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

Related Concept Videos

Deconvolution

Deconvolution

Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...

Downsampling

Downsampling

When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...

Difference from Background: Limit of Detection

Difference from Background: Limit of Detection

The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...

Upsampling

Upsampling

Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Effectiveness of Interruptive Clinical Decision Support Alerts on Intravenous vs. Oral Acetaminophen Prescribing.

Applied clinical informatics·2026

Same author

[Impact of preoperative cardiology consultation on major adverse cardiac events in non-cardiac surgery patients: a retrospective cohort study].

Nan fang yi ke da xue xue bao = Journal of Southern Medical University·2026

Same author

Equivariant Filters for Efficient Tracking in 3D Imaging.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2026

Same author

Evaluating reliability of automated quantitative brain morphometry from fetal T2-weighted MRI.

Frontiers in neuroscience·2026

Same author

Toward real-time alignment of 3D CT and 2D X-ray with multi-stage CNNs.

Computer assisted surgery (Abingdon, England)·2026

Same author

An Evaluation of the Inhibitory Effect of Apatinib on Ilaprazole Metabolism: in vitro, in vivo and in silico Study.

Drug design, development and therapy·2026

Same journal

Towards the Efficient Inference by Incorporating Automated Computational Phenotypes under Covariate Shift.

Proceedings of machine learning research·2026

Same journal

Endo-SemiS: Towards Robust Semi-Supervised Image Segmentation for Endoscopic Video.

Proceedings of machine learning research·2026

Same journal

Perspective: Machine Learning for Health Should Consider Social Drivers of Health.

Proceedings of machine learning research·2026

Same journal

Classifying Phonotrauma Severity from Vocal Fold Images with Soft Ordinal Regression.

Proceedings of machine learning research·2026

Same journal

Does Domain-Specific Retrieval Augmented Generation Help LLMs Answer Consumer Health Questions?

Proceedings of machine learning research·2026

Same journal

Quantitative Convergence Analysis of Projected Stochastic Gradient Descent for Non-Convex Losses via the Goldstein Subdifferential.

Proceedings of machine learning research·2026

See all related articles

Search research articles

Related Experiment Videos

Sample-Specific Debiasing for Better Image-Text Models.

Peiqi Wang¹, Yingcheng Liu¹, Ching-Yun Ko¹

¹Massachusetts Institute of Technology, Cambridge, MA, USA.

Proceedings of Machine Learning Research

|March 25, 2026

Summary

This summary is machine-generated.

This study introduces a novel method to improve self-supervised learning in medical AI by correcting false negatives in contrastive learning. This enhances representation quality for critical healthcare applications.

Related Experiment Videos

Area of Science:

Artificial Intelligence
Medical Informatics
Computer Vision

Background:

Self-supervised representation learning on image-text data is vital for medical AI tasks like classification and retrieval.
Contrastive learning methods often use uniform negative sampling, which introduces false negatives in healthcare datasets due to non-uniform class distributions.
These false negatives can degrade the quality of learned representations.

Purpose of the Study:

To develop a novel approach for correcting false negatives in self-supervised representation learning.
To improve the quality of learned representations from medical image-text data.

Main Methods:

The study proposes a debiased contrastive learning variant that estimates sample-specific class probabilities.
This method corrects for false negatives by adjusting the contrastive objective function.
Theoretical analysis of the objective function was performed.

Main Results:

The proposed approach demonstrates empirical advantages on both image and paired image-text datasets.
Sample-specific debiasing effectively mitigates the negative impact of false negatives.
Improved representation quality was observed.

Conclusions:

The novel method enhances self-supervised representation learning by addressing false negatives in contrastive learning.
This approach offers a significant improvement for medical AI applications relying on accurate data representations.
The technique is effective for both image and image-text data.