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

Cancer Survival Analysis

Cancer Survival Analysis

Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...

You might also read

Related Articles

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

Sort by

Same author

Mapping radiosensitivity in glioblastoma using MR elastography and biomechanical modeling.

Scientific reports·2026

Same author

Automated delineation of putative non-contrast-enhancing tumor in glioblastoma: Prognostic insights.

Neuro-oncology·2026

Same author

Biomechanical mapping of tumor growth: A novel method to quantify glioma infiltration and mass effect.

Medical physics·2026

Same author

Comparing Glymphatic Function Measures: Diffusion Tensor Image Analysis Along Perivascular Spaces (DTI-ALPS) versus Intrathecal Contrast-Enhanced MRI.

Radiology·2026

Same author

Modeling glioma-induced impairments on the glymphatic system.

Fluids and barriers of the CNS·2026

Same author

Monitoring the lateral ventricles in the presence of intracranial hemorrhage using automated dual segmentation.

Medical physics·2026

Search research articles

Related Experiment Video

Updated: Jul 13, 2025

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Interpreting deep learning models for glioma survival classification using visualization and textual explanations.

Michael Osadebey¹, Qinghui Liu², Elies Fuster-Garcia³

¹Department of Physics and Computational Radiology, Division of Radiology and Nuclear Medicine, Oslo University Hospital, Sognsvannsveien 20, 0372, Oslo, Norway. osademic@gmail.com.

BMC Medical Informatics and Decision Making

|October 18, 2023

Summary

This study enhances deep learning interpretability for glioma survival prediction by extracting domain knowledge from saliency maps. Tumors near critical brain regions indicate shorter survival, aiding clinical decisions.

Keywords:

3D Gradient Weighted Class Activation Mapping (3D-Grad-CAM)Convolutional Neural Network (CNN)Deep learning Glioblastoma Magnetic Resonance Imaging (MR1)

More Related Videos

Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections

Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections

Published on: July 31, 2017

Implementation of Minimally Invasive Brain Tumor Resection in Rodents for High Viability Tissue Collection

Implementation of Minimally Invasive Brain Tumor Resection in Rodents for High Viability Tissue Collection

Published on: May 9, 2022

Related Experiment Videos

Last Updated: Jul 13, 2025

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections

Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections

Published on: July 31, 2017

Implementation of Minimally Invasive Brain Tumor Resection in Rodents for High Viability Tissue Collection

Implementation of Minimally Invasive Brain Tumor Resection in Rodents for High Viability Tissue Collection

Published on: May 9, 2022

Area of Science:

Neuroimaging
Artificial Intelligence
Oncology

Background:

Saliency-based algorithms aid in understanding deep learning models' predictions from input images.
Assessing the clinical significance of image features and predictions from raw saliency maps can be challenging.
This study aims to improve the interpretability of deep learning models for glioma patient survival classification.

Purpose of the Study:

To enhance the interpretability of deep learning models for glioma survival classification.
To extract domain knowledge-based information from saliency maps for improved clinical value.
To correlate tumor location and volume with patient survival outcomes.

Main Methods:

Utilized a dataset of 147 glioma patients' pre-surgical MRI scans (T1, T2, T2-FLAIR) from the BraTs 2020 challenge.
Developed a 3D convolutional neural network (CNN) for classifying patients into short-term, medium-term, and long-term survival groups.
Extended 2D Gradient-weighted Class Activation Mapping (Grad-CAM) to 3D and integrated it with the SRI 24 anatomical atlas to extract region-specific information.

Main Results:

Larger tumor volume correlated with shorter overall survival (OS).
Identified specific tumor locations associated with distinct survival durations (short, medium, long-term).
Tumors in the transverse temporal gyrus, fusiform, and palladium were linked to short, medium, and long-term survival, respectively.
Highlighted tumor location and regional brain contributions to OS prediction, aiding physician analysis.

Conclusions:

Extracting domain knowledge from saliency maps significantly enhances deep learning model interpretability.
Tumors located in eloquent brain regions are associated with poorer patient survival outcomes.
The developed method provides valuable insights for understanding glioma prognosis and supporting clinical decision-making.