Development of a Mobile Diagnostic Application for Rapid Infectious Disease Identification in Livestock

Thesis Abstract

The rapid and accurate identification of infectious diseases in livestock remains a critical challenge in veterinary medicine, particularly in resource-limited rural settings where delays in diagnosis can lead to significant economic losses, widespread outbreaks, and threats to food security. Traditional diagnostic methods often require laboratory infrastructure, specialized personnel, and extended processing times, which hinder timely decision-making and intervention. This study aims to develop and evaluate a mobile-based diagnostic application designed to facilitate rapid, accessible, and cost-effective identification of prevalent infectious diseases in livestock. The specific objectives include designing the application architecture, integrating symptom-based decision algorithms, validating its diagnostic accuracy against laboratory tests, and assessing user acceptance among livestock farmers and veterinary practitioners. To achieve these objectives, a mixed-methods research design was adopted, combining qualitative system development approaches with quantitative validation and user evaluation. The study population comprised 500 livestock farmers and 50 veterinary extension officers across three rural districts, selected using stratified random sampling to ensure representativeness. Data collection instruments included structured diagnostic test cases, focus group discussion guides, user satisfaction questionnaires, and software usability scales. The diagnostic algorithms embedded within the application were informed by a comprehensive review of veterinary diagnostic protocols and the Theory of Planned Behavior (TPB) to model user acceptance and behavioral determinants influencing adoption. Validation of the application’s diagnostic performance employed a diagnostic accuracy study, comparing the app’s results to laboratory-confirmed diagnoses using sensitivity, specificity, positive predictive value, and negative predictive value metrics. Usability and acceptance data were analyzed through descriptive statistics, inferential analysis (ANOVA), and thematic analysis for qualitative feedback. The anticipated findings suggest that the mobile diagnostic application will demonstrate high diagnostic accuracy, with sensitivity and specificity exceeding 85% across targeted infectious diseases such as foot-and-mouth disease, brucellosis, and contagious bovine pleuropneumonia. User testing is expected to reveal high levels of satisfaction and willingness to adopt the tool, influenced by factors modeled through the TPB constructs, including perceived ease of use, perceived usefulness, and social influence. The integration of symptom-based decision rules within the app is projected to enable livestock farmers to make preliminary disease assessments within minutes, significantly reducing reliance on distant diagnostic laboratory facilities. The study also expects to identify key barriers and facilitators to mobile health technology adoption in rural livestock management contexts. This research contributes novel insights into the application of mobile health (mHealth) solutions for veterinary diagnostics, expanding the theoretical framework of technology acceptance in agricultural settings, rooted in the TPB. It offers a validated, user-centered prototype that enhances the capacity for early detection and response to infectious diseases, thereby mitigating economic losses and improving animal health management practices in rural communities. Furthermore, the findings provide empirical evidence to support policy development for integrating ICT-based diagnostic tools into national livestock health strategies. The study concludes that the developed mobile diagnostic application is a feasible, effective, and scalable tool that can be widely adopted in livestock health surveillance systems. Recommendations include fostering collaborations between veterinary authorities, ICT developers, and rural communities to facilitate training, improve app functionalities based on ongoing feedback, and ensure sustainability. Future research should explore longitudinal impact assessments and integration with regional disease monitoring networks, as well as adapting the application for other animal species and diseases to maximize its utility across diverse livestock management contexts.