Cardiac signal classification leveraging spectral optimization with ChebWave and deep blue particle filtering DOI

Anu Honnashamaiah,

Rathnakara Srinivasapandit

The International Journal of Artificial Organs, Год журнала: 2025, Номер unknown

Опубликована: Май 30, 2025

Electrocardiogram (ECG) signal classification plays a critical role in diagnosing various cardiac conditions by identifying irregularities heart rhythms. Despite advancements the field, existing methodologies often rely on basic techniques that inadequately filter noise, leading to degraded performance and misinterpretation of vital features. This study presents Spectral-Optimized Cardiac Framework (SOCF) approach enhance accuracy ECG through advanced noise filtering, comprehensive feature extraction, efficient selection integration hybrid modelling techniques. The proposed methodology introduces ChebWave Mean Refinement Filter (CWMRF) for effective reduction clarity while preserving essential characteristics. In Spectral Essence Extractor (SEE) captures both high order features, providing deeper insights into signals. Additionally, Deep Blue Particle Optimizer (DBPO) efficiently identify relevant features mitigating risk overfitting. Furthermore, architecture Convolutional neural network (CNN) long short-term memory (LSTM) enable model effectively capture spatial temporal dependencies, thereby improving accuracy. To optimize performance, Aquila enhances convergence speed efficiency employing diverse search strategies inspired hunting behavior bird. By integrating these techniques, SOCF achieved impressive results MIT-BIH dataset PTB with an 99.6% 99.68%, precision 99.4% 99.44%, recall 99.5% 99.51%, F1 score 99.2% 99.49%, which significantly improves robustness reliability classification, ultimately more accurate clinical better patient outcomes.

Язык: Английский

Optimized Transformer- Basde Security for Vehicular Network Communication Against Denial- of - Service Attack DOI

Ramani Gaddam,

Amarendra Kothalanka

Salud Ciencia y Tecnología - Serie de Conferencias, Год журнала: 2025, Номер 4, С. 1424 - 1424

Опубликована: Янв. 21, 2025

Objective: A Vehicular Ad-Hoc Network (VANET) is one of the crucial elements an Intelligent Transport System (ITS) and plays a significant role in security communication. VANETs are susceptible to Denial Service (DoS) attacks, which inherent threat performance such networks, requiring more sophisticated detection countermeasures. Methods: In response this problem, Spatial Hyena Security Transformer Model (SHSTM) introduced improve use Ad-hoc communication against DoS attacks. The network nodes set up enable Vehicle-to-Vehicle (V2V) communication; SHSTM constantly detects each node detect filter out attack targets. model includes effective Cluster Head (CH) selection approach based on traffic patterns enhance security.Results: Comparative measurements conducted positions before after attacks show enhanced overall terms Packet Delivery Ratio (PDR), Throughput (NT), Energy Consumption (EC), End-to-End Delay (EED), Attack Detection (ADR). attains NT 3.91 Mbps, minimal EC 1.02 mJ, highest PDR 99.04%, EED 0.0206 seconds, higher ADR 98%. Conclusions: design proposed proved improvement performance, outperforms existing state-of-the-art technique. Hence, it considered potential solution address VANET.

Язык: Английский

Процитировано

0

Cardiac signal classification leveraging spectral optimization with ChebWave and deep blue particle filtering DOI

Anu Honnashamaiah,

Rathnakara Srinivasapandit

The International Journal of Artificial Organs, Год журнала: 2025, Номер unknown

Опубликована: Май 30, 2025

Electrocardiogram (ECG) signal classification plays a critical role in diagnosing various cardiac conditions by identifying irregularities heart rhythms. Despite advancements the field, existing methodologies often rely on basic techniques that inadequately filter noise, leading to degraded performance and misinterpretation of vital features. This study presents Spectral-Optimized Cardiac Framework (SOCF) approach enhance accuracy ECG through advanced noise filtering, comprehensive feature extraction, efficient selection integration hybrid modelling techniques. The proposed methodology introduces ChebWave Mean Refinement Filter (CWMRF) for effective reduction clarity while preserving essential characteristics. In Spectral Essence Extractor (SEE) captures both high order features, providing deeper insights into signals. Additionally, Deep Blue Particle Optimizer (DBPO) efficiently identify relevant features mitigating risk overfitting. Furthermore, architecture Convolutional neural network (CNN) long short-term memory (LSTM) enable model effectively capture spatial temporal dependencies, thereby improving accuracy. To optimize performance, Aquila enhances convergence speed efficiency employing diverse search strategies inspired hunting behavior bird. By integrating these techniques, SOCF achieved impressive results MIT-BIH dataset PTB with an 99.6% 99.68%, precision 99.4% 99.44%, recall 99.5% 99.51%, F1 score 99.2% 99.49%, which significantly improves robustness reliability classification, ultimately more accurate clinical better patient outcomes.

Язык: Английский

Процитировано

0