SmartCardio: Smart IoT-Based Heart Disease Diagnosis with Optimal Feature Selection and Dense Non-Local Coordinate-Attention Network

Presently, Heart Disease (HD) is considered as the major reason of death. The prediction of HD is a challenging since that demands both sophisticated knowledge and experience. Recently, healthcare organizations have begun using Internet of Things (IoT) technology for collecting sensor data for the d...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2024 International Conference on Advancement in Renewable Energy and Intelligent Systems (AREIS) s. 1 - 6
Hlavní autoři: Venifa Mini, G., Lincy, C T, Pravin Rose, T, Reeba Rex, S, Saji, K S, John, Dayana. V.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 05.12.2024
Témata:
Accuracy
Black-Winged Kite Algorithm
Coordinate-Attention
Densenet
Distributed databases
Football Team Training Optimizer
Heart
Heart Disease Diagnosis
Internet of Things
Medical diagnosis
Prognostics and health management
Renewable energy sources
System performance
Training
Wearable Health Monitoring Systems
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Presently, Heart Disease (HD) is considered as the major reason of death. The prediction of HD is a challenging since that demands both sophisticated knowledge and experience. Recently, healthcare organizations have begun using Internet of Things (IoT) technology for collecting sensor data for the diagnosis and prognosis of HD. The diagnosis of HD is the subject of several studies; yet the accuracy of findings is poor. To solve this problem, asmart IoT-based heart disease diagnosis using Deep Learning (DL) (SmartCardio) is proposed to more precisely detectHD. The patient's data are initially tracked by the smartwatch and heart monitor gadget. The received sensor data is classified as normal or HD. The repetitive data are minimized through the use of Hadoop distributed file system (HDFS) and performed pre-processing. After that, a Black-Winged Kite Algorithm (BWiKA) is employed to choose the important features. Lastly, a Dense Non-Local Team Training Coordinate-Attention (DNL-TTCA) network is used to classify these features as HD or not HD. The Python platform is used to simulate the SmartCardio framework, and the system's performance is evaluated by comparing with existing techniques. The results demonstrate that the SmartCardio framework outperforms other classifiers with 99.51% accuracy, respectively.
DOI:10.1109/AREIS62559.2024.10893662