Attai Kingsley F, Amannah Constance, Ekpenyong Moses, Asuquo Daniel E, Akputu Oryina K, Obot Okure U, Ajuga Peterben C, Obi Jeremiah C, Maduka Omosivie, Akwaowo Christie, Uzoka Faith-Michael
Department of Mathematics and Computer Science, Ritman University, Ikot Ekpene, Nigeria.
Department of Computer Science, Ignatius Ajuru University of Education, Port Harcourt, Nigeria.
Healthc Inform Res. 2025 Apr;31(2):125-135. doi: 10.4258/hir.2025.31.2.125. Epub 2025 Apr 30.
This study proposes a mobile-based explainable artificial intelligence (XAI) platform designed for diagnosing febrile illnesses.
We integrated the interpretability offered by local interpretable model-agnostic explanations (LIME) and the explainability provided by generative pre-trained transformers (GPT) to bridge the gap in understanding and trust often created by machine learning models in critical healthcare decision-making. The developed system employed random forest for disease diagnosis, LIME for interpretation of the results, and GPT-3.5 for generating explanations in easy-to-understand language.
Our model demonstrated robust performance in detecting malaria, achieving precision, recall, and F1-scores of 85%, 91%, and 88%, respectively. It performed moderately well in detecting urinary tract and respiratory tract infections, with precision, recall, and F1-scores of 80%, 65%, and 72%, and 77%, 68%, and 72%, respectively, maintaining an effective balance between sensitivity and specificity. However, the model exhibited limitations in detecting typhoid fever and human immunodeficiency virus/acquired immune deficiency syndrome, achieving lower precision, recall, and F1-scores of 69%, 53%, and 60%, and 75%, 39%, and 51%, respectively. These results indicate missed true-positive cases, necessitating further model fine-tuning. LIME and GPT-3.5 were integrated to enhance transparency and provide natural language explanations, thereby aiding decision-making and improving user comprehension of the diagnoses.
The LIME plots revealed key symptoms influencing the diagnoses, with bitter taste in the mouth and fever showing the highest negative influence on predictions, and GPT-3.5 provided natural language explanations that increased the reliability and trustworthiness of the system, promoting improved patient outcomes and reducing the healthcare burden.
本研究提出了一个基于移动设备的可解释人工智能（XAI）平台，用于诊断发热性疾病。
我们整合了局部可解释模型无关解释（LIME）提供的可解释性和生成式预训练变换器（GPT）提供的可解释性，以弥合机器学习模型在关键医疗决策中常常造成的理解和信任差距。所开发的系统采用随机森林进行疾病诊断，LIME用于结果解释，GPT-3.5用于生成易于理解的语言解释。
我们的模型在检测疟疾方面表现出强大性能，精确率、召回率和F1分数分别达到85%、91%和88%。在检测尿路感染和呼吸道感染方面表现中等，精确率、召回率和F1分数分别为80%、65%和72%，以及77%、68%和72%，在敏感性和特异性之间保持了有效平衡。然而，该模型在检测伤寒和人类免疫缺陷病毒/获得性免疫缺陷综合征方面存在局限性，精确率、召回率和F1分数分别较低，为69%、53%和60%，以及75%、39%和51%。这些结果表明存在漏诊的真阳性病例，需要进一步对模型进行微调。LIME和GPT-3.5被整合以提高透明度并提供自然语言解释，从而辅助决策并提高用户对诊断的理解。
LIME图揭示了影响诊断的关键症状，口中苦味和发热对预测的负面影响最大，而GPT-3.5提供的自然语言解释提高了系统的可靠性和可信度，促进了更好的患者预后并减轻了医疗负担。
