Scholarlink

<p>Creutzfeldt-Jakob disease (CJD) is a spongiform encephalopathy caused by misfolded human prion proteins (PrP)s. Due to variability in presentation, the diagnosis may be missed in lieu of various psychiatric disorders. In this context, a fusion of data augmentation, feature selection along with robust multimodal deep learning (DL) architecture is presented to analyses visual inputs to increase the detection accuracy of CJD. In this study, a novel hybrid framework is proposed that leverages Generative Adversarial Networks (GANs) for data augmentation, Brood Parasitism-Based Optimization (BPBO) for feature selection, and a hybrid LSTM-BiGRU (long short term memory- gated recurrent unit) model for robust classification. The stacked model is trained on extracted features from multi-modal images, effectively distinguishing between different CJD severity levels. The performance of the proposed framework demonstrates that the GAN-enhanced dataset with improved feature space enhances classification accuracy of the LSTM-BiGRU network. Various performance metrics, including accuracy, Cohen kappa score and Jaccard Score, reported in the result section indicates that the proposed method outperforms conventional feature selection and classification techniques.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

BPBO-LSTM-BiGRU: generative adversial network with brood parasitism-based optimization for diagnosis of Creutzfeldt-Jakob disease using multiple visual modalities

  • 期刊论文

  • 发表日期：

    2025年9月26日

查看全文
  • P. Jaiendra Reddy,
  • K. V. Panduranga Rao,
  • P. Ashok Reddy

摘要

Creutzfeldt-Jakob disease (CJD) is a spongiform encephalopathy caused by misfolded human prion proteins (PrP)s. Due to variability in presentation, the diagnosis may be missed in lieu of various psychiatric disorders. In this context, a fusion of data augmentation, feature selection along with robust multimodal deep learning (DL) architecture is presented to analyses visual inputs to increase the detection accuracy of CJD. In this study, a novel hybrid framework is proposed that leverages Generative Adversarial Networks (GANs) for data augmentation, Brood Parasitism-Based Optimization (BPBO) for feature selection, and a hybrid LSTM-BiGRU (long short term memory- gated recurrent unit) model for robust classification. The stacked model is trained on extracted features from multi-modal images, effectively distinguishing between different CJD severity levels. The performance of the proposed framework demonstrates that the GAN-enhanced dataset with improved feature space enhances classification accuracy of the LSTM-BiGRU network. Various performance metrics, including accuracy, Cohen kappa score and Jaccard Score, reported in the result section indicates that the proposed method outperforms conventional feature selection and classification techniques.