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

Search research articles

Related Experiment Video

Updated: Jun 29, 2026

A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound

A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound

Published on: March 21, 2025

GUSL: A novel and efficient machine learning model for prostate segmentation on MRI.

Jiaxin Yang¹, Vasileios Magoulianitis², Catherine Aurelia Christie Alexander¹

¹Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California (USC), 3740 McClintock Ave., Los Angeles, CA 90089, USA.

Computers in Biology and Medicine

|June 27, 2026

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

Readability of sexual health educational materials: a comparative analysis of ISSM resources and large language model-generated content.

The journal of sexual medicine·2026

Same author

Enhancing the quality and trustworthiness of large language model-generated summaries of clinical oncology literature.

JAMIA open·2026

Same author

CROSS-MODAL FINE-TUNING OF 3D CONVOLUTIONAL FOUNDATION MODELS FOR ADHD CLASSIFICATION WITH LOW-RANK ADAPTATION.

Proceedings. IEEE International Symposium on Biomedical Imaging·2026

Same author

Strengthening microbiology laboratory capacity in Latin America and the Caribbean region to bolster global health security agenda.

Journal of clinical microbiology·2026

Same author

The learning curve of single-port extraperitoneal robotic radical prostatectomy: Initial experience and outcomes from a newly graduated fellowship-trained robotic surgeon.

BJUI compass·2026

Same author

From data to decision: integrating causality AI and predictive analytics in endourological practice-a descriptive guide for clinicians from EAU Endourology.

World journal of urology·2026

This summary is machine-generated.

We developed Green U-shaped Learning (GUSL), a novel, interpretable deep learning model for prostate cancer segmentation. GUSL achieves state-of-the-art results with high energy efficiency and transparency, making it clinically practical.

Area of Science:

Medical imaging analysis
Machine learning for healthcare
Prostate cancer diagnostics

Background:

Prostate and zonal segmentation are vital for prostate cancer (PCa) diagnosis.
Current deep learning (DL) models for segmentation lack transparency, hindering clinical adoption.
Physicians perceive DL models as "black-box" solutions.

Purpose of the Study:

Introduce Green U-shaped Learning (GUSL), a feed-forward machine learning model for medical image segmentation.
Develop an interpretable and energy-efficient alternative to traditional DL models.
Improve the clinical practicality of AI-driven segmentation tools.

Main Methods:

GUSL utilizes a feed-forward architecture without backpropagation.
Employs a multi-layer regression for coarse-to-fine segmentation and linear models for interpretable feature extraction.

Keywords:

Feed-forward model Machine learning Magnetic resonance Medical imaging Prostate segmentation Regression

Related Experiment Videos

Last Updated: Jun 29, 2026

A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound

A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound

Published on: March 21, 2025

Incorporates boundary attention and residue correction for improved accuracy.

A two-step pipeline addresses class imbalance in medical imaging.

Main Results:

GUSL achieved state-of-the-art performance in prostate gland and zonal segmentation across multiple datasets.
Achieved a Dice Similarity Coefficient (DSC) greater than 0.9 for gland segmentation.
Demonstrated significantly smaller model size and lower complexity compared to other DL models.
Exhibited transparent feature extraction and high energy efficiency.

Conclusions:

GUSL offers a competitive and practical solution for medical imaging segmentation.
Its interpretability, efficiency, and performance make it suitable for clinical deployment.
The model addresses key limitations of current DL approaches in medical diagnosis.