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

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...

You might also read

Related Articles

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

Sort by

Same author

Retractions of publications in radiomics: An underestimated problem?

European radiology·2025

Same author

Correction: Növer et al. Prediction of Recurrence and Rupture Risk of Ruptured and Unruptured Intracranial Aneurysms of the Posterior Circulation: A Machine Learning-Based Analysis. <i>Diagnostics</i> 2025, <i>15</i>, 2365.

Diagnostics (Basel, Switzerland)·2025

Same author

Prediction of Recurrence and Rupture Risk of Ruptured and Unruptured Intracranial Aneurysms of the Posterior Circulation: A Machine Learning-Based Analysis.

Diagnostics (Basel, Switzerland)·2025

Same author

Benchmarking feature projection methods in radiomics.

Scientific reports·2025

Same author

Rethinking feature reproducibility in radiomics: the elephant in the dark.

European radiology experimental·2025

Same author

Detecting the left atrial appendage in CT localizers using deep learning.

Scientific reports·2025

Search research articles

Related Experiment Video

Updated: May 5, 2026

Guidelines and Experience Using Imaging Biomarker Explorer IBEX for Radiomics

Guidelines and Experience Using Imaging Biomarker Explorer IBEX for Radiomics

Published on: January 8, 2018

The effect of data resampling methods in radiomics.

Aydin Demircioğlu¹

¹Institute of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Hufelandstraße 55, 45147, Essen, Germany. aydin.demircioglu@uk-essen.de.

Scientific Reports

|February 3, 2024

Summary

Resampling methods for imbalanced radiomic data did not improve predictive performance overall. While some minor gains were observed, feature selection showed significant disagreement, impacting interpretability.

More Related Videos

Use of Magnetic Resonance Imaging and Biopsy Data to Guide Sampling Procedures for Prostate Cancer Biobanking

Use of Magnetic Resonance Imaging and Biopsy Data to Guide Sampling Procedures for Prostate Cancer Biobanking

Published on: October 10, 2019

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Related Experiment Videos

Last Updated: May 5, 2026

Guidelines and Experience Using Imaging Biomarker Explorer IBEX for Radiomics

Guidelines and Experience Using Imaging Biomarker Explorer IBEX for Radiomics

Published on: January 8, 2018

Use of Magnetic Resonance Imaging and Biopsy Data to Guide Sampling Procedures for Prostate Cancer Biobanking

Use of Magnetic Resonance Imaging and Biopsy Data to Guide Sampling Procedures for Prostate Cancer Biobanking

Published on: October 10, 2019

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Area of Science:

Medical Imaging and Data Science
Machine Learning in Healthcare
Radiomics Research

Background:

Class-imbalanced datasets are common in radiomics, potentially biasing predictive models.
Resampling techniques are often used to address class imbalance.
The impact of various resampling methods on radiomic model performance and feature selection is not well understood.

Purpose of the Study:

To evaluate the effect of nine resampling methods on radiomic models.
To assess the impact of resampling on predictive performance across 15 public datasets.
To analyze the agreement and similarity of features selected by models using resampling.

Main Methods:

Utilized fifteen publicly available radiomic datasets.
Applied nine different resampling methods to address class imbalance.
Evaluated predictive performance (e.g., AUC) and feature selection consistency.

Main Results:

Resampling methods did not significantly improve average predictive performance.
Minor AUC improvements (around 0.015) were noted on specific datasets.
Significant disagreement in selected features (28.7% agreement) was observed, hindering interpretability.
Selected features showed high correlation on average (82.9%).

Conclusions:

Resampling methods offer limited benefit for improving overall radiomic model predictive performance.
The lack of feature selection consistency poses challenges for model interpretability in radiomics.
Further research is needed to optimize resampling strategies for radiomic applications.