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 Videos

Medical image classification via multiscale representation learning.

Qiling Tang¹, Yangyang Liu¹, Haihua Liu²

¹College of Biomedical Engineering, South Central University for Nationalities, Wuhan 430074, PR China.

Artificial Intelligence in Medicine

|July 14, 2017

Summary

This summary is machine-generated.

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

An integrated molecular classification system identifies distinct prognostic clusters of non-muscle-invasive bladder cancer.

Functional & integrative genomics·2026

Same author

EED-Astig: A Multimodal Dataset for Pediatric Astigmatism Severity Prediction.

Scientific data·2026

Same author

Raphe subregion seeds reveal whole-brain functional connectivity differences underlying adolescent depression.

BMC psychiatry·2026

Same author

Pueraria lobata-derived exosome-like nanovesicles alleviate rheumatoid arthritis via targeting <i>Ruminococcus gnavus</i> phenylethylamine production.

Gut·2025

Same author

Transcriptome analysis of differentially expressed genes in lily leaves under selenite application.

Scientific reports·2025

Same author

Enhanced recovery after liver transplantation-a prospective analysis focusing on quality assessment.

Hepatobiliary surgery and nutrition·2025

Same journal

Real-time EEG-based epileptic seizure prediction using artificial intelligence: A systematic review.

Artificial intelligence in medicine·2026

Same journal

R-peak detection and ECG data compression scheme based on empirical mode decomposition and wavelet transform.

Artificial intelligence in medicine·2026

Same journal

CastNet: A three-channel EEG-based deep learning model for cross-subject depression detection.

Artificial intelligence in medicine·2026

Same journal

State-of-the-art TinyML approaches for colorectal cancer detection: Current advances, challenges, and future directions.

Artificial intelligence in medicine·2026

Same journal

JRadiEvo: A Japanese radiology report generation model enhanced by evolutionary optimization of model merging.

Artificial intelligence in medicine·2026

Same journal

Causally-informed deep learning towards explainable and generalizable outcome prediction in critical care.

Artificial intelligence in medicine·2026

See all related articles

This study introduces a novel multiscale representation learning method using sparse autoencoder networks for medical image classification. The approach effectively captures intrinsic scales, improving classification performance on medical datasets.

Area of Science:

Medical Imaging
Machine Learning
Computer Vision

Background:

Natural images exhibit multiscale structures, a property also observed in medical imaging.
Analyzing medical images across various scales is crucial for accurate measurements and diagnostics.

Purpose of the Study:

To propose a multiscale representation learning method for medical image classification.
To capture intrinsic scaling phenomena in medical images using sparse autoencoder networks.

Main Methods:

Utilized sparse autoencoder networks with varying receptive field sizes to obtain multiscale feature detectors.
Generated feature maps via convolution operations and employed Fisher vector encoding for fixed-length image representation.
Applied the method to the IRMA-2009 medical collection and a mammographic patch dataset.

Keywords:

Fisher vector Image classification Multiscale feature learning Sparse autoencoder

Related Experiment Videos

Main Results:

The multiscale approach demonstrated superior characterization of structures compared to single-scale methods.
Fisher vector encoding enhanced feature representation with high-order statistics, improving descriptiveness and discriminative ability.
Achieved superior performance on the tested medical imaging datasets.

Conclusions:

The proposed multiscale representation learning method effectively captures intrinsic scales in medical images.
This technique offers enhanced feature representation for improved medical image classification accuracy.
The findings suggest significant potential for this method in clinical applications.