Preventive DAN- Using Deep Learning for Early Detection

Preventive DAN- Using Deep Learning for Early Detection

Mr. Ashish Modi1, Mr. Maviya Marediya2

1Asst. Prof / Department of Information Technology, Nagindas Khandwala College, Mumbai, Maharashtra, India,

2 Student / Department of Information Technology, Nagindas Khandwala College, Mumbai, Maharashtra, India.

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract -

Diabetes is a global health issue often linked with diabetic neuropathy, specifically diabetic autonomic neuropathy (DAN), which affects the autonomic nerve system, potentially impacting vital organs. DAN is underdiagnosed due to financial constraints, limited access to diagnostic tools, challenges in assessments, and asymptomatic stages.

This research tackles these issues by using electronic health records (EHRS) and deep learning to detect DAN, Machine learning algorithms overcome traditional diagnostic barriers, achieving 85.5% accuracy with a Random Forest (RF) model. A deep learning approach using the (LSTM) model outperforms traditional methods, reaching 95.7% accuracy.

This study advances healthcare technology by demonstrating the effectiveness of AI, particularly deep learning, in detecting DAN accurately and efficiently. It highlights the potential of EHRS in improving clinical diagnosis support, complication detection, and disease monitoring in diabetic patients.

Key Words: Diabetic Autonomic Neuropathy (DAN), Electronic Health Records (EHRs), Long Short-Term Memory (LSTM), Clinical Diagnosis, Early Detection, Cardiac Autonomic Neuropathy (DCAN).