Related Experiment Video
Updated: Oct 26, 2025
10:25
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
48.7K
Quantile Regression for Uncertainty Estimation in VAEs with Applications to Brain Lesion Detection
Haleh Akrami1, Anand Joshi1, Sergul Aydore2
1University of Southern California, Los Angeles, USA.
Information Processing in Medical Imaging : Proceedings of the ... Conference
|August 2, 2021
Summary
The Quantile-Regression Variational AutoEncoder (QR-VAE) improves anomaly detection by addressing variance underestimation. This method enhances lesion detection in medical images by accurately modeling data distributions.
Area of Science:
- Machine Learning
- Medical Imaging Analysis
- Computer Vision
Background:
- Variational Autoencoders (VAEs) are popular for anomaly detection, particularly in medical imaging for tasks like lesion detection.
- VAEs learn data distributions to identify deviations, often using reconstruction probability.
- A key challenge in VAEs is the underestimation of variance when jointly optimizing mean and variance.
Purpose of the Study:
- To introduce a novel Variational AutoEncoder model, the Quantile-Regression VAE (QR-VAE).
- To overcome the variance shrinkage problem inherent in traditional VAEs.
- To provide a principled approach for anomaly detection using reconstruction probability.
Main Methods:
- Developed the Quantile-Regression VAE (QR-VAE) model.
- QR-VAE estimates conditional quantiles instead of directly optimizing variance.
- Computed conditional mean and variance from estimated quantiles for Gaussian modeling.
- Utilized reconstruction probability for anomaly detection.
Main Results:
- The QR-VAE effectively avoids the variance shrinkage problem.
- The model enables principled anomaly detection through reconstruction probability.
- Demonstrated application in heterogeneous thresholding for brain lesion detection.
Conclusions:
- QR-VAE offers a robust solution to variance underestimation in VAEs.
- This approach enhances anomaly detection accuracy in medical imaging.
- QR-VAE shows promise for improved lesion detection in brain scans.
Related Concept Videos
00:54Uncertainty: Confidence Intervals
7.4K
The confidence interval is the range of values around the mean that contains the true mean. It is expressed as a probability percentage. The interpretation of a 95% confidence interval, for instance, is that the statistician is 95% confident that the true mean falls within the interval. The upper and lower limits of this range are known as confidence limits. The confidence limits for the true mean are estimated from the sample's mean, the standard deviation, and the statistical factor...
7.4K
00:59Propagation of Uncertainty from Random Error
1.3K
An experiment often consists of more than a single step. In this case, measurements at each step give rise to uncertainty. Because the measurements occur in successive steps, the uncertainty in one step necessarily contributes to that in the subsequent step. As we perform statistical analysis on these types of experiments, we must learn to account for the propagation of uncertainty from one step to the next. The propagation of uncertainty depends on the type of arithmetic operation performed on...
1.3K
01:10Propagation of Uncertainty from Systematic Error
1.1K
The atomic mass of an element varies due to the relative ratio of its isotopes. A sample's relative proportion of oxygen isotopes influences its average atomic mass. For instance, if we were to measure the atomic mass of oxygen from a sample, the mass would be a weighted average of the isotopic masses of oxygen in that sample. Since a single sample is not likely to perfectly reflect the true atomic mass of oxygen for all the molecules of oxygen on Earth, the mass we obtain from this...
1.1K
01:03Prediction Intervals
2.5K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
2.5K
00:59Uncertainty: Overview
1.2K
In analytical chemistry, we often perform repetitive measurements to detect and minimize inaccuracies caused by both determinate and indeterminate errors. Despite the cares we take, the presence of random errors means that repeated measurements almost never have exactly the same magnitude. The collective difference between these measurements - observed values - and the estimated or expected value is called uncertainty. Uncertainty is conventionally written after the estimated or expected value.
1.2K
01:24One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation
824
This lesson introduces two critical methods in pharmacokinetics, the Wagner-Nelson and Loo-Riegelman methods, used for estimating the absorption rate constant (ka) for drugs administered via non-intravenous routes. The Wagner-Nelson method relates ka to the plasma concentration derived from the slope of a semilog percent unabsorbed time plot. However, it is limited to drugs with one-compartment kinetics and can be impacted by factors like gastrointestinal motility or enzymatic degradation.
On...
On...
824