• Karlapalem Harshini Devi
• K.Ruchira Reddy
• Kadambari Sai Rohith
• Guttula Pavan Teja

Diabetes is one of the most prevalent chronic diseases worldwide and can lead to severe complications if not detected early. Timely identification of diabetes is crucial to initiate appropriate treatment and prevent disease progression. Additionally, distinguishing between type 1 and type 2 diabetes is essential for personalized treatment strategies. This study proposes a method that not only predicts the future development of diabetes but also determines the type of diabetes a person may have. This classification task is addressed using deep neural networks with hidden layers, and dropout regularization is employed to mitigate overfitting. Multiple parameters are optimized, and a binary cross-entropy loss function is utilized to create a highly accurate prediction model. The effectiveness and suitability of the proposed model, called DLPD (Deep Learning to Predict Diabetes), are demonstrated through extensive experiments conducted on Pima Indian diabetes and diabetes type datasets. The experimental results indicate that our proposed model outperforms existing state-of-the-art methods.

Deep Learning, Keras, Diabetes type 1, type 2, Dropout regularization, Machine learning methods, Diabetes

" DEEP LEARNING APPROACH FOR DIABETES PREDICTION ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 7, page no.d249-d252, July-2023, Available :http://www.jetir.org/papers/JETIR2307332.pdf

" DEEP LEARNING APPROACH FOR DIABETES PREDICTION ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 7, page no. ppd249-d252, July-2023, Available at : http://www.jetir.org/papers/JETIR2307332.pdf