A deep learning-based radiomic nomogram derived from visceral fat for early prediction of gastrointestinal stromal tumor risk grade - Takeaways - MDSpire

A deep learning-based radiomic nomogram derived from visceral fat for early prediction of gastrointestinal stromal tumor risk grade

  • By

  • Wei Chen

  • Long-Yu Duan

  • Kun-Ming Yi

  • Xiao-Juan Peng

  • Lian-Qin Kuang

  • June 19, 2026

  • 0 min

Share

  • 1

    The study developed a deep-learning-based radiomics nomogram (DLRN) for preoperative risk stratification of gastrointestinal stromal tumors (GISTs).

  • 2

    The DLRN utilizes visceral fat features extracted from non-contrast CT scans, making it applicable for patients unable to undergo contrast-enhanced imaging.

  • 3

    In the derivation cohort, the DLRN achieved an AUC of 0.936, with accuracy, sensitivity, and specificity values of 0.873, 0.811, and 0.904, respectively.

  • 4

    The external test cohort showed the DLRN's AUC at 0.862, with an accuracy of 0.925, sensitivity of 0.833, and specificity of 0.936.

  • 5

    The DLRN demonstrated a higher net clinical benefit compared to traditional models, but further validation in larger cohorts is necessary.

Original Source(s)

A deep learning-based radiomic nomogram derived from visceral fat for early prediction of gastrointestinal stromal tumor risk grade

  • by Wei Chen, Long-Yu Duan, Kun-Ming Yi, Xiao-Juan Peng, Lian-Qin Kuang

  • June 19, 2026

Related Content

Correction: Stereotactic radiotherapy for brain metastases: indications, dose fractionation, technological innovations, and evolving combination strategies – a comprehensive review

Ipilimumab and nivolumab for cancer treatment: a pharmacovigilance study based on the FDA adverse event reporting system database

Editorial: Innovative strategies in overcoming glioblastoma: advancements in treatment and research