Scholarlink

<p>Lung cancer remains one of the deadliest malignancies worldwide, necessitating early and accurate diagnosis to improve patient outcomes. Advanced next generation sequencing (NGS) techniques have revolutionised the molecular analysis of lung cancer, providing unparalleled insights into its genetic makeup. The vast amount of data generated by NGS has created new opportunities for applying machine learning, particularly deep learning, to extract meaningful insights and make accurate predictions. This paper comprehensively reviews state-of-the-art deep learning approaches used in conjunction with NGS data to enhance lung cancer research. An in-depth analysis of cutting-edge deep learning techniques for NGS was conducted, drawing on prominent studies from the past 10 years, sourced using well-structured keywords in reputable databases such as PubMed, Scopus, and IEEE Xplore. The distinctive capabilities of convolutional neural networks (CNNs), graph neural networks (GNNs), and DeepVariant in analysing and interpreting next generation sequencing (NGS) data were rigorously evaluated. The assessment focused on the applications in identifying critical biomarkers, predicting cancer subtypes, and assessing therapeutic responses. Notably, CNNs demonstrated a precision of 99.6% for single nucleotide polymorphisms and 97.8% for insertions and deletions). Comprehensive tables summarise the methodologies, accuracy, and performance metrics of the models across diverse datasets. We found that CNNs are highly effective in extracting features from sequencing data, while GNNs excel at capturing the intricate relationships between genetic mutations. This study highlights the potential of deep learning models to revolutionise lung cancer detection and underscores the need for further research to optimise their clinical applications.</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

Deep Learning Approaches to Enhance Lung Cancer Diagnosis Using Next Generation Sequencing: State of the Art

  • 期刊论文

  • 发表日期：

    2025年9月9日

查看全文
  • Syed Naseer Ahmad Shah,
  • Javed Aalam,
  • Rafat Parveen

摘要

Lung cancer remains one of the deadliest malignancies worldwide, necessitating early and accurate diagnosis to improve patient outcomes. Advanced next generation sequencing (NGS) techniques have revolutionised the molecular analysis of lung cancer, providing unparalleled insights into its genetic makeup. The vast amount of data generated by NGS has created new opportunities for applying machine learning, particularly deep learning, to extract meaningful insights and make accurate predictions. This paper comprehensively reviews state-of-the-art deep learning approaches used in conjunction with NGS data to enhance lung cancer research. An in-depth analysis of cutting-edge deep learning techniques for NGS was conducted, drawing on prominent studies from the past 10 years, sourced using well-structured keywords in reputable databases such as PubMed, Scopus, and IEEE Xplore. The distinctive capabilities of convolutional neural networks (CNNs), graph neural networks (GNNs), and DeepVariant in analysing and interpreting next generation sequencing (NGS) data were rigorously evaluated. The assessment focused on the applications in identifying critical biomarkers, predicting cancer subtypes, and assessing therapeutic responses. Notably, CNNs demonstrated a precision of 99.6% for single nucleotide polymorphisms and 97.8% for insertions and deletions). Comprehensive tables summarise the methodologies, accuracy, and performance metrics of the models across diverse datasets. We found that CNNs are highly effective in extracting features from sequencing data, while GNNs excel at capturing the intricate relationships between genetic mutations. This study highlights the potential of deep learning models to revolutionise lung cancer detection and underscores the need for further research to optimise their clinical applications.