Integrating Omics Data and AI for Cancer Diagnosis and Prognosis

Area of Science:

• Oncology
• Medical Informatics
• Artificial Intelligence

Background:

• Cancer remains a leading cause of mortality, underscoring the need for accurate and timely diagnosis and prognosis.
• Artificial intelligence (AI) offers powerful capabilities for organizing and analyzing complex patient data to improve cancer care.
• The clinical utility of AI in oncology is rapidly evolving, necessitating a comprehensive review of recent advancements.

Purpose of the Study:

• To review the diverse applications of artificial intelligence (AI) in cancer diagnosis and prognosis.
• To assess the clinical utility of AI tools utilizing various data types in oncology.
• To synthesize findings from recent research on AI in cancer care.

Main Methods:

• A systematic literature search was conducted on PubMed and EBSCO databases for publications from January 1, 2020, to December 22, 2023.
• Key search terms included "artificial intelligence" and "machine learning" to identify relevant studies.
• Eighty-nine studies were included, categorized by data type (multi-omics, radiomics, pathomics, clinical, laboratory) and focus (diagnosis, prognosis).

Main Results:

• AI applications in cancer diagnosis and prognosis were analyzed across 89 studies, categorized by data modalities.
• Studies utilized multi-omics data (genomics, transcriptomics, epigenomics, proteomics), radiomics, pathomics, and clinical/laboratory data.
• Eight studies integrated multiple omics data types, highlighting the potential of multi-modal data analysis with AI.

Conclusions:

• Integrating AI with omics and clinical data represents a significant advancement in cancer diagnosis and prognosis.
• AI demonstrates considerable potential to enhance the accuracy and efficiency of cancer care.
• Ongoing prospective studies are essential to improve AI algorithm interpretability and ensure safe clinical integration for patient benefit.