Hybrid Convolutional Neural Network with Intuitionistic Fuzzy Estimations for Detection of Kidney Damage in Patients with Diabetes Mellitus

The early detection of kidney damage in patients with diabetes mellitus is crucial for timely intervention and improved clinical outcomes. In this paper, we propose a novel approach for kidney damage detection by combining the power of convolutional neural networks (CNNs) with intuitionistic fuzzy estimates. The proposed hybrid CNN method leverages the hierarchical feature extraction capabilities of CNNs to automatically learn discriminative features from medical imaging data, while intuitionistic fuzzy estimates are utilized to model uncertainty and imprecision inherent in medical diagnosis. The intuitionistic fuzzy estimates enable the incorporation of expert knowledge and domain-specific information into the classification process, enhancing the interpretability and robustness of the model. We demonstrate the effectiveness of the proposed approach on a dataset of medical images obtained from patients with diabetes mellitus, achieving state-of-the-art performance in kidney damage detection. Our results highlight the potential of hybrid CNNs with intuitionistic fuzzy estimates as a promising tool for early diagnosis and management of kidney complications in diabetic patients.