Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Steps in Outbreak Investigation01:18

Steps in Outbreak Investigation

487
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
487
Detection of Gross Error: The Q Test01:00

Detection of Gross Error: The Q Test

6.9K
When one or more data points appear far from the rest of the data, there is a need to determine whether they are outliers and whether they should be eliminated from the data set to ensure an accurate representation of the measured value. In many cases, outliers arise from gross errors (or human errors) and do not accurately reflect the underlying phenomenon. In some cases, however, these apparent outliers reflect true phenomenological differences. In these cases, we can use statistical methods...
6.9K
Receiver Operating Characteristic Plot01:15

Receiver Operating Characteristic Plot

470
A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...
470
What Are Outliers?01:12

What Are Outliers?

4.9K
Outliers are observed data points that are far from the least squares line. They have unusual values and need to be examined carefully. Though an outlier may result from erroneous data, at other times, it may hold valuable information about the population under study and should be included in the data. Hence, it is crucial to examine what causes a data point to be an outlier.
The z score is used to find outliers or unusual values. It should be noted that any values beyond -2 and +2 are...
4.9K
Quantifying and Rejecting Outliers: The Grubbs Test01:02

Quantifying and Rejecting Outliers: The Grubbs Test

3.5K
Sometimes, a data set can have a recorded numerical observation that greatly  deviates from the rest of the data. Assuming that the data is normally distributed, a statistical method called the Grubbs test can be used to determine whether the observation is truly an outlier.  To perform a two-tailed Grubbs test, first, calculate the absolute difference between the outlier and the mean. Then, calculate the ratio between this difference and the standard deviation of the sample. This...
3.5K
Outliers and Influential Points01:08

Outliers and Influential Points

6.1K
An outlier is an observation of data that does not fit the rest of the data. It is sometimes called an extreme value. When you graph an outlier, it will appear not to fit the pattern of the graph. Some outliers are due to mistakes (for example, writing down 50 instead of 500), while others may indicate that something unusual is happening. Outliers are present far from the least squares line in the vertical direction. They have large "errors," where the "error" or residual is the...
6.1K

You might also read

Related Articles

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

Sort by
Same author

Leveraging Artificial Intelligence in Drug and Biological Product Development: An FDA and Clinical Trial Transformation Initiative Workshop Report.

NEJM AI·2025
Same author

Mitigating Limited Data Challenges to Improve Artificial Intelligence Integration in Rare Disease Drug Development.

NEJM AI·2025
Same author

Artificial Intelligence in Clinical and Translational Science: From Bench Insights to Bedside Impact.

Clinical and translational science·2025
Same author

US Food and Drug Administration Approval Summary: Ribociclib With an Aromatase Inhibitor in the Adjuvant Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Stage II and III High-Risk Early Breast Cancer Treatment Setting.

Journal of clinical oncology : official journal of the American Society of Clinical Oncology·2025
Same author

Using Quantitative Approaches to Optimize Dosages for New Combinations and Subsequent Indications for Oncology Drugs.

Clinical pharmacology and therapeutics·2025
Same author

AI-Driven Applications in Clinical Pharmacology and Translational Science: Insights From the ASCPT 2024 AI Preconference.

Clinical and translational science·2025

Related Experiment Video

Updated: Jan 14, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

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

Published on: August 16, 2020

7.4K

Out-of-Distribution Detection as a Risk-Control Strategy for Medical Classification Machine Learning Models.

Chu Weng1,2, Joshua Ward1,3, Wesley Lin1

  • 1Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, Maryland, USA.

Clinical and Translational Science
|October 23, 2025
PubMed
Summary
This summary is machine-generated.

Out-of-distribution (OOD) detection algorithms identify patients unlikely to be from training data, improving medical artificial intelligence (AI) reliability. These methods help mitigate risks when deploying AI in real-world healthcare settings.

Keywords:
clinical riskmachine learningout‐of‐distribution detection

More Related Videos

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.6K
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.9K

Related Experiment Videos

Last Updated: Jan 14, 2026

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

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

Published on: August 16, 2020

7.4K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.6K
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.9K

Area of Science:

  • Medical Artificial Intelligence
  • Machine Learning in Healthcare
  • Algorithm Reliability

Background:

  • Artificial intelligence (AI) and machine learning (ML) are increasingly used in medicine.
  • High-stakes medical AI requires robust performance guardrails.
  • Out-of-distribution (OOD) detection is a critical guardrail for identifying unreliable predictions.

Purpose of the Study:

  • To evaluate state-of-the-art OOD detection algorithms in medical contexts.
  • To assess the utility of OOD detection for improving medical AI performance and safety.
  • To investigate if OOD detection can identify underrepresented patient subsets.

Main Methods:

  • Assessed OOD detection algorithms on three diverse medical datasets (image, transcriptomics, time series).
  • Utilized a simulated training-deployment scenario to evaluate algorithm performance.
  • Analyzed the ability of OOD detectors to identify patients with poor model performance and underrepresented data.

Main Results:

  • Several OOD detectors consistently identified patients on whom the AI model performed poorly.
  • OOD detection methods successfully flagged patient subsets that were underrepresented in the training data.
  • The findings suggest OOD detection is a viable strategy for enhancing medical AI safety.

Conclusions:

  • OOD detection algorithms can serve as effective guardrails for medical AI.
  • Implementing OOD detection can mitigate risks associated with deploying AI in clinical practice.
  • Further investigation into OOD detection for underrepresented groups is warranted.