Smart City Gnosys
Smart city article details
| Title | Transforming Security In Internet Of Medical Things With Advanced Deep Learning-Based Intrusion Detection Frameworks |
|---|---|
| ID_Doc | 58830 |
| Authors | Sharma N.; Shambharkar P.G. |
| Year | 2025 |
| Published | Applied Soft Computing, 180 |
| DOI | http://dx.doi.org/10.1016/j.asoc.2025.113420 |
| Abstract | The Internet of Things (IoT) revolutionizes industries like healthcare, agriculture, smart cities, and weather forecasting by enabling vast device networks to interact and transmit information. However, traditional network security methods are inadequate for IoT due to the limited storage and processing power of IoT devices. As IoT systems encounter numerous cyber threats, it is essential to develop innovative security approaches. Intrusion detection systems (IDS) are a popular mechanism for identifying and preventing systems from attacks. Therefore, this paper explores integrating deep learning (DL) models into IDS frameworks to enhance their capabilities. Deep learning is a subset of Artificial Intelligence (AI) that can extract complex patterns from data, adapt to new threats, and provide high accuracy and real-time detection. To provide the solution, this research presents novel deep learning models, including Embed-Net (Categorical Embedding Neural Network), the collaborative ConvNet-SVM model (combination of Convolutional Neural Network (CNN) and Support Vector Machine (SVM)), and the DeepSVM-Net (Deep Neural Network emulated by SVM), which are designed to detect network-based attacks on the Internet of Medical Things (IoMT). These models are evaluated using real-time benchmark datasets: ECU-IoHT, NF-BoT-IoT, and Wustl-HDRL-2024 by applying advanced preprocessing techniques that demonstrate superior performance by achieving an accuracy of 0.9990, 0.9959, and 0.9987with the lowest FNR of 0.0001, 0.0059, and 0.0014 by all three proposed models as compared to traditional methods. Furthermore, this paper conducted an ablation study which shows that our models have achieved outstanding performance with minimum training and validation losses of 0.0034 and 0.0008 for EmbedNet, 0.0018 and 0.0008 for ConvNet-SVM, and 0.0044 and 0.0016 for DeepSVM-Net. This research has enhanced IDS effectiveness, accuracy, and resilience by contributing significantly to cybersecurity, with the proposed models showing a remarkable improvement percentage ranging from 3.5 % to 7.5 % over the state-of-the-art. © 2025 Elsevier B.V. |
| Author Keywords | Cyber-Attacks; Deep Learning; Internet of Medical Things (IoMT); Intrusion Detection System (IDS); Security |
Similar Articles
| Id | Similarity | Authors | Title | Published |
|---|---|---|---|---|
30732  | 0.926 | Amine M.S.; Nada F.A.; Hosny K.M. | Improved Model For Intrusion Detection In The Internet Of Things | Scientific Reports, 15, 1 (2025) |
| 11116 | 0.924 | Naeem H.; Alsirhani A.; Alserhani F.M.; Ullah F.; Krejcar O. | Augmenting Internet Of Medical Things Security: Deep Ensemble Integration And Methodological Fusion | CMES - Computer Modeling in Engineering and Sciences, 141, 3 (2024) |
| 5144 | 0.924 | Liao H.; Murah M.Z.; Hasan M.K.; Aman A.H.M.; Fang J.; Hu X.; Khan A.U.R. | A Survey Of Deep Learning Technologies For Intrusion Detection In Internet Of Things | IEEE Access, 12 (2024) |
| 9648 | 0.921 | Alsoufi M.A.; Razak S.; Siraj M.M.; Nafea I.; Ghaleb F.A.; Saeed F.; Nasser M. | Anomaly-Based Intrusion Detection Systems In Iot Using Deep Learning: A Systematic Literature Review | Applied Sciences (Switzerland), 11, 18 (2021) |
| 33032 | 0.912 | Dawoud A.; Sianaki O.A.; Shahristani S.; Raun C. | Internet Of Things Intrusion Detection: A Deep Learning Approach | 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020 (2020) |
| 33508 | 0.909 | Saini K.S.; Chaudhary S. | Investigation On Attack Detection In Iot Networks: A Study And Analysis Of The Existing Machine Learning And Deep Learning Techniques | 3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025 (2025) |
| 36080 | 0.902 | Ahanger T.A.; Ullah I.; Algamdi S.A.; Tariq U. | Machine Learning-Inspired Intrusion Detection System For Iot: Security Issues And Future Challenges | Computers and Electrical Engineering, 123 (2025) |
| 23626 | 0.902 | Hazman C.; Guezzaz A.; Benkirane S.; Azrour M. | Enhanced Ids With Deep Learning For Iot-Based Smart Cities Security | Tsinghua Science and Technology, 29, 4 (2024) |
| 23835 | 0.9 | Arabiat A.; Altayeb M. | Enhancing Internet Of Things Security: Evaluating Machine Learning Classifiers For Attack Prediction | International Journal of Electrical and Computer Engineering, 14, 5 (2024) |
| 34139 | 0.899 | Selvam R.; Velliangiri S. | Iotsdl: Internet Of Things Security For Deep Learning Techniques-A Research Perspectives | 2023 International Conference on Computer Communication and Informatics, ICCCI 2023 (2023) |