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

Denoising of low-dose chest computed tomography images using a U-net based convolutional autoencoder and transfer learning.

Biomedical physics & engineering express·2026
Same author

Disrupted Emergent Properties of the Brain in Schizophrenia: Insight From Integrated Information Decomposition of Resting State fMRI.

Brain and behavior·2026
Same author

Explainable AI-Based Hyperspectral Classification Reveals Differences in Spectral Response over Phenological Stages.

Biology·2026
Same author

Efficient machine learning models leveraging DCE-MRI morphological and dynamic features allow accurate breast lesion classification.

Biomedical physics & engineering express·2026
Same author

Radiomics and Deep Learning Interplay for Predicting MGMT Methylation in Glioblastoma: The Crucial Role of Segmentation Quality.

Cancers·2025
Same author

Unveiling complex patterns: An information-theoretic approach to high-order behaviors in microarray data.

PloS one·2025

Related Experiment Video

Updated: Jun 5, 2025

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

769

Tackling the small data problem in medical image classification with artificial intelligence: a systematic review.

Stefano Piffer1,2, Leonardo Ubaldi1,2, Sabina Tangaro3,4

  • 1Department of Experimental and Clinical Biomedical Sciences, University of Florence, Florence, Italy.

Progress in Biomedical Engineering (Bristol, England)
|December 10, 2024
PubMed
Summary

AI in medical imaging faces small dataset challenges. Techniques like transfer learning and data augmentation are common, but external validation and data sharing remain limited, impacting model generalizability and reproducibility.

Keywords:
artificial intelligenceclassificationdata augmentationmedical imagingsmall datatransfer learning

More Related Videos

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
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.4K

Related Experiment Videos

Last Updated: Jun 5, 2025

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

769
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
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.4K

Area of Science:

  • Artificial Intelligence
  • Medical Imaging
  • Data Science

Background:

  • AI research in medical imaging is growing, but constrained by limited data availability.
  • Small datasets pose significant challenges, leading to overfitting issues in model training.
  • Collecting extensive medical imaging data is often infeasible or resource-intensive.

Purpose of the Study:

  • To systematically review recent advancements in addressing small sample sizes in medical imaging AI.
  • To survey techniques used to overcome data scarcity in AI for medical imaging.
  • To assess the quality and transparency of publications in this field.

Main Methods:

  • A systematic review of 147 PubMed articles published up to July 31, 2022, following PRISMA guidelines.
  • Eligibility assessment by two independent reviewers, with 77 studies included.
  • Reporting standards adherence evaluated using the TRIPOD statement.

Main Results:

  • Transfer learning (75%), data augmentation (69%), and generative adversarial networks (14%) were primary methods for small data issues.
  • Over 60% of studies used binary classification due to data scarcity.
  • External validation was reported in only four studies; data and code access was limited (>80% unavailable).
  • Adherence to TRIPOD reporting standards was suboptimal (<50% for 13 items).

Conclusions:

  • While techniques like transfer learning are widely adopted for small medical imaging datasets, significant gaps exist in external validation and data/code sharing.
  • Improving publication transparency and adherence to reporting standards (e.g., TRIPOD) is crucial for enhancing the quality and generalizability of AI models in medical imaging.