Key result
A proposed MU-CSO-assisted hybrid clustering model using ECG signals and clinical data was developed and validated against conventional methods for automated heart disease prediction.
Why the study?
Developing an automated prediction system to detect heart disease using ECG signals and clinical data remains a challenging task in medical data mining.
Does a MU-CSO-assisted hybrid clustering model improve the detection of heart disease using clinical data and ECG signals compared to conventional methods?
Does a MU-CSO-assisted hybrid clustering model improve the detection of heart disease using clinical data and ECG signals compared to conventional methods?
A novel MU-CSO-assisted hybrid clustering model using clinical data and ECG signals was developed for automated heart disease prediction.
New hybrid clustering model shows promise for automated prediction; leaves open prospective validation before clinical use.
Detection of heart disease has become a significant topic in the medicalindustry while analyzing clinical information. Over the past years,heart disease diagnosis has been one of the emerging techniques indata mining. Besides this, developing an automated prediction systemis a challenging task, even when detecting the disease by ECG signaland data. Here, it proposes a flexible method to detect heart diseasewith the domain of both signal processing and data mining. Initially,the acquired features from the weighted feature extraction approachare fed into the hybrid clustering model; in turn, the two differentclustering results are taken to determine the finest output. Thehybrid clustering model is developed by K-Means clustering that issuperimposed with DBCAN, where the centroid and value are optimizedwith the aid of Modified Updating-based Chicken Swarm Optimization(MU-CSO). Simultaneously, the ECG signals are garnered and decomposedusing Discrete Wavelet Transform (DWT). Due to the curse ofdimensionality, the Principal Component Analysis (PCA) is deployed.Consequently, the MU-CSO-assisted hybrid clustering is employed todiagnose heart disease by ECG signal. Therefore, the efficacy of theprediction model is validated with various metrics and comparedagainst conventional methods.
No takes yet. Share an insight, caveat, or question.
Sonawane et al. (2022) studied Heart disease. MU-CSO-assisted hybrid clustering model (K-Means + DBSCAN) with DWT and PCA vs. Conventional methods was evaluated on Efficacy of the prediction model. A proposed MU-CSO-assisted hybrid clustering model using ECG signals and clinical data was developed and validated against conventional methods for automated heart disease prediction.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: