Aletheia: An Offline-First Clinical Decision Support System for Differential Diagnosis in Low-Resource Healthcare Settings
Abstract
Access to specialist clinical expertise remains severely limited across sub-Saharan Africa, where physician-to-patient ratios can fall below 1:25,000 in rural settings. Existing AI-assisted diagnostic tools predominantly require reliable internet connectivity and high-specification hardware, rendering them impractical for frontline healthcare workers in district hospitals and health centres. This paper presents Aletheia, an offline-first clinical decision support system designed for low-resource healthcare contexts across sub-Saharan Africa. Aletheia is built upon Qwen2.5-3B-Instruct, fine-tuned using Quantised Low-Rank Adaptation (QLoRA) on a curated dataset of 27,000 clinical reasoning samples spanning 50 disease conditions with elevated prevalence in East Africa. Evaluation demonstrates a Top-1 diagnostic accuracy of 80% (8 of 10 cases; 95% CI: 49.0-94.3%), Top-3 accuracy of 100% (10 of 10 cases; 95% CI: 72.2-100%), BERTScore-F1 of 0.909, and METEOR of 0.467. These diagnostic figures are computed over a deliberately small set of ten representative clinical case categories, one case each, and are therefore indicative rather than statistically robust; the wide confidence intervals should be read alongside them. The system achieves an Expected Calibration Error (ECE) of 0.275 and passes the Africa Deep Tech Challenge 2026 (ADTC 2026) memory budget constraint of 7168 MB, achieving a peak inference RAM of approximately 3630 MB on the standardised benchmark laptop. These results demonstrate the feasibility of deploying large language model-based clinical reasoning at the primary care level in resource-constrained settings without cloud infrastructure.
Related
- Enhancing Clinical Decision Support and Differential Diagnosis with Knowledge Graphs, and Retrieval Augmented Generation in Generative AI
- Clinical and Dermoscopic Presentation of Rhabdomyomatous Mesenchymal Hamartoma: Differential Diagnosis and Review of the Literature.
- Artificial Intelligence in Eye Disease: A Systematic Review of Screening, Diagnosis, Prognosis, Treatment-Response Prediction, and Clinical Decision Support
- Differential Diagnosis of Orofacial and Jaw Pain: A Clinical Review for Physicians
- Cost-Aware Active Feature Acquisition for Differential Diagnosis under Realistic Clinical Availability Constraints
BibTeX
@misc{walusimbi2026aletheia,
title = {{Aletheia: An Offline-First Clinical Decision Support System for Differential Diagnosis in Low-Resource Healthcare Settings}},
author = {Joseph Walusimbi and Ann Move Oguti and Abubakhari Sserwadda and Precious Boss Kasasira and Charles Brian Okoboi},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2607.24814},
eprint = {2607.24814},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.24814},
}
RIS
TY - UNPB
TI - Aletheia: An Offline-First Clinical Decision Support System for Differential Diagnosis in Low-Resource Healthcare Settings
AU - Joseph Walusimbi
AU - Ann Move Oguti
AU - Abubakhari Sserwadda
AU - Precious Boss Kasasira
AU - Charles Brian Okoboi
PY - 2026
DA - 2026/09/04
PB - arXiv
DO - 10.48550/arxiv.2607.24814
UR - https://arxiv.org/abs/2607.24814
AB - Access to specialist clinical expertise remains severely limited across sub-Saharan Africa, where physician-to-patient ratios can fall below 1:25,000 in rural settings. Existing AI-assisted diagnostic tools predominantly require reliable internet connectivity and high-specification hardware, rendering them impractical for frontline healthcare workers in district hospitals and health centres. This paper presents Aletheia, an offline-first clinical decision support system designed for low-resource healthcare contexts across sub-Saharan Africa. Aletheia is built upon Qwen2.5-3B-Instruct, fine-tuned using Quantised Low-Rank Adaptation (QLoRA) on a curated dataset of 27,000 clinical reasoning samples spanning 50 disease conditions with elevated prevalence in East Africa. Evaluation demonstrates a Top-1 diagnostic accuracy of 80% (8 of 10 cases; 95% CI: 49.0-94.3%), Top-3 accuracy of 100% (10 of 10 cases; 95% CI: 72.2-100%), BERTScore-F1 of 0.909, and METEOR of 0.467. These diagnostic figures are computed over a deliberately small set of ten representative clinical case categories, one case each, and are therefore indicative rather than statistically robust; the wide confidence intervals should be read alongside them. The system achieves an Expected Calibration Error (ECE) of 0.275 and passes the Africa Deep Tech Challenge 2026 (ADTC 2026) memory budget constraint of 7168 MB, achieving a peak inference RAM of approximately 3630 MB on the standardised benchmark laptop. These results demonstrate the feasibility of deploying large language model-based clinical reasoning at the primary care level in resource-constrained settings without cloud infrastructure.
ER -