Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{183288,
        author = {Jananee J and Dr. F. Emerson Solomon},
        title = {TEMPORAL CONVOLUTIONAL NETWORK WITH CWT FEATURES FOR ACCURATE SCHIZOPHRENIA CLASSIFICATION},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {3},
        pages = {960-967},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=183288},
        abstract = {This research introduces a high-performance DL framework for predicting SZ(Schizophrenia) using EEG signals by seamlessly integrating comprehensive preprocessing, sophisticated feature extraction methods, and advanced temporal classification. Utilizing the openly accessible Kaggle dataset on mental attention states—comprising non-invasive EEG recordings—the study begins with data refinement through ICA, which effectively eliminates non-neural disturbances such as ocular movements and muscular noise. Subsequently, the cleaned EEG signals are subjected to feature extraction via three approaches: Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT), and CWT. Among these, CWT proves most effective, offering superior time-frequency resolution for analyzing the inherently non-stationary nature of EEG data. The features obtained are then processed by a TCN, chosen for its strong capability to capture long-term temporal patterns within the data. Empirical evaluations affirm that the integration of ICA-based noise reduction, CWT-driven feature extraction, and TCN classification yields the best predictive outcomes, achieving top-tier performance across accuracy, precision, and recall metrics. This approach significantly outperforms configurations involving FFT or STFT with TCN, underlining the effectiveness of combining multi-resolution signal analysis with temporal DL in enhancing the reliability of EEG-based SZ diagnosis.},
        keywords = {EEG Signal Processing. Schizophrenia Prediction, Independent Component Analysis (ICA), Continuous Wavelet Transform (CWT), Temporal Convolutional Network (TCN), Feature Extraction, Deep Learning (DL), Brain Signal Classification, Time-Frequency Analysis, Signal Preprocessing},
        month = {August},
        }