Resource-limited Settings Research Articles

Transforming Healthcare in Low-Resource Settings With Artificial Intelligence: Recent Developments and Outcomes.

Artificial intelligence now encompasses technologies like machine learning, natural language processing, and robotics, allowing machines to undertake complex tasks traditionally done by humans. AI's application in healthcare has led to advancements in diagnostic tools, predictive analytics, and surgical precision. This comprehensive review aims to explore the transformative impact of AI across diverse healthcare domains, highlighting its applications, advancements, challenges, and contributions to enhancing patient care. A comprehensive literature search was conducted across multiple databases, covering publications from 2014 to 2024. Keywords related to AI applications in healthcare were used to gather data, focusing on studies exploring AI's role in medical specialties. AI has demonstrated substantial benefits across various fields of medicine. In cardiology, it aids in automated image interpretation, risk prediction, and the management of cardiovascular diseases. In oncology, AI enhances cancer detection, treatment planning, and personalized drug selection. Radiology benefits from improved image analysis and diagnostic accuracy, while critical care sees advancements in patient triage and resource optimization. AI's integration into pediatrics, surgery, public health, neurology, pathology, and mental health has similarly shown significant improvements in diagnostic precision, personalized treatment, and overall patient care. The implementation of AI in low-resource settings has been particularly impactful, enhancing access to advanced diagnostic tools and treatments. AI is rapidly changing the healthcare industry by greatly increasing the accuracy of diagnoses, streamlining treatment plans, and improving patient outcomes across a variety of medical specializations. This review underscores AI's transformative potential, from early disease detection to personalized treatment plans, and its ability to augment healthcare delivery, particularly in resource-limited settings.

Knowledge and Attitude of Women Towards Breast Cancer Screening in Resource-Limited Setting in Women Aged 18–45 Years

Knowledge and Attitude of Women Towards Breast Cancer Screening in Resource-Limited Setting in Women Aged 18–45 Years

RT-RPA as a dual tool for detection and phylogenetic analysis of epidemic arthritogenic alphaviruses

Chikungunya (CHIKV), o’nyong-nyong (ONNV), and Mayaro (MAYV) viruses are transmitted by mosquitoes and known to cause a debilitating arthritogenic syndrome. These alphaviruses have emerged and re-emerged, leading to outbreaks in tropical and subtropical regions of Asia, South America, and Africa. Despite their prevalence, there persists a critical gap in the availability of sensitive and virus-specific point-of-care (POC) diagnostics. Traditional immunoglobulin-based tests such as enzyme-linked immunosorbent assay (ELISA) often yield cross-reactive results due to the close genetic relationship between these viruses. Molecular diagnostics such as quantitative polymerase chain reaction (qPCR) offer high sensitivity but are limited by the need for specialized laboratory equipment. Recombinase polymerase amplification (RPA), an isothermal amplification method, is a promising alternative to qPCR, providing rapid results with minimal equipment requirements. Here, we report the development and validation of three virus-specific RT-RPA-based rapid tests for CHIKV, ONNV, and MAYV. These tests demonstrated both speed and sensitivity, capable of detecting 10–100 viral copies within 20 min of amplification, without exhibiting cross-reactivity. Furthermore, we evaluated the clinical potential of these tests using serum and tissue samples from CHIKV, ONNV, and MAYV-infected mice, as well as CHIKV-infected human patients. We demonstrate that the RPA amplicons derived from the patient samples can be sequenced, enabling cost-effective molecular epidemiological studies. Our findings highlight the significance of these rapid and specific diagnostics in improving the early detection and management of these arboviral infections, particularly in resource-limited settings.

Diagnosis and Management of an Acquired Factor XIII Deficiency in a Limited Resource Setting: A Case Report

Diagnosis and Management of an Acquired Factor XIII Deficiency in a Limited Resource Setting: A Case Report

The Burden of Gastric Cancer in Northern Central America.

Gastric cancer is the fourth leading cause of cancer-related death worldwide, with nearly one million new cases diagnosed in 2020. There is marked variation in gastric cancer incidence globally, with highest incidence rates reported in the United Nations regions of Eastern Asia, Eastern Europe, and Latin America. Although the United States is considered a low-incidence country, gastric cancer presents an important cancer disparity, with higher incidence in minoritized populations, including immigrants from high-incidence regions. The Northern Central America nations are low- and middle-income countries with a high gastric cancer incidence and large US immigrant populations. These countries lack comprehensive cancer registries, but recent GLOBOCAN-imputed gastric cancer estimates are in the range of 8 to 12/100,000 age-standardized incidence rates. Three epidemiologic studies carried out in El Salvador, Nicaragua, and Honduras demonstrate a higher than predicted burden of gastric cancer with low 5-year survival (<10%). The gastric cancer burden is projected to increase in the absence of changes to national and regional cancer control plans. Twelve evidence-based recommendations to reduce gastric cancer mortality in the Americas have recently been proposed, ranging from the "test-and-treat" Helicobacter pylori eradication strategy to endoscopic screening and surveillance programs. Translating these recommendations into a practical plan for this resource-limited setting could address the disproportionate gastric cancer burden. See related article by Peña-Galo et al., p. 1564 See related article by Ruiz de Campos et al., p. 1571 See related article by Dominguez et al., p. 1578.

State of the art of real-life concentration monitoring of rifampicin and its implementation contextualized in resource-limited settings: the Tanzanian case.

The unique medical and socio-economic situation in each country affected by TB creates different epidemiological contexts, thus providing exploitable loopholes for the spread of the disease. Country-specific factors such as comorbidities, health insurance, social stigma or the rigidity of the health system complicate the management of TB and the overall outcome of each patient. First-line TB drugs are administered in a standardized manner, regardless of patient characteristics other than weight. This approach does not consider patient-specific conditions such as HIV infection, diabetes mellitus and malnutrition, which can affect the pharmacokinetics of TB drugs, their overall exposure and response to treatment. Therefore, the 'one-size-fits-all' approach is suboptimal for dealing with the underlying inter-subject variability in the pharmacokinetics of anti-TB drugs, further complicated by the recent increased dosing regimen of rifampicin strategies, calling for a patient-specific methodology. In this context, therapeutic drug monitoring (TDM), which allows personalized drug dosing based on blood drug concentrations, may be a legitimate solution to address treatment failure. This review focuses on rifampicin, a critical anti-TB drug, and examines its suitability for TDM and the socio-economic factors that may influence the implementation of TDM in clinical practice in resource-limited settings, illustrated by Tanzania, thereby contributing to the advancement of personalized TB treatment.

TB programme outcomes in South Fly District, Papua New Guinea, were maintained through COVID-19.

An established response to an outbreak of drug-resistant TB (DR-TB) on Daru Island, South Fly District (SFD), Western Province, Papua New Guinea (PNG). To describe and evaluate the trends in TB case notification disaggregated by demographic and clinical characteristics, programmatic interventions for TB and COVID-19 and treatment outcomes in 2017-2022. A cohort study of routinely collected programmatic data of all patients registered for TB treatment in SFD comparing pre-COVID (2017-2019) to COVID (2020-2022) periods. Of the 3,751 TB cases registered, 19.6% had DR-TB, and the case notification rate was 1,792/100,000 for Daru and 623/100,000 for SFD. There was a 29.2% reduction in case notifications from 2019 to 2021, with recovery in 2022. During COVID, the healthcare workforce was adversely impacted, and active TB case-finding was stopped. During COVID, compared to pre-COVID, bacteriological confirmation increased (62.3% to 71.9%), whereas rates of child TB notifications (11.6% to 9.1%), pulmonary TB (60.8% to 57.4%) and DR-TB (20.7% to 18.6%) decreased. High rates of treatment success were maintained for both drug-susceptible (86.5%) and DR-TB (83.6%). Health systems strengthening and community engagement before COVID likely contributed to resilience and mitigated potential impacts on TB in this remote and resource-limited setting. Case notifications remain very high, and additional interventions are needed to interrupt transmission.

La réinstillation distale du chyme des stomies : une technique efficace pour gérer le débit stomial postopératoire

Background: High output enterostomy leads to malnutrition and fluid/electrolyte loss which may be challenging to manage despite dietary modification, anti-motility, anti-secretory drugs, and parenteral nutrition. Distal infusion stomal enteroclysis (DISE) is an alternative to restore nutritional deficit and replace parenteral nutrition in resource-limited settings where treatment cost and availability of trained nurses are limiting factors.Objective: To assess the effectiveness and feasibility of DISE in managing postoperative enterostomal output.Methods: Consecutive patients who met the inclusion criteria and underwent enterostomy in one year were included. Postoperatively, DISE was started after the return of bowel movement. Stomal effluent was collected, filtered, and reinfused through the distal limb using a Foley catheter. The patients and caregivers were trained to perform under supervision. At discharge, they were encouraged to maintain regular contact and advised to report back when necessary.Results: Twenty-five patients received DISE, of which 22 were discharged and successfully continued to manage at home, while 3 expired. The median age was 36 years, the median BMI at admission and discharge was 18.99 kg/m2 and 17.76 kg/m2 respectively (the difference was statistically significant), and the male: female ratio was 2.5: 1. Thirty-day readmission was done for 2 patients to manage minor complications. The average stomal output was 820.20 ml/day initially and 477.55 ml/day at discharge. The median duration of DISE was 12 days. Thirteen patients (52%) needed parenteral nutritional support due to inadequate oral intake.Conclusion: DISE is a feasible and effective technique for managing high output enterostomies. It can be easily taught and performed with minimal resources at home.

The Incidence of Infection Following Open Tibia Fractures in Low-Income Countries: A Systematic Review

Introduction: Open tibial fractures (OTFs) are common in low-income countries (LICs), primarily caused by road traffic collisions. High rates of infection in OTFs contribute to prolonged recovery, disability, and economic loss. Identifying infection rates and causative organisms is essential for improving outcomes and guiding antibiotic treatment in LICs. This systematic review aimed to determine the incidence of surgical site infections (SSIs) following OTF treatment in LICs and identify the most prevalent microorganisms involved. Methods: Following PRISMA and Cochrane guidelines, a systematic review was conducted using five databases. Eligible studies included randomised controlled trials, cohort studies, and case series investigating OTFs in LICs. Studies with fewer than ten participants, military medical resources, or those from high-income countries were excluded. Data were extracted for infection rates, causative organisms, and treatment modalities, with a meta-analysis performed using a random effects model to calculate incidence in person-years. Results: Five studies, including two retrospective cohorts, two prospective cohorts, and one RCT, were analysed. The overall infection incidence was 23.21 per 100 person-years (95% CI: 9.19; 58.62). Subgroup analysis showed a slightly lower incidence of 21.88 per 100 person-years for surgical interventions. A high degree of heterogeneity (I² = 91%) was noted. Common causative organisms were not consistently reported. Conclusion: The incidence of SSIs following OTFs in LICs is high, particularly in severe fractures. Establishing a baseline infection rate is critical for future studies to evaluate treatment efficacy and improve outcomes in resource-limited settings. Further research should focus on standardising definitions and reporting of infections to reduce heterogeneity.

Evaluation of a commercial Immunochromatographic Strip Assay (ICT) for rapid detection of bovine, porcine and human Rotavirus A

The most frequent diarrhoeal pathogen “Rotavirus” is commonly detected by Ribo Nucleic Acid- Polyacrylamide Gel Electrophoresis (RNA-PAGE), Reverse Transcription Polymerase Chain Reaction (RT-PCR) and Immunoassay (Enzyme Linked Immunosorbent Assay) in humans. However, use of these tests is limited in animals therefore, penside test like Immunochromatographic test/Lateral Flow Assay (ICT/LFA) is necessary. The present study was carried out to compare the diagnostic efficacy of rapid commercial ICT strip assay, RNA-PAGE and RT-PCR for specific detection of Rotavirus A (RVA) from stool samples of calves, piglets and children. A total of 313 faecal samples were collected between November 2022 to April 2023 from children below 5years of age (n=100), calves (n=100) and piglets (n=113) ≤ 3 months of age and were screened by RNA-PAGE, RT-PCR and commercial ICT kit for rotavirus A.The overall positivity of RVA from human, calves and piglets using RNA-PAGE, RT-PCR and LFA was found to be 35% (35/100), 8% (8/100) and 14.16% (16/113), respectively. Higher positivity of rotavirus was recorded in male children (44.26%, 27/61) than in female children (20.51%, 8/39). The kappa agreement between LFA and RT-PCR was 0.79 (substantial agreement); between LFA and RNA-PAGE was 0.61 (substantially low agreement) and between RNA-PAGE and RT-PCR was found to be 0.75 (substantial agreement). Relative diagnostic sensitivity and specificity of LFA in comparison with RT-PCR was 74.54% and 98.44%, respectively and in comparison with RNA-PAGE was 72.97% and 93.47%, respectively. RNA-PAGE in comparison with RT-PCR is having diagnostic sensitivity and specificity of 65.45% and 99.6%, respectively. In conclusion, the ICT strip assay is a rapid, convenient and effective method with satisfactory efficacy for detection of RVA from children, calves and piglets. ICT fulfilled the WHO ASSURED (Affordable, Sensitive, Specific, User-friendly, Rapid and robust, Equipment-free and Delivered) criteria for point-of-care testing. It can be useful in determining rotavirus outbreaks in resource-limited settings.

Real-time monitoring of cattle (Bos taurus) tissue using a novel point-of-care (POC) polymerase spiral reaction (PSR) colorimetric assay

Adulteration of cattle meat with other species raises critical public health concerns, as existing detection methods lack sensitivity, specificity and practicality. We address this gap by introducing a novel Point-of-Care (POC) polymerase spiral reaction (PSR) assay. This rapid, on-site tool enables highly specific and sensitive identification of cattle tissue, even in resource-limited settings. This innovation paves the way for improved food safety and consumer protection. The cattle-specific PSR assay employs COX3 gene-targeting primers and a chromogenic indicator triad (HNB, Malachite Green and Phenol Red) for rapid visual detection. Cross-amplification absence with other meats (buffalo, sheep, goat and pig) and concordance with SYBR Green I and agarose gel electrophoresis confirm specificity. PSR at 62 °C for 60minutes exhibits a 100 fg detection limit and identifies 0.1% meat admixing. Efficacy is demonstrated across a diverse range of cattle meat samples, blind samples and processed cattle muscle samples. Compatibility with both conventional (snap chill) and commercial DNA extraction kits enables on-site analysis within 90minutes, facilitating rapid field sample analysis.

Heavily treatment-experienced patients with HIV: are new mechanisms of action enough?

Antiretroviral (ARV) drug resistance poses a threat to ending the HIV epidemic. As the rates of integrase resistance continue to increase globally, the availability of options for HIV treatment becomes limited. Heavily treatment-experienced (HTE) people with HIV (PWH) are limited to two or fewer available fully active ARV classes and are more likely to have an AIDS-defining event. Appropriate identification and management of HTE PWH is crucial to improving patient outcomes and reducing the future spread of drug-resistant HIV. As treatment options become more limited owing to drug resistance, the availability of more potent drugs with a marked increase in virologic suppression is needed in the current ART era. The purpose of this narrative review is to review the identification of HTE PWH, novel mechanisms of resistance, and management of HTE PWH in resource-rich and resource-limited settings using novel ARVs and combination ART.

Advancing Rheumatic Heart Disease Management in Indonesia: Strengthening Early Detection, Targeted Screening, and Prevention Strategies

Background: Indonesia ranks fourth globally in rheumatic heart disease (RHD) prevalence, with a mortality rate of 4.8 per 100,000—higher than malaria. Despite global progress in RHD management, Indonesia lacks a comprehensive national prevention program. Secondary prevention strategies could significantly improve outcomes in high-risk populations by targeting disease progression. Method: This systematic review and meta-analysis examined the impact of secondary prevention on echocardiographic outcomes in Indonesian RHD patients. The review included randomized controlled trials, observational studies, quasi-experimental designs, and case-control studies, emphasizing the role of secondary prevention in mitigating disease progression. Results: Six studies meeting inclusion criteria underwent full-text analysis, with four eligible for meta-analysis. These studies revealed that secondary prevention improved echocardiographic markers, demonstrating its potential to alter disease trajectories in resource-limited settings. Conclusion: Indonesia faces challenges in RHD management due to geographical and healthcare constraints. Echocardiographic screening, a sensitive and specific diagnostic tool, is vital for early detection and monitoring. With technological advancements, cost-effective screening is increasingly feasible, even in remote areas. Strengthening early detection, targeted screening, and secondary prevention strategies is critical for reducing the RHD burden in Indonesia.

Health workers’ knowledge and practices toward vaccine cold chain management and its associated factors in a resource-limited setting of Sheger, Oromia, Ethiopia: a multicenter cross-sectional study

BackgroundCold chain management is a continuous and cohesive process of preserving vaccines to ensure their availability and potency. However, the quality control of vaccines has been neglected in resource-limited areas such as Ethiopia. Adequate levels of health workers’ knowledge and practices in vaccine cold chain management are important to maintain the potency of vaccines and achieve the intended vaccine end goal.ObjectiveTo assess the level of knowledge and practices of health workers toward vaccine cold chain management and associated factors in resource-limited settings in Ethiopia.MethodsAn institution-based cross-sectional survey was conducted from May to June 2023 in eight sub-cities of Sheger City, Oromia Region, Ethiopia. 255 health workers were selected via simple random sampling from 17 health centers and 17 health posts. Data were collected through a self-administered questionnaire and entered in to EPI DATA version 3.1 and exported to SPSS version 26.0 software for analysis. Binary and multiple logistic regression analyses were carried out to identify factors associated with the knowledge and practices of health workers toward vaccine cold chain management. A variable with a p value < 0.25 was considered a candidate for further regression analysis, whereas a variable with a 95% CI and a p value < 0.05 was considered statistically significant.ResultA total of 255 respondents participated, for a 100% response rate. The overall magnitude of good knowledge and practice of vaccine cold chain management was 58% (95% CI: 52.2%--64.3%) and 52.2% (95% CI: 46.3-58.4%), respectively. Receiving pre-service training (AOR = 8.2; 95% CI: (2.93–22.99), availability of SOP/guidelines (AOR = 6.55; 95% CI: (2.0–21.36) and receiving supportive supervision (AOR = 4.2; 95% CI: (1.67–10.59)) were significantly associated with good knowledge of cold chain management. The availability of SOP/guidelines (AOR = 2.13; 95% CI: (1.22–3.7)) and the knowledge of health workers (AOR = 2.32; 95% CI: (1.11–4.8) were significantly associated with good cold chain management practices.ConclusionThere are gaps in knowledge and practice concerning cold chain management. There is an urgent need to improve the knowledge and practice of health workers in cold chain management in the newly organized but resource-limited administrative structure (Sheger) in Ethiopia.

Factors, barriers and adherence of nurses to patient education in public hospitals of Lanao del Sur, Philippines

Introduction: Patient education is a core responsibility in nursing, vital for improving patient understanding and active participation in their healthcare decisions. There are various factors that impact nurses' ability to provide patient education, but little is known about how these factors influence adherence in public hospitals, particularly in resource-limited settings. This study investigates nurses' adherence to patient education protocols in Lanao del Sur, Philippines, while evaluating facilitating factors and barriers. Methods: This study employed a quantitative, descriptive, cross-sectional correlational research design. Seventy-six nurses from two public district hospitals participated. Data were gathered through surveys on demographics, adherence to patient education, facilitators, and barriers. Correlation analyses, including Pearson’s correlation and Spearman’s rho, assessed relationships between variables, with Cramer’s V used for categorical data. Statistical significance was set at α = 0.05. Results: Most participants were married females aged 31-40, with incomes of 10,000-39,000 PHP, college-educated, and assigned to wards, with 3 months to 3 years of tenure. Nurses acknowledged the importance of patient education and identified 14 barriers. Adherence to patient education showed negative correlations with age, income, educational attainment, and tenure, but positive correlations with gender, marital status, and area of assignment. Facilitating factors positively correlated with adherence, as did barriers. Conclusions: Identifying factors affecting nurses' adherence to patient education is vital for improving these practices. Findings will support the Nursing Patient Education Development Program, enhancing patient education in Lanao del Sur's public hospitals.

Classification of Pneumonia, Tuberculosis and Covid-19 from Chest X-Ray Images Using Convolution Neural Network Model

Accurate and timely diagnosis of respiratory ailments like pneumonia, tuberculosis (TB), and COVID-19 is pivotal for effective patient care and public health interventions. Deep learning algorithms have emerged as potent tools in medical image classification, offering promise for automated diagnosis and screening. This study presents a deep learning-based approach for categorizing chest X-ray images into three classes: pneumonia, tuberculosis, and COVID-19. Utilizing convolutional neural networks (CNNs) as the primary architecture, owing to their ability to automatically extract relevant features from raw image data. The proposed model is trained on a sizable dataset of chest X-ray images annotated with ground truth labels for pneumonia, TB, and COVID-19. Extensive experiments are conducted to evaluate the model&amp;#39;s performance in terms of classification accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC). Furthermore, we compare the performance of our deep learning model with traditional machine learning approaches and assess its generalization ability on an independent test set. Our findings demonstrate that the proposed deep learning model achieves high accuracy in classifying chest X-ray images of pneumonia, TB, and COVID-19, outperforming traditional methods and showing potential for clinical deployment as a screening tool, especially in resource-limited settings.

Remote monitoring of social behavior in children with autism: The role of digital phenotyping in public programs

Autism Spectrum Disorder (ASD) is characterized by social and behavioral challenges, making early identification and continuous monitoring crucial for effective intervention. Traditional methods of monitoring social behavior in children with ASD often face limitations, particularly in resource-limited settings. The advent of digital phenotyping—using data from smartphones, wearables, and other digital tools to track behaviors—presents new opportunities for remote monitoring. This review examines the role of digital phenotyping in ASD care, focusing on its integration into public health programs. It explores key technologies for remote monitoring, the advantages of digital phenotyping in tracking social interactions, and the challenges posed by privacy and data security concerns. Additionally, the review discusses the policy implications and potential for large-scale adoption in public programs. Finally, it highlights emerging trends, such as AI integration, which could enhance the effectiveness of digital phenotyping in diagnosing and treating children with ASD. The review underscores the potential of digital phenotyping to revolutionize ASD monitoring, enabling more personalized, data-driven approaches in public healthcare systems.

Machine learning in personalized cancer treatment: Implications for global public health

Cancer remains a leading cause of mortality worldwide, posing a significant challenge to global public health. Traditional approaches to cancer treatment often adopt a "one-size-fits-all" strategy, which may not be effective for every patient due to the complex and heterogeneous nature of the disease. Personalized cancer treatment aims to tailor therapeutic interventions to the individual characteristics of each patient, enhancing the efficacy and reducing adverse effects. In recent years, machine learning (ML) has emerged as a powerful tool in personalized cancer care, offering the potential to revolutionize diagnosis, prognosis, and treatment by leveraging vast amounts of biomedical data. This review explores the current landscape of machine learning applications in personalized cancer treatment, including its role in imaging, genomics, and drug discovery. We provide an overview of key machine learning techniques, highlight successful case studies from both developed and developing nations, and examine the challenges and limitations associated with the integration of these technologies into clinical practice. Furthermore, we discuss the implications of machine learning-driven personalized cancer care for global public health, emphasizing its potential to address disparities in access to healthcare and improve outcomes in resource-limited settings. Finally, we offer insights into future research directions and policy considerations that could accelerate the adoption of machine learning in global cancer treatment, fostering a more equitable and effective healthcare ecosystem.'

Hepatitis C Virus—Core Antigen: Implications in Diagnostic, Treatment Monitoring and Clinical Outcomes

The hepatitis C virus (HCV) infection, a global health concern, can lead to chronic liver disease. The HCV core antigen (HCVcAg), a viral protein essential for replication, offers a cost-effective alternative to HCV RNA testing, particularly in resource-limited settings. This review explores the significance of HCVcAg, a key protein in the hepatitis C virus, examining its structure, function, and role in the viral life cycle. It also evaluates its clinical use in diagnosis and treatment monitoring, comparing its performance to the standard HCV RNA assay using data from PubMed and Google Scholar. HCVcAg assays show high pooled sensitivity (93.5%) and pooled specificity (99.2%) compared to HCV RNA assays, correlating closely (r = 0.87) with HCV RNA levels. Hence, HCVcAg testing offers a cost-effective way to diagnose active HCV infections and monitor treatment, especially in resource-limited settings, but its sensitivity can vary and standardization is needed. HCVcAg also predicts liver disease progression and assesses liver damage risk, aiding patient management. It helps to identify patients at risk for fibrosis or carcinoma, making it vital in hepatitis C care. HCVcAg testing can expand access to HCV care, simplify management, and contribute to global elimination strategies, especially in low- and middle-income countries.

Comparative effectiveness of chest ultrasound, chest X-ray and computer-aided diagnostic (CAD) for tuberculosis diagnosis in low-resource setting: study protocol for a cross-sectional study from Ethiopia

IntroductionEarly and accurate diagnosis of pulmonary tuberculosis (TB) is crucial for timely treatment and prevention of transmission, but diagnostic challenges persist due to complex symptoms and limitations in diagnostic tools. Chest X-ray (CXR) is the standard imaging modality, but its sensitivity and specificity may vary. Recently, some promising alternatives emerged such as chest ultrasonography (CUS) – particularly valuable in resource-limited settings – and computer-aided diagnosis (CAD) systems – helping clinicians in the reading and interpretation of the CXR. However, direct comparisons of CUS, CXR, and CAD score in TB diagnosis are limited.Methods and analysisThis cross-sectional study will assess the diagnostic effectiveness of CUS in diagnosing TB compared to CXR and CAD score among index cases and household contacts. The study will be conducted at Wolisso St. Luke Hospital (Wolisso, Ethiopia). Index cases will be subjects with diagnosis of pulmonary tuberculosis within 7 days. Household contacts will be identified by administering a screening questionnaire to index cases. They will undergo CXR as for standard of care and consequent CAD analysis and CUS. The anticipated sample size is 136 subjects. The common accuracy metrics (sensitivity, specificity, positive and negative predictive values) will be calculated.Ethics and disseminationThe protocol was approved by the Oromia Health Bureau Research Ethics Committee (BFO/MBTFH/1-16/1908). All information obtained will be confidential. Selected investigators will have access to data, while international partners will sign a dedicated Data Protection Agreement. Eligible subjects will receive a brief information about the study before being asked to participate and they will provide a written informed consent. Results will be conveyed to stakeholders and disseminated through conferences and peer-reviewed journals.Clinical trial registrationNCT06409780, https://clinicaltrials.gov.