Key result
The proposed Deep CNN-R-LSTM hybrid network architecture achieved excellent classification performance for IoT healthcare sensor signals, with 99.02% accuracy, 99.47% specificity, and 99.56% sensitivity.
Why the study?
Automated identification and classification technologies are needed to help clinicians make more accurate and timely diagnoses from healthcare IoT sensor signals.
A novel deep learning hybrid network (CNN-R-LSTM) demonstrated high accuracy (99.02%) in classifying healthcare IoT sensor signals into health states.
Should not yet change IoT remote-monitoring practice; leaves open prospective validation of hybrid CNN-LSTM models in CV cohorts.
IoT is a fascinating technology in today's IT world, in which items may transmit data and interact through intranet or internet networks. TheInternet of Things (IoT) has shown a lot of promise in connecting various medical equipment, sensors, and healthcare specialists to provide high-quality medical services from afar. As a result, patient safety has improved, healthcare expenses have fallen, healthcare service accessibility has increased, and operational efficiency has increased in the healthcare industry. Healthcare IoT signal analysis is now widely employed in clinics as a critical diagnostic tool for diagnosing health issues. In the medical domain, automated identification and classification technologies help clinicians make more accurate and timely diagnoses. In this paper, we have proposed a Deep Learning-Based hybrid network architecture (CNN-R-LSTM (DCRL)) that combines the characteristics of a Convolutional Neural Network (CNN) and a Recurrent Neural Network (RNN) based long-short-term memory (LSTM) to diagnose IoT sensor signals and classify them into three categories: healthy, patient, and serious illness. Deep CNN-R-LSTM Algorithm is used for classify the IoT healthcare data support via a dedicated neural networking model. For our study, we have used the MIT-BIH dataset, the Pima Indians Diabetes dataset, the BP dataset, and the Cleveland Cardiology datasets. The experimental results revealed great classification performance in accuracy, specificity, and sensitivity, with 99.02 percent, 99.47 percent, and 99.56 percent, respectively. Our proposed DCLR model is based on healthcare IoT Centre inputs enhanced with the centenary, which may aid clinicians in effectively recognizing the health condition.
No takes yet. Share an insight, caveat, or question.
S. et al. (2023) studied Various (Arrhythmia, Diabetes, Hypertension, Cardiac problems). Deep CNN-R-LSTM (DCRL) architecture vs. CNN architecture was evaluated on Classification accuracy. The proposed Deep CNN-R-LSTM hybrid network architecture achieved excellent classification performance for IoT healthcare sensor signals, with 99.02% accuracy, 99.47% specificity, and 99.56% sensitivity.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: