Digital data collection tools incorporated validation checks to enhance data quality, consistent with WHO recommendations for programme monitoring ( 24 ). Weekly supervisory visits and monthly review meetings assessed protocol adherence and implementation challenges. A dedicated WhatsApp platform facilitated real-time troubleshooting, and monthly feedback was shared with stakeholders to support continuous quality improvement. Data collection and analysis Data were collected using the electronic Q-Track platform provided with the AI model, and complementary paper-based registers capturing demographic characteristics, AI screening outputs, laboratory results, and referral outcomes. Paper records were routinely cross-checked against electronic data to ensure consistency. Data were cleaned and analysed using Microsoft Excel. Descriptive statistics were used to summarise participant characteristics, screening outcomes, diagnostic yield, and linkage-to-care rates, consistent with recommended approaches for implementation-focused TB screening evaluations ( 24 ) Ethical considerations Ethical approval was obtained from the National Health Research Ethics Committee (NHREC: NHREC/2024/01/137/13-06-24). Permission to conduct the intervention was granted by State Ministries of Health and State Tuberculosis Programmes. Participation was voluntary, informed consent was obtained from all participants or their guardians, and all data were anonymised prior to analysis to ensure confidentiality, in line with WHO ethical standards for TB care and the Declaration of Helsinki ( 27 , 28 ). Results Participant characteristics and screening volume Between January 2023 and December 2024, a total of 9,585 individuals were screened using AI-integrated portable digital chest X-ray through 93 community-based outreach activities conducted across five Local Government Areas (LGAs) in Ebonyi and Nasarawa States. Of those screened, 4,053 (42%) were male, and 5,532 (58%) were female. Screening volume ranged from 80 to 120 individuals per outreach day, with variations influenced by weather conditions and community mobilisation. Communities with larger populations or higher screening yields were prioritised for repeat outreach visits. Screening volume and outcomes disaggregated by LGA are presented in Table 1 Radiographic findings and AI outputs Of the 9,585 chest radiographs acquired during community outreach activities, 6,419 (67%) were classified as normal, while 3,166 (33%) were flagged as abnormal by the AI software. Among individuals with normal radiographs, 2,958 (46%) were male, and 3461 (54%) were female. Of those with abnormal radiographs, 1,095 (35%) were male, and 2,071 (65%) were female. The proportion of abnormal radiographs was higher in Ebonyi State (35%) compared with Nasarawa State (29%). Abnormality rates further disaggregated by sex and Local Government Area are presented in Table 2 . Tuberculosis screening, diagnosis, and linkage to care A total of 1,336 (14%) individuals were classified as presumptive tuberculosis (TB) cases based on the presence of TB-related symptoms, an AI-generated qXR score ≥ 0.5, or both. Of these, 800 (60%) individuals were identified based on an AI score ≥ 0.5, while 536 (40%) individuals were identified based on symptoms alone with qXR scores < 0.5. Overall, presumptive TB cases comprised 694(52%) males and 642(48%) females. Among presumptive TB cases, 1,123 individuals (84%) were able to produce spot sputum samples. GeneXpert testing identified 204 individuals as bacteriologically confirmed TB cases, representing 18% of those tested. Of these confirmed cases, 124(61%) were male, and 80 (39%) were female. Linkage-to-care data showed that 194 (95%) of bacteriologically confirmed TB cases were successfully initiated on treatment at a DOTS facility, either at the nearest facility or at a facility preferred by the patient. In addition, 199 individuals with TB-suggestive radiographic abnormalities but negative GeneXpert results or inability to produce sputum were clinically diagnosed and initiated on TB treatment following radiologist review of chest radiographs and clinical assessment. These clinically diagnosed cases included 114 males and 85 females. Non-TB abnormalities and referral outcomes Among the 3,166 abnormal chest radiographs, the AI software identified 2,367 individuals (75%) with abnormalities suggestive of non-TB conditions, including cardiovascular diseases (CVDs) and chronic respiratory diseases (CRDs). Specifically, 852 individuals had radiographic findings suggestive of CVDs, and 1,514 individuals had findings suggestive of CRDs. All individuals with non-TB abnormalities were referred to partnering tertiary hospitals for further clinical or laboratory evaluation. Of those referred, 121 individuals with suspected CVDs and 164 individuals with suspected CRDs were confirmed to have reached treatment facilities. Clinical or laboratory assessment confirmed 107 CVD cases and 147 CRD cases. At the three-month follow-up, 12 individuals with confirmed CVDs and 11 individuals with confirmed CRDs were documented as still engaged in care. Among individuals diagnosed with TB, approximately 18% also had radiographic features suggestive of cardiomegaly. Table 1 Characteristics of participants screened in community outreach activities (Median age 41 years (range: 6–101 years). State LGA (Total screened) Male n (%) Female n (%) Ebonyi Ikwo (2,432) 914 (38%) 1,518 (62%) Ishielu (2,616) 819 (31%) 1,797(69%) Izzi (1,727) 774 (45%) 953 (55%) Ebonyi total 6,775 (71%) Nasarawa Karu (1,585) 884 (52%) 701(48%) Lafia (1,223) 662(51%) 563(49%) Nasarawa total 2808 (29%) Table 2 Abnormality characteristics by LGA and sex State / LGA Abnormality Male Abnormality Female CVD Male CVD Female CRD Male CRD Female Presumptive TB (≥ 0.5) Male Presumptive TB (≥ 0.5) Female Ebonyi – Ikwo (808) 270(34%) 538(66%) 39(16%) 219((84%) 110(32%) 209(68%) 121 (52%) 110 (48%) Ebonyi – Ishielu (926) 247(27%) 679(73%) 45(14%) 264(86%) 111(28%) 279(72%) 91(40%) 136(60%) Ebonyi – Izzi (608) 248 (40%) 360 (60%) 32(27%) 84(73%) 134(38%) 228(68%) 80(62%) 50(38%) Nasarawa – Karu (477) 197(36%) 280(66%) 28(31%) 61(69%) 95(36%) 160(64%) 74(56%) 59(44%) Nasarawa – Lafia (347) 133 (33%) 214(67%) 22(25%) 58(75%) 65(35%) 123(65%) 46(58%) 33(42%) Discussion This study demonstrates that community-based AI-enabled chest X-ray screening is a feasible platform for integrated detection of tuberculosis (TB), cardiovascular diseases (CVDs), and chronic respiratory diseases (CRDs) in underserved rural communities in Nigeria. Among the 9,585 individuals screened, 3,166 (33%) chest radiographs were flagged as abnormal by the AI system, highlighting a substantial burden of previously undiagnosed cardiopulmonary conditions that would likely remain undetected within routine primary care systems ( 29 , 30 ). The findings provide insights into the distribution of abnormalities, sex differences in screening participation and disease patterns, and the emerging burden of non-communicable diseases (NCDs) in rural LMIC settings. However, considerable attrition along referral pathways for non-TB conditions was observed. Although individuals with NCD-related abnormalities were referred for further evaluation, only 12% completed referral that is reaching the treatment facility for further evaluation and care, this reflects structural gaps in service availability. These findings highlight both the public health value of multi-disease community screening and the importance of strengthening referral systems and decentralised services to ensure that early detection translates into effective care ( 31 , 32 ) High prevalence of radiographic abnormalities in rural communities A key finding of this study is the high proportion of abnormal chest radiographs detected during community screening. Approximately one-third (33%) of all screened individuals had abnormal chest radiographs, indicating substantial unmet diagnostic need in these rural communities. Similar abnormality rates have been reported in community-based chest X-ray screening initiatives in other high TB burden settings and are often linked to delayed health-seeking behaviour, limited access to diagnostic services, and prolonged exposure to environmental and occupational risk factors ( 33 , 34 ) Chest radiography remains one of the most effective population-level tools for detecting pulmonary disease, including TB and other cardiopulmonary conditions ( 18 ). The integration of computer-aided detection (CAD) systems has further improved the feasibility of large-scale screening in resource-limited settings by enabling rapid interpretation and reducing reliance on specialist radiologists. Several studies have demonstrated that AI-based CAD systems can achieve diagnostic performance comparable to trained human readers for TB screening and are increasingly recommended for use in high-burden settings( 16 , 35 , 36 ). In communities with limited access to routine diagnostic services, AI-supported chest radiography therefore offers an effective triage tool for identifying both TB and other pulmonary abnormalities. The high abnormality yield observed in this study supports the rationale for integrated screening models that address multiple cardiopulmonary conditions rather than focusing solely on TB ( 29 , 37 ). Gender patterns in screening participation and disease detection Clear gender patterns emerged across both screening participation and disease detection. Overall, females constituted a higher proportion of individuals screened, although regional differences were observed. In Ebonyi State, females represented the majority of participants screened, whereas in Nasarawa State males accounted for a larger proportion of screened individuals. These patterns likely reflect regional differences in gender norms, health-seeking behaviour, and participation in community mobilisation activities. Evidence from southern Nigeria suggests that women often engage more actively in community-based preventive health services, while studies from northern and north-central Nigeria report lower female participation due to mobility constraints and household decision-making dynamics affecting healthcare access ( 38 ). Despite higher female participation in screening, men accounted for a greater proportion of presumptive TB cases and bacteriologically confirmed TB diagnoses (60%), consistent with global epidemiological evidence showing higher TB incidence among men( 5 ). Occupational exposures, tobacco use, alcohol consumption, and delayed care-seeking behaviour among men are recognised contributors to this disparity.( 39 ) In contrast, women represented the majority of individuals with abnormalities suggestive of CVDs and CRDs. Approximately 25% of individuals screened had abnormalities suggestive of cardiovascular or chronic respiratory disease, with women accounting for about 75% of these cases. This pattern may reflect gendered differences in environmental exposures and access to preventive care. For example, prolonged exposure to household air pollution from biomass fuel use has been linked to chronic respiratory disease among women in rural settings ( 40 ). Similarly, women in many LMICs experience lower rates of routine cardiovascular screening and treatment, contributing to delayed detection of hypertension and related conditions ( 7 , 32 ). These findings highlight the importance of incorporating gender-sensitive strategies into integrated screening programmes to address both biological and social determinants of health. Emerging burden of NCDs in rural settings Another important finding of this study is the substantial burden of NCD-related abnormalities detected during screening. Among the 3,166 abnormal radiographs, approximately 2,367 (75%) showed features suggestive of CVDs or CRDs, indicating that the majority of abnormalities identified were related to non-TB conditions. These findings challenge the perception that NCDs are primarily urban health problems and instead highlight the ongoing epidemiological transition affecting rural populations in LMICs ( 40 ) Evidence from Nigeria and other sub-Saharan African countries has documented increasing prevalence of hypertension, cardiovascular risk factors, and chronic respiratory disease in rural communities ( 7 ). Environmental exposures such as biomass fuel use, alongside changing lifestyles and ageing populations, are contributing to this growing burden ( 40 ). Integrated screening initiatives such as INTEGRATE TB therefore provide valuable community-level data that can inform health system planning and support the decentralisation of NCD services within primary healthcare platforms. Without such integrated approaches, many of these conditions would likely remain undiagnosed until advanced stages of disease( 29 ). Health system constraints and linkage to care The contrasting linkage outcomes for TB and NCD-related abnormalities highlight important differences in health system capacity. The 95% treatment initiation rate among bacteriologically confirmed TB cases reflects the relative maturity of Nigeria’s TB programme. TB services are decentralised, free at the point of care, and embedded within primary healthcare facilities through established diagnostic and treatment pathways ( 5 ). Community-based active case finding supported by AI-enabled chest X-ray screening, therefore enabled timely diagnosis and treatment initiation, consistent with findings from other TB active case-finding initiatives ( 34 ). In contrast, linkage to care for individuals with abnormalities suggestive of CVD or CRD was substantially lower. Although referrals were issued, only 12% completed the referral, defined as attendance at a treatment facility for further evaluation and care. This gap reflects systemic constraints rather than limitations of the screening approach itself. In many rural Nigerian settings, CVD and CRD services are not available at the primary healthcare level and require referral to higher-level facilities. This referral-dependent model introduces barriers related to travel distance, transport costs, and out-of-pocket expenditure, contributing to attrition along the referral cascade ( 32 ). These findings highlight a key challenge for integrated screening programmes: detection alone does not guarantee access to treatment when health systems lack sufficient capacity to manage identified conditions. Strengthening primary healthcare capacity will therefore be essential for improving the impact of integrated screening interventions. The WHO Package of Essential Noncommunicable Disease Interventions (WHO PEN) recommends decentralising basic NCD services to primary care facilities, including hypertension screening, diabetes testing, and access to essential medicines ( 41 ). Integrating such services within existing TB programme platforms could help reduce fragmentation and improve continuity of care. Operational lessons from community-based screening A small proportion of pre-treatment loss to follow-up occurred among individuals diagnosed with TB. Of the 204 bacteriologically confirmed TB cases, approximately 5% were not initiated on treatment, reflecting the challenges of tracking individuals identified through community outreach activities, particularly in large congregate settings such as markets. Similar patterns have been observed in other community-based TB case-finding programmes ( 34 ) To mitigate this risk, the intervention engaged trusted community volunteers to support participant tracking, communicate results, and facilitate referral and treatment initiation. Community-led approaches have been shown to improve linkage to care by leveraging local trust and social networks and are recommended in WHO guidance on community-based TB care ( 22 ). Implications for integrated service delivery The findings suggest that integrated screening programmes will be most effective when aligned with accessible treatment services. TB screening platforms can serve as entry points for broader cardiopulmonary health assessment, but meaningful improvements in NCD outcomes will require strengthening primary healthcare capacity to deliver basic services such as blood pressure measurement, chronic respiratory assessment, and initial treatment ( 41 ). Combining AI-enabled chest X-ray screening with complementary low-cost diagnostic tools could improve early identification of cardiovascular risk while reducing unnecessary referrals. Sustained community engagement, structured referral systems, and the involvement of trusted community volunteers should remain central components of integrated outreach models. Strengths and limitations A major strength of this study is its implementation under routine programme conditions in rural communities, generating operational evidence on the feasibility of AI-enabled integrated screening. Standardised screening protocols, digital data systems, and ongoing supervision supported implementation fidelity and data quality. Several limitations should also be considered. Imaging artefacts caused by patterned clothing produced false-positive readings during early implementation, although mitigation measures were subsequently introduced. In addition, referral completion for non-TB conditions may be underestimated because the study tracked only individuals attending designated referral facilities and some participants may have sought care elsewhere. Abbreviations AI Artificial Intelligence CVD Cardiovascular Disease CRD Chronic Respiratory Disease DOTS Directly Observed Treatment, Short-course LGA Local Government Area PDX Portable Digital X-ray TB Tuberculosis Declarations Competing interests: J.C. and T.R. are affiliated with the Stop TB Partnership, which supports the TB REACH initiative that funded this work. They were not involved in funding decisions but provided technical support. All other authors declare that they have no competing interests. Funding: The intervention was implemented by the Catholic Caritas Foundation of Nigeria (CCFN) in partnership with the Zankli Research Centre, Bingham University, Nigeria, under the TB REACH Wave 10 initiative funded by the Stop TB Partnership, in collaboration with State and Local Government Tuberculosis Programmes and tertiary referral hospitals. Author Contribution C.O., J.I., and C.U. conceptualised the study, and C.O. led implementation. C.O., J.I., and C.U. designed the study methodology. C.O., C.U., and R.F. supervised data collection and analysis. C.O. and C.U. drafted the manuscript. J.C. and other co-authors provided critical revisions and technical input. All authors reviewed and approved the final manuscript. Acknowledgement The authors acknowledge the contributions of community volunteers, field teams, and State Tuberculosis and Leprosy Control Programme staff in Ebonyi and Nasarawa States for their support in implementing the outreach activities. We particularly recognise the State Tuberculosis and Leprosy Control Programme Managers in both states, the National Tuberculosis Programme, and Qure.ai for their technical support. We also thank the participating communities for their cooperation and engagement. Data Availability The datasets generated and/or analysed are available from the corresponding author on request and with permission from the Catholic Caritas Foundation of Nigeria and the relevant State TB and leprosy control program. References World Health Organization. Noncommunicable diseases [Internet]. 2023. Available from: https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases Omran AR. The epidemiologic transition: a theory of the epidemiology of population change. Milbank Mem Fund Q. 1971;49(4):509–38. World Health Organization. The double burden of malnutrition [Internet]. World Health Organization; 2020. Report No. Available from: https://www.who.int/publications/i/item/WHO-NMH-NHD-20.6 Popkin BM, Corvalan C, Grummer-Strawn LM. Dynamics of the double burden of malnutrition. The Lancet. 2020;395:65–74. doi:10.1016/S0140-6736(19)32497-3 World Health Organization. Global tuberculosis report 2024 [Internet]. World Health Organization; 2024. Report No. Available from: https://www.who.int/publications/i/item/9789240101531 GBD 2019 Diseases and Injuries Collaborators. Global burden of diseases results tool: Nigeria. The Lancet [Internet]. 2020. Available from: https://vizhub.healthdata.org/gbd-results/ Adeloye D, Basquill C. Estimated prevalence and control of hypertension in Nigeria. J Hypertens. 2014;32(12):2307–15. doi:10.1097/HJH.0000000000000380 World Health Organization. STEPwise approach to NCD risk factor surveillance: Nigeria factsheet [Internet]. World Health Organization; 2018. Report No. Available from: https://www.who.int/teams/noncommunicable-diseases/surveillance/systems-tools/steps Mbakwem AC, Ajuluchukwu JN, Amadi C. Cardiovascular disease admissions and mortality in Lagos, Nigeria. Cardiovasc J Afr. 2016;27(2):84–9. doi:10.5830/CVJA-2015-088 World Health Organization. Global tuberculosis report 2023: Nigeria country profile [Internet]. World Health Organization; 2023. Report No. Available from: https://www.who.int/teams/global-tuberculosis-programme/data Federal Ministry of Health Nigeria. National Multi-Sectoral Action Plan for the Prevention and Control of Non-Communicable Diseases (2019–2025) [Internet]. Federal Ministry of Health Nigeria; 2019. Report No. Available from: https://www.health.gov.ng/doc/NCDs_ACTION_PLAN.pdf Akinwumi AF, Esimai OA, Arije O, Ojo TO, Esan OT. Preparedness of primary health care facilities on implementation of essential non-communicable disease interventions in osun state south-west Nigeria: a rural-urban comparative study. BMC Health Serv Res. 2023 Feb 14;23(1):154. doi:10.1186/s12913-023-09138-8 PubMed PMID: 36788557; PubMed Central PMCID: PMC9930353. https://www.who.int/publications/i/item/WHO-HTM-TB-2016.20. van’t Hoog AH. Systematic screening for tuberculosis using chest radiography. Int J Tuberc Lung Dis. 2014;18(11):1228–36. doi:10.5588/ijtld.14.0315 Qin ZZ. Using artificial intelligence to read chest radiographs for tuberculosis detection. Eur Respir J. 2019;54. doi:10.1183/13993003.01023-2019 Philipsen RH. Diagnostic accuracy of computer-aided detection for tuberculosis on chest radiography. Clin Infect Dis. 2022;74(11):2023–32. doi:10.1093/cid/ciab639 Pande T. Computer-aided detection for tuberculosis screening. PLOS ONE. 2020;15(2):e0227479. doi:10.1371/journal.pone.0227479 World Health Organization. WHO consolidated guidelines on tuberculosis: module 2 – screening [Internet]. World Health Organization; 2021. Report No. Available from: https://www.who.int/publications/i/item/9789240022676 McAdams HP. Radiographic manifestations of chronic lung and cardiac disease. Radiographics. 2015;35(5):1289–303. doi:10.1148/rg.2015140310 Hoffmann TC, Glasziou PP, Boutron I. Better reporting of interventions: the TIDieR checklist and guide. BMJ. 2014;348:g1687. World Health Organization. Systematic screening for active tuberculosis: principles and recommendations [Internet]. Geneva: World Health Organization; 2013. Available from: https://www.who.int/publications/i/item/9789241548601 World Health Organization. WHO consolidated guidelines on tuberculosis: Module 3 – Diagnosis [Internet]. Geneva: World Health Organization; 2021. Available from: https://www.who.int/publications/i/item/9789240029415 World Health Organization. The use of digital chest radiography for tuberculosis detection and diagnosis [Internet]. Geneva: World Health Organization; 2016. Available from: https://www.who.int/publications/i/item/WHO-HTM-TB-2016.20 World Health Organization. WHO operational handbook on tuberculosis: Module 2 – Screening [Internet]. Geneva: World Health Organization; 2022. Available from: https://www.who.int/publications/i/item/9789240057562 Stop TB Partnership. TB REACH: Finding the missing people with TB [Internet]. Geneva: Stop TB Partnership; 2023. Available from: https://stoptb.org/what-we-do/tb-reach McAdams HP, Erasmus JJ, Rosado-de-Christenson ML. Radiographic manifestations of chronic lung and cardiac disease. Radiographics. 2015;35(5):1289–303. World Medical Association. Declaration of Helsinki: Ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191–4. World Health Organization. Ethical standards for tuberculosis care and prevention [Internet]. Geneva: World Health Organization; 2017. Available from: https://www.who.int/publications/i/item/WHO-HTM-TB-2017.28 Foo CD, Shrestha P, Wang L, Du Q, García-Basteiro AL, Abdullah AS, et al. Integrating tuberculosis and noncommunicable diseases care in low- and middle-income countries (LMICs): A systematic review. PLoS Med. 2022 Jan;19(1):e1003899. doi:10.1371/journal.pmed.1003899 PubMed PMID: 35041654; PubMed Central PMCID: PMC8806070. Calderwood CJ, Timire C, Mavodza C, Kavenga F, Ngwenya M, Madziva K, et al. Beyond tuberculosis: a person-centred and rights-based approach to screening for household contacts. Lancet Glob Health. 2024 Mar 1;12(3):e509–15. doi:10.1016/S2214-109X(23)00544-2 PubMed PMID: 38365421. Nugent R, Bertram MY, Jan S, Niessen LW, Sassi F, Jamison DT, et al. Investing in non-communicable disease prevention and management to advance the Sustainable Development Goals. The Lancet. 2018 May 19;391(10134):2029–35. doi:10.1016/S0140-6736(18)30667-6 PubMed PMID: 29627167. Atun R, Jaffar S, Nishtar S, Knaul FM, Barreto ML, Nyirenda M, et al. Improving responsiveness of health systems to non-communicable diseases. The Lancet. 2013 Feb 23;381(9867):690–7. doi:10.1016/S0140-6736(13)60063-X Oni T, Burke R, Tsekela R, Bangani N, Seldon R, Gideon HP, et al. High prevalence of subclinical tuberculosis in HIV-1-infected persons without advanced immunodeficiency: implications for TB screening. Thorax. 2011 Aug;66(8):669–73. doi:10.1136/thx.2011.160168 PubMed PMID: 21632522; PubMed Central PMCID: PMC3142344. Hamada Y, Mukora R, Pelusa R, Ntshiqa T, Shedrawy J, Velen K, et al. Costs and cost-effectiveness of integrated screening for non-communicable diseases in TB contacts. IJTLD Open. 2025 Mar;2(3):160–5. doi:10.5588/ijtldopen.24.0625 PubMed PMID: 40092516; PubMed Central PMCID: PMC11906026. Qin ZZ, Ahmed S, Sarker MS. Using artificial intelligence to read chest radiographs for tuberculosis detection: a multi-site evaluation of diagnostic accuracy. Lancet Digit Health. 2021;3(9):e595–604. Scott AJ, Perumal T, Pooran A, Oelofse S, Jaumdally S, Swanepoel J, et al. Clinical evaluation of computer-aided digital x-ray detection of pulmonary tuberculosis during community-based screening or active case-finding: a case–control study. Lancet Glob Health. 2025 Mar 1;13(3):e517–27. doi:10.1016/S2214-109X(24)00516-3 PubMed PMID: 40021309. Beaglehole R, Bonita R, Horton R, Adams C, Alleyne G, Asaria P, et al. Priority actions for the non-communicable disease crisis. The Lancet. 2011 Apr 23;377(9775):1438–47. doi:10.1016/S0140-6736(11)60393-0 PubMed PMID: 21474174. Adeloye D, Basquill C, Aderemi AV, Thompson JY, Obi FA. An estimate of the prevalence of hypertension in Nigeria: a systematic review and meta-analysis. J Hypertens. 2015 Feb;33(2):230–42. doi:10.1097/HJH.0000000000000413 PubMed PMID: 25380154. Kc H, P M, Rm H, Rg W, El C. Sex Differences in Tuberculosis Burden and Notifications in Low- and Middle-Income Countries: A Systematic Review and Meta-analysis. PLoS Med. 2016 Sep 6;13(9). doi:10.1371/journal.pmed.1002119 PubMed PMID: 27598345. Bigna JJ, Noubiap JJ. The rising burden of non-communicable diseases in sub-Saharan Africa. Lancet Glob Health. 2019 Oct 1;7(10):e1295–6. doi:10.1016/S2214-109X(19)30370-5 PubMed PMID: 31537347. WHO package of essential noncommunicable (PEN) disease interventions for primary health care [Internet]. [cited 2025 Jun 7]. Available from: https://www.who.int/publications/i/item/9789240009226 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 28 Apr, 2026 Reviews received at journal 24 Apr, 2026 Reviews received at journal 14 Apr, 2026 Reviewers agreed at journal 12 Apr, 2026 Reviewers agreed at journal 12 Apr, 2026 Reviews received at journal 09 Apr, 2026 Reviews received at journal 06 Apr, 2026 Reviewers agreed at journal 06 Apr, 2026 Reviewers agreed at journal 01 Apr, 2026 Reviewers invited by journal 01 Apr, 2026 Editor assigned by journal 23 Mar, 2026 Submission checks completed at journal 23 Mar, 2026 First submitted to journal 20 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9180615","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":615762657,"identity":"77e8555d-35d5-4d79-a8ac-25926027d9d7","order_by":0,"name":"Chidimma Okoye","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyUlEQVRIiWNgGAWjYLACxgYJOSD5gDQtxgwMzAYkaWFIbCBaC397+8MHH3dYpPe3H2Z7+IXBJl/egYAWiTMHkg1nnpHInXEmmd1YhiHNcuMBAloMJBKOSfO2SeRukOA/Ji3BcNjAsIGQFvmH7b+BWtINJJjZiNQCVMkM1JIAYkh+AGqRJ6AD6Jc0ZkmgXwyBfmGTZjBIMyAYbvztxx9++LijTp4fGGKSPypsDOQJOQwFMPMArTA4QIoWxh9AgjRbRsEoGAWjYCQAAKO0OEwZL5UpAAAAAElFTkSuQmCC","orcid":"","institution":"Catholic Caritas foundation of Nigeria","correspondingAuthor":true,"prefix":"","firstName":"Chidimma","middleName":"","lastName":"Okoye","suffix":""},{"id":615762659,"identity":"5c7546cb-fe03-4f1c-95cb-a8fcaade9a87","order_by":1,"name":"Jude Ilozumba","email":"","orcid":"","institution":"Catholic Caritas foundation of Nigeria","correspondingAuthor":false,"prefix":"","firstName":"Jude","middleName":"","lastName":"Ilozumba","suffix":""},{"id":615762661,"identity":"1c58148a-2884-4e16-b1a6-306421327733","order_by":2,"name":"John Okoh","email":"","orcid":"","institution":"Catholic Caritas foundation of Nigeria","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"","lastName":"Okoh","suffix":""},{"id":615762663,"identity":"58114c1c-9a7b-43fd-951a-cb15a5fa7b3f","order_by":3,"name":"Amana Effiong","email":"","orcid":"","institution":"Catholic Caritas foundation of Nigeria","correspondingAuthor":false,"prefix":"","firstName":"Amana","middleName":"","lastName":"Effiong","suffix":""},{"id":615762665,"identity":"0b478e9e-ffd0-4e13-a2ae-2d378beef561","order_by":4,"name":"Young Oluokun","email":"","orcid":"","institution":"Catholic Caritas foundation of Nigeria","correspondingAuthor":false,"prefix":"","firstName":"Young","middleName":"","lastName":"Oluokun","suffix":""},{"id":615762666,"identity":"1625d56a-2c36-4eb8-99cd-ef9de81d42d2","order_by":5,"name":"Chinwe Eze","email":"","orcid":"","institution":"PRIME Data Nexus","correspondingAuthor":false,"prefix":"","firstName":"Chinwe","middleName":"","lastName":"Eze","suffix":""},{"id":615762667,"identity":"f0b2f129-aaeb-4002-9d09-fbd36d24fb6a","order_by":6,"name":"Obioma akaniro","email":"","orcid":"","institution":"National TB and Leprosy control Program","correspondingAuthor":false,"prefix":"","firstName":"Obioma","middleName":"","lastName":"akaniro","suffix":""},{"id":615762668,"identity":"a14eba17-94d2-46f6-affc-52c5c80dd923","order_by":7,"name":"Emperor Ubochioma","email":"","orcid":"","institution":"National TB and Leprosy control Program","correspondingAuthor":false,"prefix":"","firstName":"Emperor","middleName":"","lastName":"Ubochioma","suffix":""},{"id":615762669,"identity":"be8d49c4-663b-41a4-8ca1-91abe3b06287","order_by":8,"name":"Rena Fukunanga","email":"","orcid":"","institution":"Centers for Disease Control and Prevention","correspondingAuthor":false,"prefix":"","firstName":"Rena","middleName":"","lastName":"Fukunanga","suffix":""},{"id":615762670,"identity":"2096d8ef-9754-4515-8e31-69a9c0e23c04","order_by":9,"name":"Toufiq Raham","email":"","orcid":"","institution":"Stop TB Partnership Geneva","correspondingAuthor":false,"prefix":"","firstName":"Toufiq","middleName":"","lastName":"Raham","suffix":""},{"id":615762671,"identity":"40e966b7-ac10-466c-8772-5f04d3c648e3","order_by":10,"name":"John Bimba","email":"","orcid":"","institution":"Zankli Research Center Bingham University,Nigeria","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"","lastName":"Bimba","suffix":""},{"id":615762672,"identity":"dba181c6-99f9-4519-a42d-dadf0e83a51e","order_by":11,"name":"Jacob Creswell","email":"","orcid":"","institution":"Stop TB Partnership Geneva","correspondingAuthor":false,"prefix":"","firstName":"Jacob","middleName":"","lastName":"Creswell","suffix":""},{"id":615762673,"identity":"fd10fd54-2bd5-44d1-bce8-f46f56090c71","order_by":12,"name":"Chukwuebuka Ugwu","email":"","orcid":"","institution":"Center for Tuberculosis Research Liverpool school of Tropical Medicine UK","correspondingAuthor":false,"prefix":"","firstName":"Chukwuebuka","middleName":"","lastName":"Ugwu","suffix":""}],"badges":[],"createdAt":"2026-03-20 16:08:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9180615/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9180615/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106308512,"identity":"305c6205-367e-4ff6-8384-9fc178244087","added_by":"auto","created_at":"2026-04-07 10:12:10","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":181101,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Result section.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9180615/v1/a9bc2a2cf78ece65297a9970.png"},{"id":106308894,"identity":"159a93f4-4f78-4f8f-8d82-f4adad384632","added_by":"auto","created_at":"2026-04-07 10:14:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1106184,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9180615/v1/460dd66e-3484-43b3-a786-51a1f79f677c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Implementing AI-enabled chest X-ray for community-based integrated screening for tuberculosis, chronic respiratory diseases, and cardiovascular diseases in Nigeria","fulltext":[{"header":"Background","content":"\u003cp\u003eLow- and middle-income countries (LMICs) are experiencing a growing double burden of disease, in which communicable diseases coexist with an increasing prevalence of non-communicable diseases (NCDs) within the same population(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). This reflects overlapping demographic and epidemiological transitions, including population growth, increased life expectancy, urbanisation, and lifestyle changes(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). In parallel, many LMICs face a double burden of malnutrition, with undernutrition and overnutrition occurring concurrently, further increasing vulnerability to both infectious and chronic conditions (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). These converging trends place sustained pressure on health systems that remain largely organised around vertical, disease-specific programmes.\u003c/p\u003e \u003cp\u003eNigeria exemplifies this challenge, bearing the highest tuberculosis (TB) burden in Africa, alongside persistent gaps in the prevention and management of drug-resistant TB and TB/HIV co-infection(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). TB remains among the leading causes of death nationally. The World Health Organization estimated approximately 56,000 TB deaths among HIV-negative individuals in Nigeria in 2024, with an additional 5,800 deaths among people living with HIV. (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e)\u0026rdquo; At the same time, the burden of NCDs has increased steadily. In 2019, NCDs accounted for approximately 29% of all deaths in Nigeria(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Cardiovascular diseases (CVDs) contributed about 11% of all deaths, while chronic respiratory diseases (CRDs) accounted for approximately 2%, alongside a substantial loss of disability-adjusted life years (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHypertension, the leading modifiable risk factor for cardiovascular disease, affects over 30% of Nigerian adults and contributes substantially to myocardial infarction and stroke(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). National evidence also indicates a high prevalence of additional CVD risk factors, including overweight and obesity, diabetes mellitus, physical inactivity, tobacco use, exposure to household air pollution, and unhealthy dietary patterns(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Hospital-based studies further demonstrate a sustained rise in CVD-related admissions and mortality. A 15-year review from Lagos reported that CVDs accounted for 20.8% of all medical admissions and 30.4% of in-hospital mortality, underscoring the growing clinical and economic burden of chronic diseases in Nigeria (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNigeria\u0026rsquo;s primary healthcare system is relatively better resourced for the surveillance and management of endemic communicable diseases such as malaria, TB, and HIV, largely through established vertical programmes and sustained donor support (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). While TB case detection and treatment coverage improved between 2019 and 2023 (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e), Access to NCD prevention, diagnosis, and long-term care remains limited, particularly at the primary healthcare level. For example, a recent assessment of primary healthcare facility readiness in Nigeria found that many facilities lacked essential diagnostic tools, trained personnel, and medicines required to implement basic WHO Package of Essential NCD Interventions (PEN), limiting routine screening and management of conditions such as hypertension and diabetes (\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Services for CVDs and CRDs are often unavailable or fragmented at facilities closest to communities, resulting in delayed diagnoses, high out-of-pocket expenditure, and poor continuity of care. These gaps undermine progress toward universal health coverage and disproportionately affect rural and hard-to-reach populations (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIntegrated, community-based approaches that combine screening for communicable and non-communicable diseases are therefore increasingly important. Systematic chest X-ray\u0026ndash;based screening has re-emerged as an effective strategy for TB case finding in high-burden settings (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). AI\u0026ndash;enabled chest X-ray interpretation systems have demonstrated high accuracy for TB screening and increasing utility in identifying non-TB abnormalities, while reducing reliance on scarce specialist expertise (\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). In 2021, the World Health Organization endorsed the use of AI-based chest X-ray interpretation for TB screening, recognising its potential to expand access, improve efficiency, and strengthen case detection in resource-limited settings (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Evidence also shows that chest radiography can identify abnormalities suggestive of CRDs and CVDs, including cardiomegaly and chronic lung changes (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe INTEGRATE TB project was designed to apply community-based, AI-enabled chest X-ray screening for tuberculosis, while identifying radiographic abnormalities suggestive of cardiovascular and chronic respiratory diseases in remote rural communities in Nigeria. By using TB screening as an entry point for broader lung and cardiovascular health assessment, the project aimed to expand access to diagnostic services for underserved populations and to document implementation experiences within routine programme settings. This paper describes the implementation approach, summarises findings across the screening and referral cascade, and highlights key operational considerations, guided by the Template for Intervention Description and Replication (TIDieR) framework (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e).\u003c/p\u003e"},{"header":"Methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy design and setting\u003c/h2\u003e \u003cp\u003eThis study was a non-randomised implementation study using routinely collected programme data to evaluate a community-based screening intervention. It was conducted across five Local Government Areas (LGAs) in Nigeria: three LGAs in Ebonyi State (South-East) and two LGAs in Nasarawa State (North-Central), between January 2023 and December 2024.\u003c/p\u003e \u003cp\u003e Communities were selected using the State Primary Health Care Development Agency annual Updated Directory of Settlements, 2022. Ministry of Health; 2022 and restricted to those classified as hard-to-reach, based on geographic remoteness, limited transport connectivity, and poor access to health facilities, consistent with WHO systematic screening guidance (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIntervention description\u003c/h3\u003e\n\u003cp\u003eWe evaluated the screening cascade of a community-based TB screening programme that used AI-enabled chest X-ray to simultaneously screen for tuberculosis and identify radiographic abnormalities suggestive of CVDs and CRDs. The intervention used portable digital chest X-ray (PDX) machines equipped with artificial intelligence (AI) software to enable real-time interpretation during outreach activities, in line with WHO recommendations on systematic TB screening and use of chest radiography (\u003cspan additionalcitationids=\"CR22 CR23\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTuberculosis screening served as the primary entry point, with non-TB abnormalities identified and referred through established pathways. The intervention aimed to improve early detection, strengthen linkage to care, and generate operational evidence on the feasibility of AI-enabled, multi-disease screening in underserved settings (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eCommunity engagement and delivery\u003c/h3\u003e\n\u003cp\u003ePrior to implementation, structured engagement was conducted with community leaders, women\u0026rsquo;s and youth groups, religious institutions, and local government health teams to introduce the intervention and adapt implementation plans to local contexts, consistent with WHO recommendations for community-based TB screening (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Outreach schedules were aligned with community calendars, including market days and cultural events.\u003c/p\u003e \u003cp\u003eCommunity volunteers were recruited from respected community members to support mobilisation, health education, and participant flow. Health education sessions were delivered in local languages using visual aids and public address systems, following best practices for community TB care and prevention (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eOutreach implementation and eligibility\u003c/h3\u003e\n\u003cp\u003eOutreach teams comprised a radiographer, a linkage coordinator, the Local Government TB and Leprosy Supervisor (LGTBLS), community volunteers, an administrative assistant responsible for real-time data entry, and CCFN supervisory staff. Outreach activities were conducted in accessible public spaces such as markets, town halls, and school compounds.\u003c/p\u003e \u003cp\u003eIndividuals aged six years and above were eligible for screening, excluding pregnant women. Verbal informed consent was obtained from adult participants, and parental consent was obtained for minors. Following consent, participants underwent symptom screening and chest radiography, with AI-generated results available within an average of 3 minutes.\u003c/p\u003e\n\u003ch3\u003eScreening tools and procedures\u003c/h3\u003e\n\u003cp\u003ePortable digital chest X-ray machines (Mini X-ray, Northbrook, IL, USA) were integrated with qXR Version 3, a CE-certified and WHO recommended AI software developed by Qure.ai (Mumbai, India). The software automatically analysed chest radiographs, generating a TB abnormality score ranging from 0.01 to 0.99 and also identifies non-TB abnormalities.\u003c/p\u003e \u003cp\u003e Chest radiographs were analysed using qXR Version 3 (Qure.ai), which generates a TB abnormality score ranging from 0.01 to 0.99 and provides localisation of both TB-related and non-TB radiographic findings.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eAbnormal chest radiograph:\u003c/h2\u003e \u003cp\u003eA chest radiograph was classified as abnormal if the AI software localized any radiographic abnormalities, including TB-related abnormalities (e.g., cavities, nodules, consolidation) or non-TB abnormalities (e.g., cardiomegaly, fibrosis, pleural effusion, or structural changes), based on predefined classification outputs of the qXR algorithm.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePresumptive TB:\u003c/h3\u003e\n\u003cp\u003eDefined as the presence of TB-related symptoms (using the WHO four-symptom screening tool: cough, fever, weight loss, or night sweats) and/or an AI-generated qXR score\u0026thinsp;\u0026ge;\u0026thinsp;0.5, in line with the manufacturer's default threshold and prior validation studies of AI-based chest X-ray triage\u003c/p\u003e\n\u003ch3\u003eNon-TB abnormalities (CVDs and CRDs):\u003c/h3\u003e\n\u003cp\u003eRadiographic abnormalities were categorised as suggestive of CVDs or CRDs based on AI-generated outputs. A decision-support algorithm, developed in collaboration with cardiopulmonary specialists, guided the categorisation and referral of non-TB abnormalities based on structured outputs generated by the qXR artificial intelligence software.\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eCVD-related findings included features such as cardiomegaly.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eCRD-related findings included pulmonary and pleural abnormalities such as fibrosis, consolidation, reticulonodular patterns, pleural effusion, and other chronic lung changes.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThese classifications were used to guide referral for further clinical evaluation and were not considered definitive diagnoses.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDiagnostic and referral pathways\u003c/h2\u003e \u003cp\u003ePresumptive TB cases provided sputum samples on site, which were transported to nearby GeneXpert laboratories following WHO-recommended biosafety and triple-packaging protocols(\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). Bacteriologically confirmed TB cases were linked to treatment at the nearest DOTS facility or a facility preferred by the participant, in line with Nigeria\u0026rsquo;s national TB guidelines.\u003c/p\u003e \u003cp\u003eFor individuals with TB-suggestive radiographic abnormalities identified by AI but with negative GeneXpert results or inability to produce sputum, chest radiographs and clinical histories were reviewed remotely by radiologists via a secure cloud-based platform. Based on this review, a clinical decision was made to initiate TB treatment or pursue alternative management in accordance with national TB guidelines and WHO diagnostic recommendations (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e Participants with non-TB abnormalities suggestive of CVDs or CRDs were referred to partnering tertiary hospitals: Federal Teaching Hospital Abakaliki (Ebonyi State) and Dalhatu Araf Specialist Hospital (Nasarawa State). Pre-implementation advocacy established expedited referral pathways, allowing referred clients to bypass routine general outpatient procedures and access specialist care directly, informed by radiographic criteria for chronic lung and cardiac disease. While no financial support for treatment was provided, logistical assistance was offered to reduce access barriers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eFidelity, supervision, and quality assurance\u003c/h2\u003e \u003cp\u003eStandard operating procedures and job aids guided all field activities. Field teams received initial and refresher training, supported by routine supervision. Digital data collection tools incorporated validation checks to enhance data quality, consistent with WHO recommendations for programme monitoring (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWeekly supervisory visits and monthly review meetings assessed protocol adherence and implementation challenges. A dedicated WhatsApp platform facilitated real-time troubleshooting, and monthly feedback was shared with stakeholders to support continuous quality improvement.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eData collection and analysis\u003c/h2\u003e \u003cp\u003e Data were collected using the electronic Q-Track platform provided with the AI model, and complementary paper-based registers capturing demographic characteristics, AI screening outputs, laboratory results, and referral outcomes. Paper records were routinely cross-checked against electronic data to ensure consistency.\u003c/p\u003e \u003cp\u003eData were cleaned and analysed using Microsoft Excel. Descriptive statistics were used to summarise participant characteristics, screening outcomes, diagnostic yield, and linkage-to-care rates, consistent with recommended approaches for implementation-focused TB screening evaluations (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e)\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eEthical considerations\u003c/h2\u003e \u003cp\u003eEthical approval was obtained from the National Health Research Ethics Committee (NHREC: NHREC/2024/01/137/13-06-24). Permission to conduct the intervention was granted by State Ministries of Health and State Tuberculosis Programmes. Participation was voluntary, informed consent was obtained from all participants or their guardians, and all data were anonymised prior to analysis to ensure confidentiality, in line with WHO ethical standards for TB care and the Declaration of Helsinki (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e).\u003c/p\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n\u003ch2\u003eParticipant characteristics and screening volume\u003c/h2\u003e\n\u003cp\u003eBetween January 2023 and December 2024, a total of 9,585 individuals were screened using AI-integrated portable digital chest X-ray through 93 community-based outreach activities conducted across five Local Government Areas (LGAs) in Ebonyi and Nasarawa States. Of those screened, 4,053 (42%) were male, and 5,532 (58%) were female.\u003c/p\u003e\n\u003cp\u003eScreening volume ranged from 80 to 120 individuals per outreach day, with variations influenced by weather conditions and community mobilisation. Communities with larger populations or higher screening yields were prioritised for repeat outreach visits. Screening volume and outcomes disaggregated by LGA are presented in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n\u003ch2\u003eRadiographic findings and AI outputs\u003c/h2\u003e\n\u003cp\u003eOf the 9,585 chest radiographs acquired during community outreach activities, 6,419 (67%) were classified as normal, while 3,166 (33%) were flagged as abnormal by the AI software. Among individuals with normal radiographs, 2,958 (46%) were male, and 3461 (54%) were female. Of those with abnormal radiographs, 1,095 (35%) were male, and 2,071 (65%) were female.\u003c/p\u003e\n\u003cp\u003eThe proportion of abnormal radiographs was higher in Ebonyi State (35%) compared with Nasarawa State (29%). Abnormality rates further disaggregated by sex and Local Government Area are presented in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n\u003ch2\u003eTuberculosis screening, diagnosis, and linkage to care\u003c/h2\u003e\n\u003cp\u003eA total of 1,336 (14%) individuals were classified as presumptive tuberculosis (TB) cases based on the presence of TB-related symptoms, an AI-generated qXR score\u0026thinsp;\u0026ge;\u0026thinsp;0.5, or both. Of these, 800 (60%) individuals were identified based on an AI score\u0026thinsp;\u0026ge;\u0026thinsp;0.5, while 536 (40%) individuals were identified based on symptoms alone with qXR scores\u0026thinsp;\u0026lt;\u0026thinsp;0.5. Overall, presumptive TB cases comprised 694(52%) males and 642(48%) females. Among presumptive TB cases, 1,123 individuals (84%) were able to produce spot sputum samples. GeneXpert testing identified 204 individuals as bacteriologically confirmed TB cases, representing 18% of those tested. Of these confirmed cases, 124(61%) were male, and 80 (39%) were female.\u003c/p\u003e\n\u003cp\u003eLinkage-to-care data showed that 194 (95%) of bacteriologically confirmed TB cases were successfully initiated on treatment at a DOTS facility, either at the nearest facility or at a facility preferred by the patient. In addition, 199 individuals with TB-suggestive radiographic abnormalities but negative GeneXpert results or inability to produce sputum were clinically diagnosed and initiated on TB treatment following radiologist review of chest radiographs and clinical assessment. These clinically diagnosed cases included 114 males and 85 females.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n\u003ch2\u003eNon-TB abnormalities and referral outcomes\u003c/h2\u003e\n\u003cp\u003eAmong the 3,166 abnormal chest radiographs, the AI software identified 2,367 individuals (75%) with abnormalities suggestive of non-TB conditions, including cardiovascular diseases (CVDs) and chronic respiratory diseases (CRDs). Specifically, 852 individuals had radiographic findings suggestive of CVDs, and 1,514 individuals had findings suggestive of CRDs. All individuals with non-TB abnormalities were referred to partnering tertiary hospitals for further clinical or laboratory evaluation. Of those referred, 121 individuals with suspected CVDs and 164 individuals with suspected CRDs were confirmed to have reached treatment facilities. Clinical or laboratory assessment confirmed 107 CVD cases and 147 CRD cases.\u003c/p\u003e\n\u003cp\u003eAt the three-month follow-up, 12 individuals with confirmed CVDs and 11 individuals with confirmed CRDs were documented as still engaged in care. Among individuals diagnosed with TB, approximately 18% also had radiographic features suggestive of cardiomegaly.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab1\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003e\u003cstrong\u003eCharacteristics of participants screened in community outreach activities\u003c/strong\u003e \u003cem\u003e(Median age 41 years (range: 6\u0026ndash;101 years).\u003c/em\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eState\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eLGA (Total screened)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMale n (%)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eFemale n (%)\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eEbonyi\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003eIkwo (2,432)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e914 (38%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e1,518 (62%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003eIshielu (2,616)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e819 (31%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e1,797(69%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003eIzzi (1,727)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e774 (45%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e953 (55%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eEbonyi total\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e6,775 (71%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eNasarawa\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003eKaru (1,585)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e884 (52%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e701(48%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003eLafia (1,223)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e662(51%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e563(49%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eNasarawa total\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e2808 (29%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003ctable id=\"Tab2\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eAbnormality characteristics by LGA and sex\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eState / LGA\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eAbnormality Male\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eAbnormality Female\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eCVD Male\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eCVD Female\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eCRD Male\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eCRD Female\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003ePresumptive TB (\u0026ge;\u0026thinsp;0.5) Male\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003ePresumptive TB (\u0026ge;\u0026thinsp;0.5) Female\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eEbonyi \u0026ndash; Ikwo (808)\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e270(34%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e538(66%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e39(16%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e219((84%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e110(32%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e209(68%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e121 (52%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e110 (48%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eEbonyi \u0026ndash; Ishielu (926)\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e247(27%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e679(73%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e45(14%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e264(86%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e111(28%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e279(72%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e91(40%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e136(60%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eEbonyi \u0026ndash; Izzi (608)\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e248 (40%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e360 (60%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e32(27%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e84(73%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e134(38%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e228(68%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e80(62%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e50(38%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eNasarawa \u0026ndash; Karu (477)\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e197(36%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e280(66%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e28(31%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e61(69%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e95(36%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e160(64%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e74(56%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e59(44%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003eNasarawa \u0026ndash; Lafia (347)\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e133 (33%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e214(67%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e22(25%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e58(75%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e65(35%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e123(65%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e46(58%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e33(42%)\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study demonstrates that community-based AI-enabled chest X-ray screening is a feasible platform for integrated detection of tuberculosis (TB), cardiovascular diseases (CVDs), and chronic respiratory diseases (CRDs) in underserved rural communities in Nigeria. Among the 9,585 individuals screened, 3,166 (33%) chest radiographs were flagged as abnormal by the AI system, highlighting a substantial burden of previously undiagnosed cardiopulmonary conditions that would likely remain undetected within routine primary care systems (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). The findings provide insights into the distribution of abnormalities, sex differences in screening participation and disease patterns, and the emerging burden of non-communicable diseases (NCDs) in rural LMIC settings. However, considerable attrition along referral pathways for non-TB conditions was observed. Although individuals with NCD-related abnormalities were referred for further evaluation, only 12% completed referral that is reaching the treatment facility for further evaluation and care, this reflects structural gaps in service availability. These findings highlight both the public health value of multi-disease community screening and the importance of strengthening referral systems and decentralised services to ensure that early detection translates into effective care (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eHigh prevalence of radiographic abnormalities in rural communities\u003c/p\u003e \u003cp\u003eA key finding of this study is the high proportion of abnormal chest radiographs detected during community screening. Approximately one-third (33%) of all screened individuals had abnormal chest radiographs, indicating substantial unmet diagnostic need in these rural communities. Similar abnormality rates have been reported in community-based chest X-ray screening initiatives in other high TB burden settings and are often linked to delayed health-seeking behaviour, limited access to diagnostic services, and prolonged exposure to environmental and occupational risk factors (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eChest radiography remains one of the most effective population-level tools for detecting pulmonary disease, including TB and other cardiopulmonary conditions (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). The integration of computer-aided detection (CAD) systems has further improved the feasibility of large-scale screening in resource-limited settings by enabling rapid interpretation and reducing reliance on specialist radiologists. Several studies have demonstrated that AI-based CAD systems can achieve diagnostic performance comparable to trained human readers for TB screening and are increasingly recommended for use in high-burden settings(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). In communities with limited access to routine diagnostic services, AI-supported chest radiography therefore offers an effective triage tool for identifying both TB and other pulmonary abnormalities. The high abnormality yield observed in this study supports the rationale for integrated screening models that address multiple cardiopulmonary conditions rather than focusing solely on TB (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGender patterns in screening participation and disease detection\u003c/p\u003e \u003cp\u003eClear gender patterns emerged across both screening participation and disease detection. Overall, females constituted a higher proportion of individuals screened, although regional differences were observed. In Ebonyi State, females represented the majority of participants screened, whereas in Nasarawa State males accounted for a larger proportion of screened individuals. These patterns likely reflect regional differences in gender norms, health-seeking behaviour, and participation in community mobilisation activities. Evidence from southern Nigeria suggests that women often engage more actively in community-based preventive health services, while studies from northern and north-central Nigeria report lower female participation due to mobility constraints and household decision-making dynamics affecting healthcare access (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite higher female participation in screening, men accounted for a greater proportion of presumptive TB cases and bacteriologically confirmed TB diagnoses (60%), consistent with global epidemiological evidence showing higher TB incidence among men(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Occupational exposures, tobacco use, alcohol consumption, and delayed care-seeking behaviour among men are recognised contributors to this disparity.(\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eIn contrast, women represented the majority of individuals with abnormalities suggestive of CVDs and CRDs. Approximately 25% of individuals screened had abnormalities suggestive of cardiovascular or chronic respiratory disease, with women accounting for about 75% of these cases. This pattern may reflect gendered differences in environmental exposures and access to preventive care. For example, prolonged exposure to household air pollution from biomass fuel use has been linked to chronic respiratory disease among women in rural settings (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). Similarly, women in many LMICs experience lower rates of routine cardiovascular screening and treatment, contributing to delayed detection of hypertension and related conditions (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). These findings highlight the importance of incorporating gender-sensitive strategies into integrated screening programmes to address both biological and social determinants of health.\u003c/p\u003e \u003cp\u003eEmerging burden of NCDs in rural settings\u003c/p\u003e \u003cp\u003eAnother important finding of this study is the substantial burden of NCD-related abnormalities detected during screening. Among the 3,166 abnormal radiographs, approximately 2,367 (75%) showed features suggestive of CVDs or CRDs, indicating that the majority of abnormalities identified were related to non-TB conditions. These findings challenge the perception that NCDs are primarily urban health problems and instead highlight the ongoing epidemiological transition affecting rural populations in LMICs (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eEvidence from Nigeria and other sub-Saharan African countries has documented increasing prevalence of hypertension, cardiovascular risk factors, and chronic respiratory disease in rural communities (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Environmental exposures such as biomass fuel use, alongside changing lifestyles and ageing populations, are contributing to this growing burden (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). Integrated screening initiatives such as INTEGRATE TB therefore provide valuable community-level data that can inform health system planning and support the decentralisation of NCD services within primary healthcare platforms. Without such integrated approaches, many of these conditions would likely remain undiagnosed until advanced stages of disease(\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHealth system constraints and linkage to care\u003c/p\u003e \u003cp\u003eThe contrasting linkage outcomes for TB and NCD-related abnormalities highlight important differences in health system capacity. The 95% treatment initiation rate among bacteriologically confirmed TB cases reflects the relative maturity of Nigeria\u0026rsquo;s TB programme. TB services are decentralised, free at the point of care, and embedded within primary healthcare facilities through established diagnostic and treatment pathways (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Community-based active case finding supported by AI-enabled chest X-ray screening, therefore enabled timely diagnosis and treatment initiation, consistent with findings from other TB active case-finding initiatives (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn contrast, linkage to care for individuals with abnormalities suggestive of CVD or CRD was substantially lower. Although referrals were issued, only 12% completed the referral, defined as attendance at a treatment facility for further evaluation and care. This gap reflects systemic constraints rather than limitations of the screening approach itself. In many rural Nigerian settings, CVD and CRD services are not available at the primary healthcare level and require referral to higher-level facilities. This referral-dependent model introduces barriers related to travel distance, transport costs, and out-of-pocket expenditure, contributing to attrition along the referral cascade (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese findings highlight a key challenge for integrated screening programmes: detection alone does not guarantee access to treatment when health systems lack sufficient capacity to manage identified conditions. Strengthening primary healthcare capacity will therefore be essential for improving the impact of integrated screening interventions. The WHO Package of Essential Noncommunicable Disease Interventions (WHO PEN) recommends decentralising basic NCD services to primary care facilities, including hypertension screening, diabetes testing, and access to essential medicines (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). Integrating such services within existing TB programme platforms could help reduce fragmentation and improve continuity of care.\u003c/p\u003e \u003cp\u003eOperational lessons from community-based screening\u003c/p\u003e \u003cp\u003eA small proportion of pre-treatment loss to follow-up occurred among individuals diagnosed with TB. Of the 204 bacteriologically confirmed TB cases, approximately 5% were not initiated on treatment, reflecting the challenges of tracking individuals identified through community outreach activities, particularly in large congregate settings such as markets. Similar patterns have been observed in other community-based TB case-finding programmes (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eTo mitigate this risk, the intervention engaged trusted community volunteers to support participant tracking, communicate results, and facilitate referral and treatment initiation. Community-led approaches have been shown to improve linkage to care by leveraging local trust and social networks and are recommended in WHO guidance on community-based TB care (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eImplications for integrated service delivery\u003c/p\u003e \u003cp\u003eThe findings suggest that integrated screening programmes will be most effective when aligned with accessible treatment services. TB screening platforms can serve as entry points for broader cardiopulmonary health assessment, but meaningful improvements in NCD outcomes will require strengthening primary healthcare capacity to deliver basic services such as blood pressure measurement, chronic respiratory assessment, and initial treatment (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCombining AI-enabled chest X-ray screening with complementary low-cost diagnostic tools could improve early identification of cardiovascular risk while reducing unnecessary referrals. Sustained community engagement, structured referral systems, and the involvement of trusted community volunteers should remain central components of integrated outreach models.\u003c/p\u003e \u003cp\u003eStrengths and limitations\u003c/p\u003e \u003cp\u003eA major strength of this study is its implementation under routine programme conditions in rural communities, generating operational evidence on the feasibility of AI-enabled integrated screening. Standardised screening protocols, digital data systems, and ongoing supervision supported implementation fidelity and data quality.\u003c/p\u003e \u003cp\u003eSeveral limitations should also be considered. Imaging artefacts caused by patterned clothing produced false-positive readings during early implementation, although mitigation measures were subsequently introduced. In addition, referral completion for non-TB conditions may be underestimated because the study tracked only individuals attending designated referral facilities and some participants may have sought care elsewhere.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eArtificial Intelligence\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCVD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCardiovascular Disease\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCRD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eChronic Respiratory Disease\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDOTS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDirectly Observed Treatment, Short-course\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLGA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLocal Government Area\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePDX\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ePortable Digital X-ray\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTB\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eTuberculosis\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting interests:\u003c/h2\u003e \u003cp\u003eJ.C. and T.R. are affiliated with the Stop TB Partnership, which supports the TB REACH initiative that funded this work. They were not involved in funding decisions but provided technical support. All other authors declare that they have no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding:\u003c/h2\u003e \u003cp\u003e The intervention was implemented by the Catholic Caritas Foundation of Nigeria (CCFN) in partnership with the Zankli Research Centre, Bingham University, Nigeria, under the TB REACH Wave 10 initiative funded by the Stop TB Partnership, in collaboration with State and Local Government Tuberculosis Programmes and tertiary referral hospitals.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eC.O., J.I., and C.U. conceptualised the study, and C.O. led implementation. C.O., J.I., and C.U. designed the study methodology. C.O., C.U., and R.F. supervised data collection and analysis. C.O. and C.U. drafted the manuscript. J.C. and other co-authors provided critical revisions and technical input. All authors reviewed and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors acknowledge the contributions of community volunteers, field teams, and State Tuberculosis and Leprosy Control Programme staff in Ebonyi and Nasarawa States for their support in implementing the outreach activities. We particularly recognise the State Tuberculosis and Leprosy Control Programme Managers in both states, the National Tuberculosis Programme, and Qure.ai for their technical support. We also thank the participating communities for their cooperation and engagement.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003e The datasets generated and/or analysed are available from the corresponding author on request and with permission from the Catholic Caritas Foundation of Nigeria and the relevant State TB and leprosy control program.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eWorld Health Organization. Noncommunicable diseases [Internet]. 2023. Available from: https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases\u003c/li\u003e\n\u003cli\u003eOmran AR. The epidemiologic transition: a theory of the epidemiology of population change. Milbank Mem Fund Q. 1971;49(4):509\u0026ndash;38.\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. The double burden of malnutrition [Internet]. World Health Organization; 2020. Report No. Available from: https://www.who.int/publications/i/item/WHO-NMH-NHD-20.6\u003c/li\u003e\n\u003cli\u003ePopkin BM, Corvalan C, Grummer-Strawn LM. Dynamics of the double burden of malnutrition. The Lancet. 2020;395:65\u0026ndash;74. doi:10.1016/S0140-6736(19)32497-3\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. Global tuberculosis report 2024 [Internet]. World Health Organization; 2024. Report No. Available from: https://www.who.int/publications/i/item/9789240101531\u003c/li\u003e\n\u003cli\u003eGBD 2019 Diseases and Injuries Collaborators. Global burden of diseases results tool: Nigeria. The Lancet [Internet]. 2020. Available from: https://vizhub.healthdata.org/gbd-results/\u003c/li\u003e\n\u003cli\u003eAdeloye D, Basquill C. Estimated prevalence and control of hypertension in Nigeria. J Hypertens. 2014;32(12):2307\u0026ndash;15. doi:10.1097/HJH.0000000000000380\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. STEPwise approach to NCD risk factor surveillance: Nigeria factsheet [Internet]. World Health Organization; 2018. Report No. Available from: https://www.who.int/teams/noncommunicable-diseases/surveillance/systems-tools/steps\u003c/li\u003e\n\u003cli\u003eMbakwem AC, Ajuluchukwu JN, Amadi C. Cardiovascular disease admissions and mortality in Lagos, Nigeria. Cardiovasc J Afr. 2016;27(2):84\u0026ndash;9. doi:10.5830/CVJA-2015-088\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. Global tuberculosis report 2023: Nigeria country profile [Internet]. World Health Organization; 2023. Report No. Available from: https://www.who.int/teams/global-tuberculosis-programme/data\u003c/li\u003e\n\u003cli\u003eFederal Ministry of Health Nigeria. National Multi-Sectoral Action Plan for the Prevention and Control of Non-Communicable Diseases (2019\u0026ndash;2025) [Internet]. Federal Ministry of Health Nigeria; 2019. Report No. Available from: https://www.health.gov.ng/doc/NCDs_ACTION_PLAN.pdf\u003c/li\u003e\n\u003cli\u003eAkinwumi AF, Esimai OA, Arije O, Ojo TO, Esan OT. Preparedness of primary health care facilities on implementation of essential non-communicable disease interventions in osun state south-west Nigeria: a rural-urban comparative study. BMC Health Serv Res. 2023 Feb 14;23(1):154. doi:10.1186/s12913-023-09138-8 PubMed PMID: 36788557; PubMed Central PMCID: PMC9930353.\u003c/li\u003e\n\u003cli\u003ehttps://www.who.int/publications/i/item/WHO-HTM-TB-2016.20.\u003c/li\u003e\n\u003cli\u003evan\u0026rsquo;t Hoog AH. Systematic screening for tuberculosis using chest radiography. Int J Tuberc Lung Dis. 2014;18(11):1228\u0026ndash;36. doi:10.5588/ijtld.14.0315\u003c/li\u003e\n\u003cli\u003eQin ZZ. Using artificial intelligence to read chest radiographs for tuberculosis detection. Eur Respir J. 2019;54. doi:10.1183/13993003.01023-2019\u003c/li\u003e\n\u003cli\u003ePhilipsen RH. Diagnostic accuracy of computer-aided detection for tuberculosis on chest radiography. Clin Infect Dis. 2022;74(11):2023\u0026ndash;32. doi:10.1093/cid/ciab639\u003c/li\u003e\n\u003cli\u003ePande T. Computer-aided detection for tuberculosis screening. PLOS ONE. 2020;15(2):e0227479. doi:10.1371/journal.pone.0227479\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. WHO consolidated guidelines on tuberculosis: module 2 \u0026ndash; screening [Internet]. World Health Organization; 2021. Report No. Available from: https://www.who.int/publications/i/item/9789240022676\u003c/li\u003e\n\u003cli\u003eMcAdams HP. Radiographic manifestations of chronic lung and cardiac disease. Radiographics. 2015;35(5):1289\u0026ndash;303. doi:10.1148/rg.2015140310\u003c/li\u003e\n\u003cli\u003eHoffmann TC, Glasziou PP, Boutron I. Better reporting of interventions: the TIDieR checklist and guide. BMJ. 2014;348:g1687.\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. Systematic screening for active tuberculosis: principles and recommendations [Internet]. Geneva: World Health Organization; 2013. Available from: https://www.who.int/publications/i/item/9789241548601\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. WHO consolidated guidelines on tuberculosis: Module 3 \u0026ndash; Diagnosis [Internet]. Geneva: World Health Organization; 2021. Available from: https://www.who.int/publications/i/item/9789240029415\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. The use of digital chest radiography for tuberculosis detection and diagnosis [Internet]. Geneva: World Health Organization; 2016. Available from: https://www.who.int/publications/i/item/WHO-HTM-TB-2016.20\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. WHO operational handbook on tuberculosis: Module 2 \u0026ndash; Screening [Internet]. Geneva: World Health Organization; 2022. Available from: https://www.who.int/publications/i/item/9789240057562\u003c/li\u003e\n\u003cli\u003eStop TB Partnership. TB REACH: Finding the missing people with TB [Internet]. Geneva: Stop TB Partnership; 2023. Available from: https://stoptb.org/what-we-do/tb-reach\u003c/li\u003e\n\u003cli\u003eMcAdams HP, Erasmus JJ, Rosado-de-Christenson ML. Radiographic manifestations of chronic lung and cardiac disease. Radiographics. 2015;35(5):1289\u0026ndash;303.\u003c/li\u003e\n\u003cli\u003eWorld Medical Association. Declaration of Helsinki: Ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191\u0026ndash;4.\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. Ethical standards for tuberculosis care and prevention [Internet]. Geneva: World Health Organization; 2017. Available from: https://www.who.int/publications/i/item/WHO-HTM-TB-2017.28\u003c/li\u003e\n\u003cli\u003eFoo CD, Shrestha P, Wang L, Du Q, Garc\u0026iacute;a-Basteiro AL, Abdullah AS, et al. Integrating tuberculosis and noncommunicable diseases care in low- and middle-income countries (LMICs): A systematic review. PLoS Med. 2022 Jan;19(1):e1003899. doi:10.1371/journal.pmed.1003899 PubMed PMID: 35041654; PubMed Central PMCID: PMC8806070.\u003c/li\u003e\n\u003cli\u003eCalderwood CJ, Timire C, Mavodza C, Kavenga F, Ngwenya M, Madziva K, et al. Beyond tuberculosis: a person-centred and rights-based approach to screening for household contacts. Lancet Glob Health. 2024 Mar 1;12(3):e509\u0026ndash;15. doi:10.1016/S2214-109X(23)00544-2 PubMed PMID: 38365421.\u003c/li\u003e\n\u003cli\u003eNugent R, Bertram MY, Jan S, Niessen LW, Sassi F, Jamison DT, et al. Investing in non-communicable disease prevention and management to advance the Sustainable Development Goals. The Lancet. 2018 May 19;391(10134):2029\u0026ndash;35. doi:10.1016/S0140-6736(18)30667-6 PubMed PMID: 29627167.\u003c/li\u003e\n\u003cli\u003eAtun R, Jaffar S, Nishtar S, Knaul FM, Barreto ML, Nyirenda M, et al. Improving responsiveness of health systems to non-communicable diseases. The Lancet. 2013 Feb 23;381(9867):690\u0026ndash;7. doi:10.1016/S0140-6736(13)60063-X\u003c/li\u003e\n\u003cli\u003eOni T, Burke R, Tsekela R, Bangani N, Seldon R, Gideon HP, et al. High prevalence of subclinical tuberculosis in HIV-1-infected persons without advanced immunodeficiency: implications for TB screening. Thorax. 2011 Aug;66(8):669\u0026ndash;73. doi:10.1136/thx.2011.160168 PubMed PMID: 21632522; PubMed Central PMCID: PMC3142344.\u003c/li\u003e\n\u003cli\u003eHamada Y, Mukora R, Pelusa R, Ntshiqa T, Shedrawy J, Velen K, et al. Costs and cost-effectiveness of integrated screening for non-communicable diseases in TB contacts. IJTLD Open. 2025 Mar;2(3):160\u0026ndash;5. doi:10.5588/ijtldopen.24.0625 PubMed PMID: 40092516; PubMed Central PMCID: PMC11906026.\u003c/li\u003e\n\u003cli\u003eQin ZZ, Ahmed S, Sarker MS. Using artificial intelligence to read chest radiographs for tuberculosis detection: a multi-site evaluation of diagnostic accuracy. Lancet Digit Health. 2021;3(9):e595\u0026ndash;604.\u003c/li\u003e\n\u003cli\u003eScott AJ, Perumal T, Pooran A, Oelofse S, Jaumdally S, Swanepoel J, et al. Clinical evaluation of computer-aided digital x-ray detection of pulmonary tuberculosis during community-based screening or active case-finding: a case\u0026ndash;control study. Lancet Glob Health. 2025 Mar 1;13(3):e517\u0026ndash;27. doi:10.1016/S2214-109X(24)00516-3 PubMed PMID: 40021309.\u003c/li\u003e\n\u003cli\u003eBeaglehole R, Bonita R, Horton R, Adams C, Alleyne G, Asaria P, et al. Priority actions for the non-communicable disease crisis. The Lancet. 2011 Apr 23;377(9775):1438\u0026ndash;47. doi:10.1016/S0140-6736(11)60393-0 PubMed PMID: 21474174.\u003c/li\u003e\n\u003cli\u003eAdeloye D, Basquill C, Aderemi AV, Thompson JY, Obi FA. An estimate of the prevalence of hypertension in Nigeria: a systematic review and meta-analysis. J Hypertens. 2015 Feb;33(2):230\u0026ndash;42. doi:10.1097/HJH.0000000000000413 PubMed PMID: 25380154.\u003c/li\u003e\n\u003cli\u003eKc H, P M, Rm H, Rg W, El C. Sex Differences in Tuberculosis Burden and Notifications in Low- and Middle-Income Countries: A Systematic Review and Meta-analysis. PLoS Med. 2016 Sep 6;13(9). doi:10.1371/journal.pmed.1002119 PubMed PMID: 27598345.\u003c/li\u003e\n\u003cli\u003eBigna JJ, Noubiap JJ. The rising burden of non-communicable diseases in sub-Saharan Africa. Lancet Glob Health. 2019 Oct 1;7(10):e1295\u0026ndash;6. doi:10.1016/S2214-109X(19)30370-5 PubMed PMID: 31537347.\u003c/li\u003e\n\u003cli\u003eWHO package of essential noncommunicable (PEN) disease interventions for primary health care [Internet]. [cited 2025 Jun 7]. Available from: https://www.who.int/publications/i/item/9789240009226\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-global-and-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BMC Global and Public Health](https://bmcglobalpublichealth.biomedcentral.com/)","snPcode":"44263","submissionUrl":"https://submission.springernature.com/new-submission/44263/3","title":"BMC Global and Public Health","twitterHandle":"@BMC_GPH","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9180615/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9180615/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLow- and middle-income countries face a growing dual burden of communicable and non-communicable diseases, while health services remain largely organised around vertical programmes. In Nigeria, tuberculosis (TB) services are relatively established at the primary healthcare level, whereas access to cardiovascular disease (CVD) and chronic respiratory disease (CRD) care remains limited in rural settings. Artificial intelligence (AI)–enabled chest X-ray offers an opportunity to integrate TB screening with the identification of other cardiopulmonary abnormalities at the community level. This study describes the implementation and outcomes of a community-based, AI-enabled integrated screening intervention in rural Nigeria.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a non-randomised implementation study across five Local Government Areas in Ebonyi and Nasarawa States between January 2023 and December 2024. Community outreach activities used portable digital chest X-ray integrated with AI software to screen individuals aged six years and above. Presumptive TB cases underwent GeneXpert testing, while non-TB radiographic abnormalities were referred for further evaluation. Descriptive analyses summarised screening yield, diagnostic outcomes, and linkage to care.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA total of 9,585 individuals were screened through 93 community-based outreach activities. Overall, 3,166 (33%) chest radiographs were flagged as abnormal by AI. In total, 1,336 individuals were classified as presumptive TB cases, of whom 1,123 (84%) produced sputum samples for GeneXpert testing. 204 individuals were bacteriologically confirmed with TB, and 194 (95%) were initiated on treatment. An additional 199 individuals were clinically diagnosed with TB following radiologist review.\u003c/p\u003e\n\u003cp\u003eAmong abnormal radiographs, 2,367 (75%) showed features suggestive of CVDs or CRDs. All clients with such conditions were referred to tertiary facilities; however, only 12% completed the referral. Implementation adaptations, including improved imaging protocols and community-based follow-up strategies, supported TB linkage but had a limited impact on non-TB referral completion.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e\u003cbr\u003e\nAI-enabled community chest X-ray screening is feasible for TB case finding in rural Nigeria and achieves high linkage to TB treatment. However, limited decentralisation of non-communicable disease services constrains care continuity for CVDs and CRDs. Integrated screening programmes should be paired with strengthened primary healthcare capacity, complementary tools such as blood pressure measurement, and context-specific community engagement strategies.\u003c/p\u003e","manuscriptTitle":"Implementing AI-enabled chest X-ray for community-based integrated screening for tuberculosis, chronic respiratory diseases, and cardiovascular diseases in Nigeria","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-07 10:09:45","doi":"10.21203/rs.3.rs-9180615/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-28T05:34:51+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-24T13:37:51+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-14T08:28:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"288168377984434664596314897123003045293","date":"2026-04-12T17:30:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"289302558016702648500195675366039616908","date":"2026-04-12T10:22:07+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-10T03:17:24+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-06T18:46:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"155571319364878460337834608618059891077","date":"2026-04-06T18:28:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"280263256576173029978056892604147775800","date":"2026-04-01T07:57:09+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-01T07:16:06+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-23T05:43:48+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-23T05:08:43+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Global and Public Health","date":"2026-03-20T15:57:13+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-global-and-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BMC Global and Public Health](https://bmcglobalpublichealth.biomedcentral.com/)","snPcode":"44263","submissionUrl":"https://submission.springernature.com/new-submission/44263/3","title":"BMC Global and Public Health","twitterHandle":"@BMC_GPH","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"714e6fc5-966f-4a11-8264-6b72948551b5","owner":[],"postedDate":"April 7th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-04-28T05:40:12+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-07 10:09:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9180615","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9180615","identity":"rs-9180615","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}