Self-Aware AI-Based Anomaly Detection for AI IoT Enabled Smart Healthcare System
The Internet of Things (IoT) based smart healthcare systems can significantly enhance the quality of patient care. These systems can enable healthcare providers to monitor patient conditions in real time. IoT-based smart healthcare systems have the potential to revolutionize the way we provide care. Anomaly detection is crucial in IoT-based smart healthcare systems due to numerous reasons. It plays a dominant role in proactive health-care analysis, especially in brain diseases. In brain Magnetic Res-onance Imaging (MRI), anomaly detection helps in identifying several abnormalities, such as cysts, strokes, infections, vascular and developmental malformations, and neurological disorders like Alzheimer's and multiple sclerosis. The existing approaches of brain MRI anomaly detection, and anomaly detection in IoT-based smart healthcare systems utilizes static AI models. Conventional AI based models can be trained once and cannot tackle the changes in real time. The population health and the nature of disease consistently change, and conventional AI based models can't adapt to those changes. Conventional AI based system cannot tackle the changes caused by variations like diseases and public health and become ineffective over the time. Self-aware AI models can address these issues as they can learn and adapt to their environment. Self-aware AI-based systems have the potential to change lives through their ability to learn, adapt, and improve over time. In this paper, we proposed a novel approach to anomaly detection based on self-aware AI in IoT-based smart healthcare systems. Additionally, an algorithm for anomaly detection in MRI images, namely Self-Aware AI based Anomaly Detection (SAAD) is proposed. Our proposed system can detect anomalies in brain MRI images. We achieved 98% accuracy and 0.99 precision and recall, outperforming the existing approaches. The proposed self-aware AI based anomaly detection framework for IoT-based smart healthcare systems has the potential to revolutionize healthcare.
Authors
- Muhammad Farrukh Shahid (ORCID: https://orcid.org/0009-0004-8787-1868)
- M. Hassan Tanveer (ORCID: https://orcid.org/0000-0001-9266-6368)
- Yusra Shereen
Institutions
- Kennesaw State University (US)
- National University of Computer and Emerging Sciences (PK)
Publication Details
- Published
- 2025-04-13
- DOI
- https://doi.org/10.1109/iccit63348.2025.10989421
- Citations
- 1
- Primary Topic
- Anomaly Detection Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 2.48