Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Computed Tomography

Computed Tomography

Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...

Positron Emission Tomography

Positron Emission Tomography

Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body being...

You might also read

Related Articles

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

Sort by

Same author

Blood biomarkers of inflammation and brain-derived neurotrophic factor can not be used to assess the effectiveness of collagen mesotherapy in chronic cervical myofascial pain syndrome. A pilot single-blind randomized study.

Rheumatology international·2026

Same author

Consensus statement on mesotherapy for clinical and regulatory practice.

Rheumatology international·2026

Same author

Instability in the distal-to-stabilization segments after occipito-cervical fixation in patients with rheumatoid arthritis: the challenge of spine surgery.

Rheumatology international·2025

Same author

Comparative Evaluation of CNN and Transformer Architectures for Flowering Phase Classification of <i>Tilia cordata</i> Mill. with Automated Image Quality Filtering.

Sensors (Basel, Switzerland)·2025

Same author

Change in the angle of cervical lordosis after occipito-cervical fixation in the course of instability of C1-C2 in patients with rheumatoid arthritis.

Rheumatology international·2025

Same author

A Retrospective Analysis of the Effectiveness and Safety of Collagen Mesotherapy in the Course of Chronic Cervical Myofascial Pain Syndrome.

Biomedicines·2025

Search research articles

Related Experiment Video

Updated: Jun 20, 2026

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

Zero-Shot Vertebral Instance Segmentation on DICOM Spine Radiographs Using Promptable Segment Anything Models.

Alexander Sieradzki¹, Kamil Koszela², Szymon Koszykowski¹

¹Department of Artificial Intelligence, Institute of Information Technology, Warsaw University of Life Sciences, ul. Nowoursynowska 159, 02-776 Warsaw, Poland.

Journal of Clinical Medicine

|March 14, 2026

Summary

This summary is machine-generated.

Promptable foundation models can perform zero-shot vertebral instance segmentation on spine radiographs. Rectangle prompts enable accurate segmentation without task-specific training, outperforming point prompts.

Keywords:

DICOM radiographs MedSAM SAM2 Segment Anything Model foundation models medical image analysis promptable segmentation vertebral instance segmentation zero-shot segmentation

More Related Videos

Author Spotlight: Improving Percutaneous Vertebroplasty Surgical Accuracy and Efficiency Through Advanced Puncture Techniques

Author Spotlight: Improving Percutaneous Vertebroplasty Surgical Accuracy and Efficiency Through Advanced Puncture Techniques

Published on: August 9, 2024

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

Related Experiment Videos

Last Updated: Jun 20, 2026

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

Author Spotlight: Improving Percutaneous Vertebroplasty Surgical Accuracy and Efficiency Through Advanced Puncture Techniques

Author Spotlight: Improving Percutaneous Vertebroplasty Surgical Accuracy and Efficiency Through Advanced Puncture Techniques

Published on: August 9, 2024

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

Area of Science:

Medical Imaging
Artificial Intelligence
Computer Vision

Background:

Accurate vertebral instance segmentation is crucial for spinal parameter assessment.
Supervised methods for segmentation demand extensive, costly annotations and can struggle with domain shifts.
Exploring automated segmentation methods is vital for efficient clinical workflows.

Purpose of the Study:

To evaluate the zero-shot generalization capabilities of promptable segmentation foundation models on raw DICOM full-spine radiographs.
To compare the performance of different models (SAM-ViT-Huge, SAM2-Hiera-Large, MedSAM-ViT-Base) and prompting strategies (point vs. rectangle).
To determine the feasibility of using these models without task-specific fine-tuning.

Main Methods:

Evaluated SAM-ViT-Huge, SAM2-Hiera-Large, and MedSAM-ViT-Base on 144 full-spine radiographs with 1309 vertebral masks.
Utilized a standardized pipeline for DICOM decoding, intensity normalization, and automatic prompt generation.
Assessed performance using oracle and model-score protocols, measuring IoU, Dice, precision, recall, ASSD, and HD95.

Main Results:

SAM-ViT-Huge with rectangle prompting achieved the highest mean IoU/Dice (0.782/0.870 oracle; 0.737/0.837 model-score).
SAM2-Hiera-Large also performed well with rectangle prompts, outperforming MedSAM-ViT-Base.
Point prompting resulted in significantly lower overlap and higher boundary errors, despite high recall.

Conclusions:

Zero-shot vertebral instance segmentation on raw DICOM spine radiographs is achievable using promptable foundation models.
Rectangle prompting is demonstrably more effective than point prompting for this task.
Foundation models show promise for reducing annotation burden in spinal imaging analysis.