Enhancing the Interpretability of Malaria and Typhoid Diagnosis with Explainable AI and Large Language Models
importance 7/10
ai assisted
Explainable Artificial Intelligence (XAI) tropical disease diagnostics stage: research mixed
Problem
Malaria and typhoid are hard to diagnose accurately due to similar symptoms, and ML predictions are opaque to healthcare workers.
Why it matters
The abstract states prompt, accurate diagnosis is crucial to reduce mortality in tropical regions.
What the AI did
Explainable AI (LIME) and LLMs like GPT were used to explain which symptoms drove model diagnostic decisions for healthcare workers.
Results
not stated in source
Impact
Aims to build trust and transparency in diagnostics; adoption or measured benefit not stated.
Record
| Venue | Tropical Medicine and Infectious Disease |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Kingsley Friday Attai, Moses E. Ekpenyong, Constance Amannah, Daniel Ekpenyong Asuquo, Peterben Ajuga, Okure Udo Obot, Ekemini Johnson, Anietie E. John et al. |
| Citations | 27 (OpenAlex, at import time) |
| References | 50 |
| Model or method | LIME, GPT (large language models) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/tropicalmed9090216 |
| Open access | gold |
| Keywords | malaria, typhoid, explainable AI, LIME, large language models, diagnosis |
Importance rating
7
significant — thousands to millions affected
Abstract states malaria and typhoid are prevalent in tropical regions and prompt accurate diagnosis is crucial to reduce mortality; large affected population implied.