Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Interpretable Deep Learning Radiomics Model for Preoperative Prediction of High-Grade Soft Tissue Sarcomas: A

Miaomiao Yang1,2, Jiyang Jin2, Rui Wen3

  • 1Department of Medical Imaging, East Hospital, Tongji University, 150 Jimo Road, Pudong New Area, Shanghai, 200120, China.

Journal of Imaging Informatics in Medicine
|May 27, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Frailty and prefrailty are associated with increased risk of aortic aneurysm but not aortic dissection: a prospective cohort study of UK biobank participants.

BMC cardiovascular disorders·2026
Same author

Rewiring of de novo serine synthesis supports tumorigenesis under deficiency of TET2.

British journal of cancer·2026
Same author

Comparative Efficacy of Laser Acupuncture and Direct Photobiomodulation for Inferior Alveolar Nerve Hypoesthesia After Mandibular Third Molar Extraction: A Randomized, Double-Blind, Sham-Controlled Clinical Trial.

Photobiomodulation, photomedicine, and laser surgery·2026
Same author

Association of Triglyceride-Glucose-Frailty Index with Cardiovascular Disease and All-Cause Mortality Incidence in Individuals with Cardiovascular-Kidney-Metabolic Syndrome Stages 0-3: A Nationwide Prospective Cohort Study.

Journal of clinical medicine·2026
Same author

Case Report: ERAT-based management of appendiceal abscess in early pregnancy.

Frontiers in surgery·2026
Same author

Peroxymonosulfate activation by NiCoFe-layered double hydroxides for enhancing contaminant elimination in water.

Journal of environmental sciences (China)·2026
Same journal

Kolmogorov-Arnold Guided Local-Global Attention for Medical Image Classification.

Journal of imaging informatics in medicine·2026
Same journal

Artificial Intelligence-Assisted Inner Ear Computed Tomography Analysis: Radiomics-Based Comparison of Affected and Unaffected Ears in Idiopathic Sudden Sensorineural Hearing Loss.

Journal of imaging informatics in medicine·2026
Same journal

High Adoption, Higher Expectations: A Cross-Sectional Survey of Radiologist Engagement with Artificial Intelligence in the United Arab Emirates.

Journal of imaging informatics in medicine·2026
Same journal

Complex-valued Multi-scale Hybrid Attention Network for Fast MRI via Sparsified Data Learning.

Journal of imaging informatics in medicine·2026
Same journal

Automatic Phase and Sequence Identification in Gd-EOB-DTPA-Enhanced Liver MRI Using Deep Convolutional and Sequential Learning.

Journal of imaging informatics in medicine·2026
Same journal

Ultrasound-Based AI in Predicting Hormone Receptor Status in Breast Cancer: Is "Digital Biopsy" Possible.

Journal of imaging informatics in medicine·2026
See all related articles
This summary is machine-generated.

A new deep learning radiomics model accurately predicts high-grade soft tissue sarcomas (STSs) before surgery. This AI approach analyzes MRI scans to identify aggressive tumors, improving preoperative diagnosis and patient care.

Area of Science:

  • Radiology and Medical Imaging
  • Artificial Intelligence in Medicine
  • Oncology

Background:

  • Accurate preoperative grading of soft tissue sarcomas (STSs) is crucial for treatment planning.
  • Distinguishing high-grade STSs from low-grade or benign lesions preoperatively remains challenging with conventional imaging.
  • Deep learning radiomics offers a novel approach to extract complex imaging features for improved diagnostic accuracy.

Purpose of the Study:

  • To develop and validate a preoperative deep learning radiomics (DLR) model for predicting high-grade soft tissue sarcomas (STSs).
  • To evaluate the predictive performance and interpretability of the DLR model using external validation data.
  • To assess the contribution of intratumoral and peritumoral regions in predicting high-grade STSs.

Main Methods:

Keywords:
Deep learningHistological gradeModel interpretabilityRadiomicsSoft tissue sarcomas

Related Experiment Videos

  • Development of a DLR model using fat-suppressed T2-weighted imaging (FS-T2WI) from 129 patients (training set) and 74 patients (external validation set).
  • Extraction of radiomic features from manually segmented intratumoral and peritumoral regions of interest (ROIs).
  • Integration of deep learning features from whole-image FS-T2WI using a 3D ResNet18 model, followed by feature selection (RFE, LASSO) and model construction.

Main Results:

  • The DLR model achieved a high predictive performance in the external validation set with an area under the ROC curve (AUC) of 0.939 and an F1-score of 0.89.
  • No significant performance differences were observed across subgroups stratified by age, sex, or tumor site.
  • Interpretability analyses (Grad-CAM, SHAP) highlighted the importance of peritumoral features and tumor heterogeneity in predicting high-grade STSs.

Conclusions:

  • The developed interpretable DLR model demonstrates high accuracy and reliability for preoperative prediction of high-grade STSs.
  • This AI-driven approach offers a clinically applicable tool to enhance preoperative diagnosis and guide treatment strategies for soft tissue sarcomas.
  • The study underscores the value of integrating deep learning with radiomics for complex medical image analysis in oncology.