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 Video

Updated: Jan 10, 2026

Cell-based Assay Protocol for the Prognostic Prediction of Idiopathic Scoliosis Using Cellular Dielectric Spectroscopy
08:08

Cell-based Assay Protocol for the Prognostic Prediction of Idiopathic Scoliosis Using Cellular Dielectric Spectroscopy

Published on: October 16, 2013

11.0K

Predicting Nonidiopathic Scoliosis from Plain Radiographs: A Deep-Learning Approach.

Kellen L Mulford1, Hans K Nugraha2, Julia E Todderud2

  • 1Orthopedic Surgery Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota.

JB & JS Open Access
|November 24, 2025
PubMed
Summary
This summary is machine-generated.

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

Predictors of Interbody Fusion and Adjacent Segment Disease Following Anterior Lumbar Interbody Fusion for Degenerative Pathologies.

Journal of clinical medicine·2026
Same author

Surgical Spine Techniques for Small-Statured Surgeons and Assistants.

The Journal of the American Academy of Orthopaedic Surgeons·2026
Same author

Lumbar, thoracic, and bilateral vertebral body tethering: do surgical outcomes differ?

Spine deformity·2026
Same author

Recurrence of Ankle Deformity Following Reconstruction in Patients With Fibular Hemimelia and Dynamic Valgus.

Journal of pediatric orthopedics·2026
Same author

Comparison of Large Language Models with Rules-Based Natural Language Processing Algorithms for Extracting Data from Operative Notes.

The Journal of bone and joint surgery. American volume·2026
Same author

Global Perspectives on Artificial Intelligence in Orthopaedic Surgery: Findings from the 2025 ABC Travelling Fellowship Survey.

JB & JS open access·2026
Same journal

The Effect of Bearing Surface On the Survival of Cementless and Hybrid Total Hip Arthroplasty: An Analysis of Linked National Joint Registry and Hospital Episode Statistics Data.

JB & JS open access·2026
Same journal

Should We Reconstruct Coronal Plane Fractures in Multifragmentary Pertrochanteric Fractures? A Randomized Controlled Trial.

JB & JS open access·2026
Same journal

Timing of Return to Driving After Volar Locking-Plate Fixation of Distal Radius Fractures Without Wrist Immobilization: A Prospective Driving Simulator Study.

JB & JS open access·2026
Same journal

Early Brace Treatment: The New Standard for Adolescent Idiopathic Scoliosis?

JB & JS open access·2026
Same journal

Prevalence and Implications of Incidental Findings on Computed Tomography Scans Performed for Mako Robotic Arm-Assisted Joint Arthroplasty.

JB & JS open access·2026
Same journal

Patient-Specific Medial Patellofemoral Ligament Femoral Footprint Mapping Using Compressed Lateral and Sagittal Sequence MRI in Patients With Trochlear Dysplasia.

JB & JS open access·2026
See all related articles

A deep learning (DL) model accurately classifies pediatric scoliosis etiology from radiographs, outperforming experienced surgeons. This AI tool aids in identifying the need for advanced imaging, improving diagnostic decisions for spinal cord pathology.

Area of Science:

  • Artificial Intelligence in Medicine
  • Medical Imaging Analysis
  • Pediatric Orthopedics

Background:

  • Scoliosis presents diverse etiologies, including idiopathic, congenital, and spinal cord abnormalities.
  • Advanced imaging for scoliosis diagnosis is often limited by cost, feasibility, and risks in pediatric patients.
  • Artificial intelligence (AI) offers potential for improved radiographic classification of scoliosis etiology.

Purpose of the Study:

  • To develop and evaluate a deep learning (DL) based image classifier for pediatric scoliosis etiology.
  • To compare the diagnostic performance of the DL classifier against experienced spine surgeons using spinal radiographs.
  • To assess the potential of AI in improving the diagnostic accuracy and efficiency of scoliosis classification.

Main Methods:

Related Experiment Videos

Last Updated: Jan 10, 2026

Cell-based Assay Protocol for the Prognostic Prediction of Idiopathic Scoliosis Using Cellular Dielectric Spectroscopy
08:08

Cell-based Assay Protocol for the Prognostic Prediction of Idiopathic Scoliosis Using Cellular Dielectric Spectroscopy

Published on: October 16, 2013

11.0K
  • A dataset of 1036 pediatric patients with scoliosis and paired radiographs was utilized.
  • Patients were manually classified into idiopathic, congenital, or spinal cord pathology categories.
  • A DL classifier (EfficientNet B4) was trained on randomized radiographs for classification and performance evaluation using precision, recall, and F1-score.

Main Results:

  • The DL classifier achieved a high F1-Score of 0.97, demonstrating excellent performance in identifying scoliosis etiologies.
  • The model exhibited superior overall precision (0.96), recall (0.96), and F1-score (0.96) compared to experienced spine surgeons.
  • Surgeons' accuracies were lower (F1-scores of 0.79 and 0.71), with no clear pattern in their diagnostic errors, particularly for spinal cord pathology.

Conclusions:

  • A DL-convolutional-neural-network classifier accurately distinguishes between three main scoliosis etiologies on pediatric spine radiographs.
  • The AI model demonstrated superior diagnostic performance compared to experienced spine surgeons.
  • This AI tool can assist clinicians in determining the need for further axial imaging, potentially aiding in the early detection of spinal cord pathology.