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

Predicting Chronic Kidney Disease from Biomarkers: An Explainable Machine Learning Approach.

Diagnostics (Basel, Switzerland)·2026
Same author

Solitary bone cyst in a 57-year-old woman - A case report.

International journal of surgery case reports·2025
Same author

Bilateral congenital hallux varus with metatarsus adductus in a 6-year-old girl: A case report.

World journal of clinical cases·2025
Same author

University Students' Understanding and Utilization of Food Labels: A Cross-Sectional Study.

International journal of food science·2025
Same author

Total hip arthroplasty for adult patients with hip arthritis in Jordan: clinical profiles and patient characteristics.

International orthopaedics·2023
Same author

<i>Ex vivo</i> assays show human gamma-delta T cells specific for common allergens are Th1-polarized in allergic donors.

Cell reports methods·2023

Related Experiment Video

Updated: Mar 29, 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

Deep Learning Based Computer-Aided Detection of Prostate Cancer Metastases in Bone Scintigraphy: An Experimental

Eslam Jabali1, Omar Almomani2, Louai Qatawneh1

  • 1Department of Nuclear Medicine, King Hussein Medical Center, Royal Medical Services, Amman 11855, Jordan.

Journal of Imaging
|March 27, 2026
PubMed
Summary

Convolutional neural network (CNN) models were evaluated for detecting bone metastases in prostate cancer from scintigraphy scans. DenseNet121 demonstrated a strong balance of diagnostic performance and reliability for automated detection.

Keywords:
bone scintigraphycancercomputer-aided detection (CAD)convolutional neural networks (CNNs)deep learning

More Related Videos

Intra-Cardiac Injection of Human Prostate Cancer Cells to Create a Bone Metastasis Xenograft Mouse Model
06:32

Intra-Cardiac Injection of Human Prostate Cancer Cells to Create a Bone Metastasis Xenograft Mouse Model

Published on: November 4, 2022

7.2K
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
08:05

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence

Published on: June 10, 2025

1.4K

Related Experiment Videos

Last Updated: Mar 29, 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
Intra-Cardiac Injection of Human Prostate Cancer Cells to Create a Bone Metastasis Xenograft Mouse Model
06:32

Intra-Cardiac Injection of Human Prostate Cancer Cells to Create a Bone Metastasis Xenograft Mouse Model

Published on: November 4, 2022

7.2K
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
08:05

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence

Published on: June 10, 2025

1.4K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Bone scintigraphy is crucial for detecting skeletal metastases in prostate cancer.
  • Visual interpretation challenges include heterogeneous uptake, benign mimics, and high workload.
  • Computer-aided decision support is needed to improve accuracy and efficiency.

Purpose of the Study:

  • To experimentally evaluate fourteen convolutional neural network (CNN) architectures for binary metastasis classification in planar bone scintigraphy.
  • To identify the most reliable and deployable CNN model for automated detection of skeletal metastases.

Main Methods:

  • Fourteen CNN models were trained and tested on 600 planar bone scintigraphy images (300 normal, 300 metastatic).
  • A unified protocol with identical preprocessing, augmentation, and stratified five-fold cross-validation was used.
  • Performance was assessed using AUC-ROC, accuracy, sensitivity, specificity, F1-score, Cohen's κ, Brier score, and deployment indicators.

Main Results:

  • DenseNet121 achieved the best overall balance, with AUC-ROC 96.0%, accuracy 89.2%, and strong calibration (Brier 0.080).
  • DenseNet121-attention yielded the highest AUC-ROC (96.3%) but showed greater variability in specificity.
  • DenseNet121 is deployable with ~7.0 M parameters and ~92 ms/image inference time, offering clinical value without excessive complexity.

Conclusions:

  • DenseNet121 is a reliable backbone for automated metastasis detection in planar scintigraphy.
  • The model offers a good balance of diagnostic performance, reliability, and computational efficiency.
  • Future work includes external validation, threshold optimization, interpretability, and model compression for clinical adoption.