@article{jcsa20251323,
author={{Ogunseye, Elizabeth O. and Oladejo, Ezekiel A. and Olaleye, Isaac and Adeyemo, Adesesan B.},
title={Towards Domain-Aware Language Models for Low-Resource Healthcare: Fine-Tuning LLMs and SLMs with Parallel English¨CYoruba Maternal Health Data},
journal={Journal of Computer Sciences and Applications},
volume={13},
number={2},
pages={54--58},
year={2025},
url={https://pubs.sciepub.com/jcsa/13/2/3},
issn={2328-725X},
abstract={Healthcare communication barriers significantly impact maternal health outcomes in multilingual communities, particularly in Sub-Saharan Africa where indigenous languages dominate daily communication while medical resources remain primarily in colonial languages. This study presents a systematic approach to developing domain-aware language models specifically tailored for maternal health communication in low-resource settings. We introduce a comprehensive parallel English¨CYoruba maternal health dataset comprising 7,000 translated and verified sentence pairs covering prenatal care, childbirth and postnatal support. Our methodology involves fine-tuning both Large Language Models (LLMs) including GPT-3.5-turbo and LLaMA-2-7B, and Small Language Models (SLMs) such as DistilBERT, mBERT, and XLM-R across multiple evaluation dimensions including translation quality, domain-specific terminology accuracy, and clinical relevance metrics. Results demonstrate that domain-specific fine-tuning significantly improves performance over general-purpose models, with the GPT-3.5-turbo variant achieving a BLEU score of 0.78 on the held-out test set and medical terminology accuracy of 89.3%. Fine-tuned models demonstrate substantial improvements in handling culture-specific maternal health concepts and traditional medicine terminology. This work contributes to bridging the digital health divide in low-resource settings and provides a replicable framework for developing multilingual healthcare AI systems that can effectively serve diverse linguistic communities while maintaining cultural and clinical sensitivity.},
doi={10.12691/jcsa-13-2-3}
publisher={Science and Education Publishing}
}