Evaluating the Diagnostic Utility of Advanced CT Radiomics and Deep Learning for Distinguishing Pediatric Peripheral Neuroblastoma from Ganglioneuroblastoma

1

Neuroblastoma is the most common extracranial solid tumor in children, often diagnosed before age five, necessitating accurate histological subtyping for treatment.
2

Differentiating between malignant neuroblastoma and benign ganglioneuroblastoma is challenging due to similar clinical presentations and imaging characteristics.
3

Conventional contrast-enhanced CT has limited effectiveness in distinguishing neuroblastoma from ganglioneuroblastoma, highlighting the need for advanced diagnostic methods.
4

This study aims to develop radiomic and deep learning models using CT, clinical data, and biochemical indicators to improve differential diagnosis in pediatric patients.
5

A total of 225 pediatric patients were included in the study, with data divided into training and validation sets to enhance the accuracy of the diagnostic models.

1