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

Actuarial Approach01:20

Actuarial Approach

223
The actuarial approach, a statistical method originally developed for life insurance risk assessment, is widely used to calculate survival rates in clinical and population studies. This method accounts for participants lost to follow-up or those who die from causes unrelated to the study, ensuring a more accurate representation of survival probabilities.
Consider the example of a high-risk surgical procedure with significant early-stage mortality. A two-year clinical study is conducted,...
223
Kaplan-Meier Approach01:24

Kaplan-Meier Approach

456
The Kaplan-Meier estimator is a non-parametric method used to estimate the survival function from time-to-event data. In medical research, it is frequently employed to measure the proportion of patients surviving for a certain period after treatment. This estimator is fundamental in analyzing time-to-event data, making it indispensable in clinical trials, epidemiological studies, and reliability engineering. By estimating survival probabilities, researchers can evaluate treatment effectiveness,...
456
Assumptions of Survival Analysis01:15

Assumptions of Survival Analysis

301
Survival models analyze the time until one or more events occur, such as death in biological organisms or failure in mechanical systems. These models are widely used across fields like medicine, biology, engineering, and public health to study time-to-event phenomena. To ensure accurate results, survival analysis relies on key assumptions and careful study design.
301
Comparing the Survival Analysis of Two or More Groups01:20

Comparing the Survival Analysis of Two or More Groups

467
Survival analysis is a cornerstone of medical research, used to evaluate the time until an event of interest occurs, such as death, disease recurrence, or recovery. Unlike standard statistical methods, survival analysis is particularly adept at handling censored data—instances where the event has not occurred for some participants by the end of the study or remains unobserved. To address these unique challenges, specialized techniques like the Kaplan-Meier estimator, log-rank test, and...
467
Life Tables01:22

Life Tables

391
A life table is a statistical tool that summarizes the mortality and survival patterns of a population, providing detailed insights into the likelihood of survival or death across different age intervals within a cohort. By organizing data on survival probabilities and mortality rates, life tables offer a clear snapshot of population dynamics over time. They are extensively used in demography, public health, actuarial science, and ecology to analyze life expectancy, design health interventions,...
391
Mechanistic Models: Compartment Models in Individual and Population Analysis01:23

Mechanistic Models: Compartment Models in Individual and Population Analysis

181
Mechanistic models are utilized in individual analysis using single-source data, but imperfections arise due to data collection errors, preventing perfect prediction of observed data. The mathematical equation involves known values (Xi), observed concentrations (Ci), measurement errors (εi), model parameters (ϕj), and the related function (ƒi) for i number of values. Different least-squares metrics quantify differences between predicted and observed values. The ordinary least...
181

You might also read

Related Articles

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

Sort by
Same author

General-purpose large language models outperform specialized clinical AI tools on medical benchmarks.

Nature medicine·2026
Same author

CNS-Obsidian: A Neurosurgical Vision-Language Model Built From Scientific Publications.

Neurosurgery·2026
Same author

Generalist Foundation Models Are Not Clinical Enough for Hospital Operations.

Research square·2026
Same author

Enhancing the prediction of hospital discharge disposition with extraction-based language model classification.

npj health systems·2026
Same author

Automating the Referral of Bone Metastases Patients With and Without the Use of Large Language Models.

Neurosurgery·2025
Same author

The TRIPOD-LLM reporting guideline for studies using large language models: a Korean translation.

Ewha medical journal·2025

Related Experiment Video

Updated: Dec 13, 2025

An R-Based Landscape Validation of a Competing Risk Model
05:37

An R-Based Landscape Validation of a Competing Risk Model

Published on: September 16, 2022

2.4K

Estimating real-world performance of a predictive model: a case-study in predicting mortality.

Vincent J Major1, Neil Jethani1, Yindalon Aphinyanaphongs1

  • 1Department of Population Health, NYU Langone Health, New York, New York, USA.

JAMIA Open
|August 1, 2020
PubMed
Summary

Choosing the right patient data selection and validation methods is crucial for accurate predictive model performance. A forwards-from-admission approach with temporal validation provides a more realistic estimate of real-world effectiveness.

Keywords:
data scienceexperimental designmachine learningmortalityreproducibility of results

More Related Videos

Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

1.9K
Establishing a Competing Risk Regression Nomogram Model for Survival Data
04:57

Establishing a Competing Risk Regression Nomogram Model for Survival Data

Published on: October 23, 2020

10.6K

Related Experiment Videos

Last Updated: Dec 13, 2025

An R-Based Landscape Validation of a Competing Risk Model
05:37

An R-Based Landscape Validation of a Competing Risk Model

Published on: September 16, 2022

2.4K
Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

1.9K
Establishing a Competing Risk Regression Nomogram Model for Survival Data
04:57

Establishing a Competing Risk Regression Nomogram Model for Survival Data

Published on: October 23, 2020

10.6K

Area of Science:

  • Healthcare data science
  • Machine learning in medicine
  • Clinical predictive modeling

Background:

  • Electronic health records (EHRs) are increasingly used for developing AI/ML applications in healthcare.
  • Experimental design choices, such as cohort selection and validation methods, significantly impact model performance estimates.
  • Data-driven risk assessment tools can enhance routine patient care.

Purpose of the Study:

  • To quantify the impact of experimental design choices on predictive model performance.
  • To compare the real-world performance estimation of "backwards-from-outcome" versus "forwards-from-admission" cohort selection.
  • To evaluate the effect of random versus temporal validation on model generalizability.

Main Methods:

  • Developed a 1-year mortality prediction model using four years of hospitalization data.
  • Employed "backwards-from-outcome" and "forwards-from-admission" cohort selection strategies.
  • Utilized random and temporal validation cohorts to assess model performance on a separate, real-world cohort.

Main Results:

  • "Forwards-from-admission" selection included significantly more admissions (92,148) than "backwards-from-outcome" (23,579).
  • Both methods showed similar performance on random test sets.
  • On a temporal validation set, "forwards-from-admission" significantly outperformed "backwards-from-outcome" (AUC 88.3% vs. 83.2%).

Conclusions:

  • "Backwards-from-outcome" selection simplifies experiments but leads to optimistic performance estimates due to data manipulation.
  • "Forwards-from-admission" selection with temporal validation provides a more conservative and realistic estimation of real-world model performance.
  • Careful experimental design is essential for accurately gauging the clinical utility of predictive models.