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

FlapMap Visual Language System for Vascular Imaging Prior to Microvascular Free Tissue Transfer.

Plastic and reconstructive surgery. Global open·2022
Same author

Development of a machine learning model using electrocardiogram signals to improve acute pulmonary embolism screening.

European heart journal. Digital health·2022
Same author

Lymphoepithelioma-like neoplasm of the biliary tract with 'probable low malignant potential'.

Histopathology·2021
Same author

Connecting Solution-Phase to Single-Molecule Properties of Ni(Salophen).

The journal of physical chemistry letters·2019
Same author

Intercalating Single-Atom Metal Centers into an Organic Monolayer with a Full-Sample Coverage.

Langmuir : the ACS journal of surfaces and colloids·2018
Same author

Preoperative Evaluation of a Pancreas Mass: Diagnostic Options.

The Surgical clinics of North America·2017

Related Experiment Video

Updated: Mar 13, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.9K

Machine Learning Interface for Medical Image Analysis.

Yi C Zhang1, Alexander C Kagen2

  • 1Translational and Molecular Imaging Institute, Icahn School of Medicine at Mount Sinai, 1470 Madison Avenue, 1st Floor, New York, NY, 10029, USA. yi.zhang1@mssm.edu.

Journal of Digital Imaging
|October 13, 2016
PubMed
Summary
This summary is machine-generated.

This study adapted TensorFlow for medical imaging by enabling DICOM input for DaTscan analysis. The developed neural network achieved high accuracy in classifying Parkinson's disease, showing TensorFlow's potential in clinical diagnostics.

Keywords:
Artificial intelligenceClassificationComputer visionImage analysis

More Related Videos

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

1.3K
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

7.6K

Related Experiment Videos

Last Updated: Mar 13, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.9K
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

1.3K
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

7.6K

Area of Science:

  • Medical Imaging
  • Machine Learning
  • Neurology

Background:

  • TensorFlow is a popular machine learning library, but its integration with medical imaging formats like DICOM is limited.
  • Developing robust machine learning models for neurological disorders requires effective handling of medical image data.

Purpose of the Study:

  • To extend the TensorFlow API for direct input of DICOM images, specifically for DaTscan analysis.
  • To develop and evaluate a neural network classifier for distinguishing between normal and Parkinson's disease groups using DaTscan images.

Main Methods:

  • Extracted pixel intensities from 1513 DaTscan DICOM images obtained from the Parkinson's Progression Markers Initiative (PPMI) database.
  • Processed DICOM data into tensors for training, validation, and testing a neural network classifier.
  • Employed gradient descent and Adagrad optimization algorithms over 1000 iterations for model training and cross-validation.

Main Results:

  • Achieved a mean accuracy of 0.938 ± 0.047 in classifying Parkinson's disease.
  • Demonstrated high mean sensitivity (0.974 ± 0.043) and acceptable mean specificity (0.822 ± 0.207).
  • The neural network's diagnostic performance was comparable to existing machine learning models.

Conclusions:

  • Successfully extended TensorFlow's API to incorporate DICOM image compatibility for DaTscan analysis.
  • The developed neural network shows significant potential as an adjunct diagnostic tool in clinical settings for Parkinson's disease detection.
  • This work facilitates the use of TensorFlow in medical image analysis, paving the way for advanced diagnostic applications.