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: Nov 2, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

Dual attention multiple instance learning with unsupervised complementary loss for COVID-19 screening.

Philip Chikontwe¹, Miguel Luna¹, Myeongkyun Kang¹

¹Department of Robotics Engineering, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu, South Korea.

Medical Image Analysis

|June 8, 2021

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

Significance of chorioretinal atrophy using ultra-widefield reflectance image as a risk factor for myopic choroidal neovascularization.

Korean journal of ophthalmology : KJO·2026

Same author

Pregnancy Outcomes and Associated Complications in Patients Undergoing Hemodialysis and Their Neonates: A Nationwide Study in South Korea (2014-2022).

Journal of clinical medicine·2026

Same author

Comparative effectiveness of Orsiro, Xience, Synergy, and Resolute stent in patients with acute myocardial infarction undergoing percutaneous coronary intervention: a nationwide registry study.

Coronary artery disease·2026

Same author

Clinical safety and breast symmetry after injectable acellular dermal fibers reconstruction: an MR-based volumetric study.

Gland surgery·2026

Same author

AptoDetect™-Lung Assay as a Blood-Based Predictor of Advanced-Stage Lung Cancer in Patients with Lung-RADS 3-4 Pulmonary Nodules: A Multicenter Prospective Cohort Study.

Biomedicines·2026

Same author

Dental caries risk in patients with chronic airway diseases using dry powder inhalers: a retrospective cohort study.

BMC oral health·2026

Same journal

Multi-class segmentation of aortic branches and zones in computed tomography angiography: The AortaSeg24 challenge.

Medical image analysis·2026

Same journal

HiVLR: Hierarchical Vision-Language Reasoning for interpretable zero-shot radiography image understanding.

Medical image analysis·2026

Same journal

FAA-Net: Fetal abdominal anomaly diagnosis in prenatal ultrasound via LLM-enhanced multi-instance learning.

Medical image analysis·2026

Same journal

Wavelet-inspired diffusion model with near-field constraint for real-time echocardiography dehazing.

Medical image analysis·2026

Same journal

Co-assistant networks by pathology foundation model and convolutional neural network for gigapixel whole slide image analysis.

Medical image analysis·2026

Same journal

MBAS2024: A large-scale benchmark for multi-class bi-atrial segmentation in multi-center contrast-enhanced MRIs.

Medical image analysis·2026

See all related articles

This summary is machine-generated.

This study introduces a novel AI framework for diagnosing COVID-19 and bacterial pneumonia using chest CT scans. The Dual Attention Contrastive based Multiple Instance Learning (DA-CMIL) model achieves high accuracy, aiding rapid and reliable disease detection.

Area of Science:

Medical Imaging
Artificial Intelligence
Infectious Disease Diagnostics

Background:

Chest computed tomography (CT) is crucial for diagnosing Coronavirus Disease 2019 (COVID-19), especially given limitations of RT-PCR.
Accurate CT-based diagnosis is challenging due to annotation difficulties and similarities between COVID-19 and other pneumonias.

Purpose of the Study:

To develop an advanced, weakly supervised framework for rapid COVID-19 and bacterial pneumonia diagnosis using chest CT.
To enhance diagnostic accuracy through attention mechanisms and contrastive learning.

Main Methods:

Proposed Dual Attention Contrastive based Multiple Instance Learning (DA-CMIL) framework.
Utilized attention mechanisms for spatial and latent context, combined with unsupervised contrastive learning.

Keywords:

COVID-19 CT images Deep learning Multiple instance learning Unsupervised complementary loss

More Related Videos

Development of a Gaze-Contingent Display Framework Designed for Perceptual and Oculomotor Research with Simulated Central Vision Loss

Development of a Gaze-Contingent Display Framework Designed for Perceptual and Oculomotor Research with Simulated Central Vision Loss

Published on: April 11, 2025

Related Experiment Videos

Last Updated: Nov 2, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

Development of a Gaze-Contingent Display Framework Designed for Perceptual and Oculomotor Research with Simulated Central Vision Loss

Development of a Gaze-Contingent Display Framework Designed for Perceptual and Oculomotor Research with Simulated Central Vision Loss

Published on: April 11, 2025

Trained on multiple patient CT slices (instances) to output a single diagnostic label.

Main Results:

Achieved an overall accuracy of 98.6% and an Area Under the Curve (AUC) of 98.4%.
Demonstrated the effectiveness of contrastive learning integrated with Multiple Instance Learning (MIL).
Spatial attention provided interpretable diagnostic insights.

Conclusions:

The DA-CMIL framework offers a highly accurate and efficient method for diagnosing COVID-19 and bacterial pneumonia from chest CT scans.
Attention and contrastive learning significantly improve diagnostic performance in weakly supervised settings.