Artificial Intelligence and Radiogenomics for Pediatric CNS Neoplasms

Area of Science:

• Neuro-oncology
• Molecular Pathology
• Medical Imaging

Background:

• The 5th edition of the World Health Organization (WHO) Classification of CNS Tumors (2021) highlights the importance of molecular alterations.
• Pediatric central nervous system (CNS) tumors require precise classification for effective treatment.

Purpose of the Study:

• To review the integration of molecular profiling and advanced technologies in pediatric neuro-oncology.
• To discuss the impact of new diagnostic tools on tumor classification and targeted therapies.

Main Methods:

• Integration of histological data with molecular profiling techniques such as DNA methylation profiling and Next Generation Sequencing.
• Application of radiogenomics to correlate imaging features with genetic profiles.
• Utilizing artificial intelligence (AI), including machine learning and deep learning, for enhanced image analysis and molecular marker prediction.

Main Results:

• Refined classification of pediatric CNS tumors based on molecular subtypes.
• Enabled non-invasive tumor characterization through radiogenomics, reducing the need for risky biopsies in children.
• Improved diagnostic accuracy and prediction of molecular markers for personalized treatment strategies.

Conclusions:

• Molecular profiling and advanced imaging techniques are transforming pediatric neuro-oncology.
• AI and radiogenomics offer promising avenues for non-invasive diagnosis and personalized treatment of pediatric CNS tumors.
• Despite challenges, these technological advancements are poised to revolutionize the field.