Deep learning-based methods have made significant

The gap

Evidence profile

Sourced from the future work and abstract of the source papers, classified as general, spanning 3 journals.

Supporting evidence — 3 representative gaps

• Arrhythmia Classification Using Machine Learning Techniques on ECG Data (2026) · International Journal of Scientific Engineering and Research · doi

  Despite the effectiveness of the proposed solution, there is still much room for improvement and further advancements in the field Systems for Real-Time ECG Monitoring Research can be conducted to design systems for detecting arrhythmias in real time by monitoring the ECG data in an ongoing fashion and alerting users to any abnormality detected in heart rhythms. Personalized Arrhythmia Predictions Heart rhythms can vary among different patients. It would be interesting to investigate how personalized machine learning algorithms can help increase prediction accuracy Combining Smartwatch Volume 14 Issue 5, May 2026 www.ijser.in Licensed Under Creative Commons Attribution CC BY Paper ID: SE26508173840DOI: https://dx.doi.org/10.70729/SE265081738407 of 8 International Journal of Scientific Engineering and Research (IJSER) ISSN (Online): 2347-3878 SJIF (2025): 8.036 [16] M. G. e. al., “Hybrid Deep Learning Framework for ECG Arrhythmia Classification,” Knowledge-Based Systems, vol. 302, no. --, p. 111234, 2025. - [17] D. K. e. al., “Hybrid CNN-Transformer Model for ECG Arrhythmia Classification,” Scientific Reports, vol. 15, no. 1, p. 92582, 2025. - [18] S. B. e. al., “Hybrid Machine Learning Models for ECG Classification,” PLOS ONE, vol. 20, no. 5, p. e0334607, 2025. - [19] G. a. R. Mark, “The Impact of the MIT-BIH Arrhythmia Database,” IEEE Engineering in Medicine and Biology Magazine, vol. 20, no. 3, pp. 45-50, 2001. [20] M. O. a. R. B. R. P. de Chazal, “Automatic Classification of Heartbeats Using ECG Morphology and Heartbeat Interval Features,” IEEE Transactions on Biomedical Engineering, vol. 51, no. 7, pp. 1196-1206, 2004. [21] R. A. e. al., “Automated Diagnosis of Arrhythmia Using Combination of CNN and ECG Signals,” Information Sciences, no. 145-416, pp. 190-198, 2017. [22] P. R. e. al., “Cardiologist-Level Arrhythmia Detection with Deep Neural Networks,” Nature Medicine, vol. 25, no. 1, pp. 65-69, 2017. - [23] T. I. a. M. G. S. Kiranyaz, “Real-Time Patient-Specific ECG Classification by 1-D CNN,” IEEE Transactions on Biomedical Engineering, vol. 63, no. 3, pp. 664-675, 2016. with the ECG Data from Recent advances in wearable technologies allow tracking of heart rate via ECG sensors embedded into smartwatches. The proposed arrhythmia model could be integrated with those sensors, and this would benefit millions of users. Deep Learning Techniques Application Another area where studies could be carried out is the application of deep learning techniques such as the Long Short Term Memory technique, Transformers, and deep reinforcement learning. Big Data Analysis for Hospitals There are many medical databases available that can be used to train the ECG model.

  generalfuture workevidence 5/5
  Keywords: arrhythmia learning deep classification engineering systems real time heart hybrid model ieee proposed there monitoring
• Pressure Ulcer Detection and Prediction Using Sensor Fusion and AI-Based System (2026) · International Journal of Drug Delivery Technology · doi

  Future improvements of the proposed system can focus on enhancing accuracy, usability, and practicality. Adding more physiological sensors, such as heart rate, oxygen saturation, or skin pH sensors, could offer a broader view of patient health. Machine learning methods, including deep learning models like LSTM or hybrid CNN-LSTM architectures, may capture complex patterns in sensor IJDDT, Volume 16 Issue 32s, 2026 Page 146 Pressure Ulcer Detection and Prediction Using Sensor Fusion and AI-Based System caregivers data to better predict early-stage pressure ulcers. Mobile and cloud-based platforms can be used to send real-time alerts and allow remote monitoring by and healthcare professionals. Additionally, conducting larger clinical studies with improve different patient groups will help generalization system’s and effectiveness in real-world situations. Personalized tailored predictive analytics can also support monitoring and preventive strategies based on individual patient risk profiles. These changes aim to make the system more reliable, efficient, and useful for both hospital and home-care settings. confirm the VI. REFERENCES S. Alkhalefah, I. AlTuraiki, and N. Altwaijry, “Advancing Diabetic Foot Ulcer Care: AI and Generative AI Approaches for Classification, Prediction, Segmentation, and Detection,” Healthcare, vol. 13, no. 6, 2025. “AI-Enhanced Multimodal Sensing for Early Detection of Diabetic Foot Ulcers,” Int. J. Res. Appl. Sci. Eng. Technol., vol. 13, May 2025. L. Wei, H. Lv, C. Yue, Y. Yao, N. Gao, and Q. Chai, “A Machine Learning Algorithm-Based Predictive Model for Pressure Injury Risk in Emergency Patients: A Prospective Cohort Study,” Int. Emerg. Nurs., 2024. S. Morad, A. M. Biju, and A. H. Morad, “A Wearable Pad for Detecting and Monitoring Heel Pressure Ulcer: Preliminary Study,” Int. J. Eng. Technol., vol. 17, no. 2, pp. 95–102, Apr. 2025. S. Bao, Y. Wang, L. Yao, S. Chen, X. Wang et al., “Research Trends and Hot Topics of Wearable Sensors in Wound Care: A Bibliometric Analysis,” Heliyon, 2024. Y. Chang, J. H. Kim, H. W. Shin, C. Ha, S. Y. Lee, and T. Go, “Diagnosis of Pressure Ulcer Stage Using On-Device AI,” Appl. Sci., vol. 14, no. 16, art. 7124, Aug. 2024. “Pressure Ulcers Classification and Detection,” Int. J. Innov. Res. Sci. Eng. Technol., vol. 12, no. 4, Apr. 2023. R.

  generalfuture workevidence 5/5
  Keywords: pressure system ulcer detection based sensors patient learning ulcers monitoring care technol machine lstm sensor
• A deformable lead-attention fusion network for multi-label ECG classification integrating clinical metadata (2026) · Biomedical Physics & Engineering Express · doi

  While deep learning-based methods have made significant progress in automating ECG classification, they are often limited by the variability of ECG signals, the frequent presence of multiple cardiac abnormalities, and the inadequate integration of diverse clinical data.

  generalabstractevidence 3/5
  Keywords: deep learning based made significant progress automating classification often limited variability signals frequent presence multiple

Questions about this gap