Development of a Mobile App for Automated Hematology Test Result Interpretation

Thesis Abstract

Hematology testing remains a cornerstone of diagnostic medicine, yet the interpretative process of test results often relies heavily on manual analysis by laboratory personnel, which may introduce delays, inconsistencies, and potential errors, particularly in resource-constrained settings. This study aims to develop a mobile application that automates the interpretation of hematology test results, thereby enhancing accuracy, efficiency, and accessibility in clinical decision-making. The specific objectives include designing the app architecture based on algorithmic interpretation models, integrating relevant clinical reference ranges, and evaluating the app’s performance in real-world settings. Employing a mixed-method research design, the study combines quantitative and qualitative approaches. The quantitative phase involves the development of a prototype mobile app utilizing a rule-based decision engine, aligned with established clinical guidelines such as the World Health Organization's hematological reference standards. A total sample size of 200 hematology test reports, collected from the hematology laboratories of a tertiary hospital, will be used to validate the app's interpretative accuracy. Data collection will utilize a structured data extraction sheet for laboratory results and a user-feedback questionnaire to assess usability and interpretative confidence. The app’s performance will be analyzed through sensitivity, specificity, positive predictive value, and negative predictive value calculations, employing descriptive statistics and receiver operating characteristic (ROC) curve analysis. The qualitative data from user feedback will undergo thematic analysis to identify usability strengths and challenges. It is expected that the developed app will demonstrate interpretative accuracy comparable to that of experienced hematologists, with sensitivity and specificity exceeding 90%. Furthermore, the app is anticipated to significantly reduce the turnaround time for test result interpretation and improve consistency across different user groups. The integration of machine-readable algorithms and clinical guidelines within the app is expected to facilitate rapid, standardized interpretations, thereby reducing diagnostic errors attributable to manual analysis or misinterpretation. Theoretically, the study adopts the Health Belief Model (HBM) to understand users’ acceptance and perceived usefulness of the mobile app, alongside the Technology Acceptance Model (TAM) to examine factors influencing user adoption. The empirical findings are expected to contribute novel insights into the application of ICT solutions in hematology diagnostics, filling existing gaps in the literature regarding automated interpretation tools tailored for resource-limited healthcare environments. This research will demonstrate that mobile health technologies can significantly augment clinical workflows, particularly where laboratory expertise or infrastructure is limited. The study’s main conclusion underscores the feasibility and potential benefits of integrating mobile app solutions into hematology diagnostics, emphasizing improvements in interpretive accuracy, standardization, and accessibility. Recommendations arising from the research include the adoption of the app in routine diagnostic workflows, further refinement through feedback-driven updates, and expansion of its functionality to incorporate other hematological tests. The study advocates for the broader implementation of ICT-based diagnostic tools as a means to enhance healthcare delivery outcomes and reduce diagnostic errors in diverse clinical settings. Future research should explore longitudinal assessments of clinical impact, integration with electronic health records, and adaptations for broader global health applications.