Smart Health Monitoring System with Wireless Networks to Detect Kidney Diseases

Area of Science:

• Health Informatics
• Artificial Intelligence in Medicine
• Remote Healthcare Technologies

Background:

• Rising healthcare costs and emerging global illnesses necessitate a shift towards patient-centric healthcare models.
• Traditional hospital-based services face challenges in providing efficient and scalable patient monitoring.
• The Internet of Things (IoT) and cloud computing offer potential solutions for remote patient care and data management.

Purpose of the Study:

• To develop an optimal decision support system for identifying Chronic Kidney Disease (CKD) using cloud and IoT technologies.
• To enhance remote healthcare services by enabling efficient and accurate CKD detection.
• To improve patient outcomes through timely and accessible medical data.

Main Methods:

• A novel algorithm, Improved Simulated Annealing-Root Mean Square -Logistic Regression (ISA-RMS-LR), was developed for CKD identification.
• The methodology involved four subprocesses: data collection, preprocessing, feature selection (FS), and classification.
• Simulated Annealing (SA) was integrated into the feature selection stage to optimize the classifier's performance.

Main Results:

• The ISA-RMS-LR model demonstrated high efficacy in classifying CKD using a benchmark dataset.
• Performance metrics included: 99.46% sensitivity, 99.26% accuracy, 98% specificity, 99.63% F-score, and 98.29% kappa value.
• The system facilitates rapid medical data transmission, real-time patient tracking, and efficient record management.

Conclusions:

• The proposed ISA-RMS-LR model provides an effective and accurate solution for remote CKD identification.
• The cloud and IoT-based system supports patient-centric healthcare, improving data accessibility and patient monitoring.
• Future enhancements could focus on optimizing hospital capacity and managing large patient populations with minimal delay.