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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Fetus-Specific Hematopoietic Stem Cell Dosimetry Framework for Leukemia-Relevant Target Cells During Prenatal Development.

Physics in medicine and biology·2026
Same author

Experimental measurement of k<sub>Q</sub> values for 150 MeV proton beams in the middle of SOBP using water calorimetry and comparison with TRS-398 rev.1.

Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB)·2026
Same author

Quantitative assessment of CBCT image quality metrics and their translation to clinical Art workflow performance in breast cancer patients.

Journal of applied clinical medical physics·2026
Same author

Technical evaluation of non-coplanar lattice radiotherapy: achieving directional VPDR uniformity with a 5-mm leaf width multi-leaf collimator.

Frontiers in oncology·2026
Same author

Assessment of Upper Airway Dimensions From Preadolescence to Young Adulthood in Correlation With Skeletal Maturation Stages.

The Journal of craniofacial surgery·2026
Same author

Effects of materials and beam currents on ozone generation in proton beam irradiation: a phantom study.

Journal of radiation research·2026
Same journal

Feasibility of uniportal thoracoscopic sublobar resection without chest tube drainage: a retrospective cohort study.

Frontiers in oncology·2026
Same journal

Real-world effectiveness and safety of carfilzomib, pomalidomide, and dexamethasone in relapsed/refractory multiple myeloma: a retrospective analysis from China.

Frontiers in oncology·2026
Same journal

Caregiver satisfaction with early integrated palliative care in oncology: secondary outcomes from the PALLiON cluster-RCT.

Frontiers in oncology·2026
Same journal

Intracranial mesenchymal tumor with FET::CREB fusion: a rare case report.

Frontiers in oncology·2026
Same journal

The multifaceted roles of mitochondria and their therapeutic transformation: a new perspective on triple-negative breast cancer treatment.

Frontiers in oncology·2026
Same journal

Trastuzumab emtansine versus trastuzumab plus pertuzumab for HER2-positive breast cancer with residual disease after neoadjuvant therapy: a real-world study.

Frontiers in oncology·2026
See all related articles

Related Experiment Video

Updated: Apr 14, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.7K

Performance enhancement method by using the probabilistic estimation for kidney tumor segmentation.

Wonjoong Cheon1, Meangee Kim2,3, Mira Han4

  • 1Department of Radiation Oncology, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.

Frontiers in Oncology
|April 13, 2026
PubMed
Summary
This summary is machine-generated.

The STAPLE algorithm significantly improved kidney tumor segmentation accuracy over individual models and other ensemble methods. This probabilistic approach enhances segmentation performance, particularly for medium-sized tumors.

Keywords:
STAPLE algorithmdeep learningensemble methodmedical image segmentationtumor segmentation

More Related Videos

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

290
Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
09:16

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis

Published on: June 18, 2020

7.4K

Related Experiment Videos

Last Updated: Apr 14, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.7K
Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

290
Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
09:16

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis

Published on: June 18, 2020

7.4K

Area of Science:

  • Medical Imaging Analysis
  • Artificial Intelligence in Medicine
  • Computational Pathology

Background:

  • Ensemble methods enhance medical image segmentation but require effective integration strategies.
  • Model diversity, achieved through varied loss functions, is crucial for ensemble performance.
  • The STAPLE algorithm offers a probabilistic framework for integrating diverse segmentation models.

Purpose of the Study:

  • To evaluate the STAPLE algorithm's effectiveness in improving kidney tumor segmentation accuracy.
  • To compare STAPLE's performance against individual models and conventional ensemble techniques like soft voting.
  • To assess the impact of leveraging model diversity from different loss functions within a probabilistic framework.

Main Methods:

  • Utilized CT scans from 210 patients (KiTS19 dataset) with expert annotations.
  • Developed five nnU-Net model variants (2D and 3D U-Nets) trained with hybrid loss functions (LCE+Dice, LTopK+Dice, LCE+GDice).
  • Compared five segmentation approaches: individual models, majority voting, soft voting, and STAPLE ensemble, using 5-fold cross-validation and evaluating DSC, JI, HD95, precision, and recall.

Main Results:

  • STAPLE achieved a higher Dice Similarity Coefficient (DSC) of 0.74 ± 0.23 compared to individual models (0.64-0.70) and soft voting (0.71 ± 0.26) (adjusted p<0.05).
  • STAPLE improved the Jaccard Index (JI) to 0.63 ± 0.24 and reduced Hausdorff Distance 95th percentile (HD95) to 11.81 ± 13.43.
  • The method demonstrated consistent superior performance on the LiTS17 liver tumor dataset, outperforming soft voting (DSC: 0.76 vs. 0.71).

Conclusions:

  • The STAPLE algorithm significantly enhances kidney tumor segmentation accuracy, outperforming individual models and conventional ensemble methods.
  • Probabilistic integration via STAPLE effectively leverages model diversity, particularly benefiting medium-sized tumors by reducing performance variability.
  • The demonstrated effectiveness across datasets suggests broad clinical applicability for STAPLE in medical image segmentation.