• The Integration of Artificial Intelligence in Radiomics and Radiogenomics for Lung Cancer: A Comprehensive Review
  • Fatemeh Mazaheri,1,* Amirreza SadeghiNasab,2 Mahmoud Mohammadi-Sadr,3 Marziyeh Tahmasb,4
    1. Medical Physics & Biomedical Engineering Department, Tehran University of Medical Sciences (TUMS) Tehran, Iran.
    2. 1Department of Radiologic Technology, School of Allied Medical Sciences, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran. 2 Student Research Committee, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran.
    3. Department of Medical Physics, School of Medicine, Isfahan University of Medical Sciences, Isfahan, Iran.
    4. Department of Radiologic Technology, School of Allied Medical Sciences, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran.


  • Introduction: Lung cancer, which originates in the lung tissue, is the leading cause of cancer-related deaths, posing a significant global public health challenge. Cancer imaging has traditionally played a crucial role in diagnosing, staging, and monitoring disease. The advent of quantitative methods for evaluating medical images has introduced radiomics, the approach that evaluates image biomarkers to provide deeper insights into disease characteristics. The integration of conventional imaging techniques with molecular features at the genomic, transcriptomic, and proteomic levels, known as radiogenomics, also aims to uncover the biological foundations of imaging phenotypes. In recent years, artificial intelligence (AI) technology has introduced data-driven analysis models that have significantly advanced information-processing techniques in the radiomics and radiogenomics of cancer. This review aims to investigate the advantages of integrating AI models into radiomics and radiogenomics approaches for lung cancer detection and treatment outcomes.
  • Methods: Utilizing different combinations of keywords "Artificial Intelligence", "Deep Learning", "Radiomic", "Radiogenomic", "lung Cancer", "Detection" and "Diagnosis", PubMed, Science Direct, Web of Science, and Google Scholar databases were explored until July 2024, Ultimately, 20 recent and relevant records were reviewed.
  • Results: Based on the results of the reviewed papers, the radiomic and genomic data has enabled personalized treatment planning, reducing the need for biopsies and improving patient outcomes, representing a major leap forward in lung cancer management. Moreover, AI methods such as deep learning techniques, particularly convolutional neural networks (CNNs), have enhanced lung cancer diagnostic accuracy of non-invasive detection methods such as computed tomography (CT) and positron emission tomography (PET) modalities by improving image feature extraction and analysis. AI has also strengthened the correlation between radiomic features and genetic mutations, accurately predicting mutations such as Epidermal Growth Factor Receptor (EGFR), Kirsten Rat Sarcoma Viral Oncogene Homolog (KRAS), and Anaplastic Lymphoma Kinase (ALK), and tumor recurrence.
  • Conclusion: The integration of AI in radiomics and radiogenomics has shown significant promise in revolutionizing lung cancer diagnosis and treatment. AI-driven models provide detailed insights that are critical for personalized medicine, potentially leading to improved patient outcomes. However, for these technologies to be fully integrated into clinical practice, interdisciplinary collaboration, data standardization, and addressing ethical considerations are essential. Continued research and development in this field are imperative to overcome existing challenges and to utilize the full potential of AI in lung cancer management.
  • Keywords: Artificial Intelligence, Deep Learning, Radiomic, Radiogenomic, Lung Cancer.