Linear regressive weighted Gaussian kernel liquid neural network for brain tumor disease prediction using time series data

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 18, 2025

A brain tumor is an abnormal growth of cells within the or surrounding tissues, which can be either benign malignant. Brain tumors develop in various regions brain, each affecting different functions such as movement, speech, and vision, depending on their location. Early prediction crucial for improving survival rates treatment outcomes. Advanced techniques, including medical imaging machine learning, are widely used early diagnosis. However, conventional learning deep detection models face challenges achieving high accuracy disease while minimizing time complexity. To address this, a novel Linear Regressive Weighted Gaussian Kernel Liquid Neural Network (LRWGKLNN) model developed. The proposed LRWGKLNN comprises four major steps, namely data acquisition, preprocessing, feature selection, classification. In initial step, large volume time-series samples collected from comprehensive dataset. Following collection, preprocessing performed, involving two key processes: handling missing outlier detection. First, handles values using linear regression method. After imputation process, identified removed Generalized Extreme Studentized Deviation test. Once complete, Cosine Congruence Majority Algorithm employed to select significant features dataset removing irrelevant features. This step helps minimize time. Finally, classification process performed selected with Kernelized Network. approach enhances samples. experimental evaluation carried out performance metrics accuracy, precision, recall, F1 score, respect number obtained results demonstrate that achieves higher 4%, 4% 5%, specificity score prediction. Furthermore, realizes substantial reduction consumption selection by 16% compared existing methods.

Language: Английский

---

Frequency-band specific directed connectivity networks reveal functional disruptions and pathogenic patterns in temporal lobe epilepsy: a MEG study

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: April 10, 2025

Language: Английский

---

Adaptive weighted median filtering for time-varying graph signals

Signal Image and Video Processing, Journal Year: 2024, Volume and Issue: 19(1)

Published: Dec. 7, 2024

Language: Английский

---

Dynamic Neural Network States During Social and Non-Social Cueing in Virtual Reality Working Memory Tasks: A Leading Eigenvector Dynamics Analysis Approach

Brain Sciences, Journal Year: 2024, Volume and Issue: 15(1), P. 4 - 4

Published: Dec. 24, 2024

This research investigates brain connectivity patterns in reaction to social and non-social stimuli within a virtual reality environment, emphasizing their impact on cognitive functions, specifically working memory. Employing the LEiDA framework with EEG data from 47 participants, I examined dynamic network states elicited by avatars compared stick cues during VR memory task. Through integration of deep learning graph theory analyses, unique associated cue type were discerned, underscoring substantial influence processes. LEiDA, conventionally utilized fMRI, was creatively employed detect swift alterations states, offering insights into processing dynamics. The findings indicate distinct neural for cues; notably, correlated state characterized increased self-referential memory-processing networks, implying greater engagement. Moreover, attained approximately 99% accuracy differentiating contexts, highlighting efficacy prominent eigenvectors analysis. Analysis also uncovered structural disparities, signifying enhanced contexts involving cues. multi-method approach elucidates cognition, establishing basis VR-based rehabilitation immersive learning, wherein signals may significantly enhance function.

Language: Английский

---