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

POT-PISA: core-degradable particles <i>via</i> aqueous radical ring-opening polymerisation-induced self-assembly using 7-phenyloxepane-2-thione.

Chemical communications (Cambridge, England)·2026
Same author

Degradable Vinyl-Based Polymers by Radical Ring-Opening Polymerization: A User Guide.

ACS polymers Au·2026
Same author

An integrated restricted mean survival time-based evaluation of efficacy, toxicity, and cost in first-line immunotherapy regimens for metastatic renal cell carcinoma.

Urologic oncology·2026
Same author

Model-Driven Optimization of Subcutaneous Polymer Prodrugs Achieves Cancer Remission in Mice.

Advanced healthcare materials·2026
Same author

Role of Spatial Heterogeneity in Muscle-Invasive Bladder Cancer on Overall Survival and Immunotherapy Response.

Cancers·2026
Same author

Towards safer medicines: management of residual solvents and green alternatives.

Journal of controlled release : official journal of the Controlled Release Society·2026

Related Experiment Video

Updated: Jun 21, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K

Deep Learning Algorithms for Bladder Cancer Segmentation on Multi-Parametric MRI.

Kazim Z Gumus1, Julien Nicolas2, Dheeraj R Gopireddy1

  • 1Department of Radiology, College of Medicine-Jacksonville, University of Florida, Jacksonville, FL 32209, USA.

Cancers
|July 13, 2024
PubMed
Summary
This summary is machine-generated.

Deep learning models MAnet and PSPnet show improved bladder cancer segmentation on MRI scans. Using a combined cross-entropy and dice similarity coefficient loss function enhances performance across various imaging sequences.

Keywords:
MAnetMRIPSPnetUnetbladder cancercross-entropydeep learningexpected calibration errorfocal lossloss functionsegmentation

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

385
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.3K

Related Experiment Videos

Last Updated: Jun 21, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

385
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.3K

Area of Science:

  • Radiology
  • Artificial Intelligence
  • Oncology

Background:

  • Accurate bladder cancer (BC) segmentation on magnetic resonance imaging (MRI) is crucial for assessing muscular invasion.
  • Multi-parametric MRI (mp-MRI) provides detailed anatomical and functional information for BC detection.

Purpose of the Study:

  • To evaluate the tumor segmentation performance of three deep learning (DL) models: Unet, MAnet, and PSPnet.
  • To compare the effectiveness of different loss functions (cross-entropy, dice similarity coefficient loss, focal loss) on mp-MRI data.

Main Methods:

  • Segmentation of bladder tumors was performed on T2-weighted (T2WI), diffusion-weighted imaging/apparent diffusion coefficient (DWI/ADC), and contrast-enhanced T1-weighted (T1WI) images from 53 patients.
  • Three DL models (Unet, MAnet, PSPnet) were trained using hybrid loss functions (CE+DSC, FL).
  • Performance was assessed using Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), and Expected Calibration Error (ECE).

Main Results:

  • MAnet with CE+DSC achieved the highest DSC values across ADC, T2WI, and T1WI images.
  • PSPnet with CE+DSC yielded the smallest HD values on ADC, T2WI, and T1WI.
  • Segmentation accuracy was superior on ADC and T1WI compared to T2WI; PSPnet with FL showed the lowest ECE on ADC, while MAnet with CE+DSC had the lowest ECE on T2WI and T1WI.

Conclusions:

  • MAnet and PSPnet, particularly with a hybrid CE+DSC loss function, demonstrate superior performance in bladder cancer segmentation compared to Unet.
  • The choice of evaluation metric influences the optimal DL model and loss function for BC segmentation on mp-MRI.