Early detection of diabetes is crucial for preventing long-term complications and improving patient outcomes. Conventional prediction models often rely on single-feature analysis or traditional machine learning techniques, which may not capture complex interactions among clinical parameters. This paper proposes an attention-assisted deep learning framework for diabetes prediction using only structured clinical data, including glucose level, BMI, insulin, age, blood pressure, and skin thickness. A feed-forward neural network is employed to model non-linear relationships between features, while an attention mechanism emphasizes the most relevant attributes for each patient. Experimental results demonstrate that the proposed approach outperforms traditional machine learning models in terms of accuracy, F1-score, and AUC, making it an effective and computationally efficient solution for real-time clinical decision support.
Sonali et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: