Search research articles

Search research articles

Related Experiment Video

Updated: Oct 4, 2025

Modeling the Functional Network for Spatial Navigation in the Human Brain

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

On Training Deep 3D CNN Models with Dependent Samples in Neuroimaging.

Yunyang Xiong¹, Hyunwoo J Kim², Bhargav Tangirala¹

¹University of Wisconsin Madison, Madison WI 53706, USA.

Information Processing in Medical Imaging : Proceedings of the ... Conference

|February 7, 2022

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

Overview of the Head and Neck Tumor Segmentation for Magnetic Resonance Guided Applications (HNTS-MRG) 2024 Challenge.

ArXiv·2024

Same author

Trajectories from Distribution-valued Functional Curves: A Unified Wasserstein Framework<sup>★</sup>.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2022

Same author

Simultaneous Multi-Heartbeat ECGI Solution with a Time-Varying Forward Model: a Joint Inverse Formulation.

Functional imaging and modeling of the heart : ... International Workshop, FIMH ..., proceedings. FIMH·2021

See all related 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

This study modifies 3D convolutional neural networks (CNNs) to handle dependent samples in biomedical imaging, improving predictions of cognitive trajectories from brain scans.

Area of Science:

Neuroimaging
Machine Learning
Biomedical Data Analysis

Background:

Standard machine learning assumes independent training samples, which is often violated in biomedical studies with limited, multi-sample datasets.
Violation of the independence assumption is particularly problematic for small-to-medium sized datasets common in medical research.
Existing 3D CNNs may not adequately address sample dependency in brain imaging analysis.

Purpose of the Study:

To adapt 3D convolutional neural networks (CNNs) for analyzing dependent samples in biomedical imaging.
To enhance the performance of 3D CNNs by incorporating information about sample dependency.
To improve the prediction of cognitive trajectories using morphometric changes from longitudinal brain imaging data.

Main Methods:

More Related Videos

Author Spotlight: Advancing 3D Cytoarchitecture Analysis - Rapid Volumetric Reconstruction of the Human Brain

Author Spotlight: Advancing 3D Cytoarchitecture Analysis - Rapid Volumetric Reconstruction of the Human Brain

Published on: January 26, 2024

Related Experiment Videos

Last Updated: Oct 4, 2025

Modeling the Functional Network for Spatial Navigation in the Human Brain

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

Author Spotlight: Advancing 3D Cytoarchitecture Analysis - Rapid Volumetric Reconstruction of the Human Brain

Author Spotlight: Advancing 3D Cytoarchitecture Analysis - Rapid Volumetric Reconstruction of the Human Brain

Published on: January 26, 2024

Modified 3D CNN architectures to explicitly model dependencies between training samples.
Incorporated terms encoding sample dependency into the network's learning objective.
Utilized morphometric change images derived from multiple time points for predicting cognitive trajectories (slope and intercept).

Main Results:

Demonstrated consistent performance improvements by augmenting standard 3D CNNs to account for sample dependency.
Showcased the value of exploiting dependency information even within standard 3D CNN frameworks.
Achieved better prediction of cognitive trajectories compared to a baseline model that ignored sample dependency.

Conclusions:

Modifying 3D CNNs to handle dependent samples offers significant advantages in biomedical imaging analysis.
Accounting for sample dependency is crucial for robust performance, especially with limited or clustered datasets.
The proposed approach enhances the predictive power of neuroimaging models for understanding cognitive trajectories.