Related Experiment Video
Updated: Oct 16, 2025
15:18
Making MR Imaging Child's Play - Pediatric Neuroimaging Protocol, Guidelines and Procedure
Published on: July 30, 2009
18.4K
Artificial Intelligence Applications in Pediatric Brain Tumor Imaging: A Systematic Review
Jonathan Huang1, Nathan A Shlobin1, Sandi K Lam1
1Department of Neurological Surgery, Northwestern University Feinberg School of Medicine, Division of Pediatric Neurosurgery, Ann and Robert H. Lurie Children's Hospital, Chicago, Illinois, USA.
World Neurosurgery
|October 14, 2021
Summary
Artificial intelligence (AI) shows promise in analyzing pediatric brain tumor imaging, often outperforming clinical experts in diagnosis. However, AI tools are not yet integrated into routine clinical practice.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Pediatric Oncology
Background:
- Recent advancements in computational capacity and data availability have propelled AI applications in medical imaging.
- While AI shows potential in brain tumor imaging, its clinical impact requires further investigation.
- This systematic review focuses on AI's role in analyzing pediatric brain tumor imaging.
Purpose of the Study:
- To systematically review the role of artificial intelligence (AI) in the analysis of pediatric brain tumor imaging.
- To assess the current applications and performance of AI in pediatric neuro-oncology imaging.
- To identify gaps in research and clinical implementation of AI for pediatric brain tumors.
Main Methods:
- A systematic literature search was conducted across PubMed, Embase, and Scopus databases.
- The search included studies published up to January 27, 2021.
- 22 studies were included from an initial pool of 298 records.
Main Results:
- Posterior fossa tumors were most frequently studied (68%), including brainstem glioma, medulloblastoma, and pilocytic astrocytoma.
- AI was most commonly applied to tumor diagnosis (64%), followed by segmentation (14%) and detection (14%).
- AI demonstrated superiority over clinical experts in tumor diagnosis in 5 out of 6 comparative studies, with comparable accuracy in other tasks like segmentation and radiotherapy planning.
Conclusions:
- AI methods for pediatric brain tumor imaging analysis are rapidly advancing.
- Clinical adoption requires further validation of AI's utility and validity.
- AI implementation could enhance diagnostic accuracy and automate imaging tasks, potentially streamlining clinical workflows.
