ALGORITHMS FOR INTELLECTUAL PROCESSING OF IoT DATA BASED ON NEURAL NETWORKS FOR DIABETES CONTROL AND PROGNOSIS
DOI:
https://doi.org/10.47390/ts-v3i7y2025N2Keywords:
IoT, diabetes mellitus, neural networks, intelligent data processing, monitoring and prediction, artificial intelligence.Abstract
Diabetes mellitus is a rapidly spreading chronic disease. IoT sensors enable real-time monitoring of glucose, heart rate, and other physiological parameters. This study compared CNN, RNN, and LSTM models, with LSTM achieving the highest accuracy (92.1%). Eliminating duplicate records improved model performance.
References
1. Farooq, M. S., & Khan, M. A. (2023). Role of Internet of Things in diabetes healthcare: Network architecture and challenges. *Journal of Medical Systems*, 47(1), 1–15.
2. Valsalan, P., & Kumar, S. (2022). IoT based expert system for diabetes diagnosis and management. *Journal of Healthcare Engineering*, 2022, 1–10.
3. Ahmed, A., et al. (2023). Performance of AI models in estimating blood glucose level using non-invasive wearable devices. *Computers in Biology and Medicine*, 165, 105234.
4. Mansour, M., & Al-Kahtani, M. (2024). Wearable devices for glucose monitoring: A review. *Sensors*, 24(3), 1–20.
5. Liu, Y., Zhang, L., & Li, J. (2025). Advanced applications in chronic disease monitoring using IoT devices. *Frontiers in Public Health*, 13, 1510456.()
6. Shukurillayev, K. S. (2023). Sun’iy neyron tarmoqlarga asoslangan tibbiy diagnostikaning asosiy bosqichlari. Zenodo. [https://zenodo.org/record/7886536] (https://zenodo.org/record/7886536).
This work is licensed under a 