• meenakshi
• Harjinder kaur

Diabetes is a global health concern affecting millions, with the number of cases expected to rise significantly in the future. Early detection of diabetes is critical to prevent complications and ensure a better quality of life for affected individuals. This paper presents an innovative approach for the early detection of diabetes using deep learning techniques, specifically Convolutional Long Short-term Memory (CLSTM) networks. We conducted experiments using the Pima Indians Diabetes Database (PIDD) to evaluate our model's performance and compared it with existing methods.Our approach incorporates an efficient data preprocessing technique called multivariate imputation by chained equations to enhance the accuracy of diabetes prediction. The results demonstrate that our CLSTM-based model outperforms traditional machine learning algorithms, such as Naïve Bayes, Support Vector Machine, and Decision Trees, in terms of accuracy. This suggests that deep learning holds great promise in improving diabetes prediction and early intervention.

Keywords: Convolutional Long Short-term Memory (CLSTM), Diabetes prediction, Deep learning, Data preprocessing, Early detection.

"Using Deep Learning and LSTM to Detect Diabetes Early", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 9, page no.g94-g105, September-2023, Available :http://www.jetir.org/papers/JETIR2309629.pdf

"Using Deep Learning and LSTM to Detect Diabetes Early", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 9, page no. ppg94-g105, September-2023, Available at : http://www.jetir.org/papers/JETIR2309629.pdf