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

End Point Prediction: Gran Plot01:07

End Point Prediction: Gran Plot

881
A Gran plot is used to predict the equivalence volume or endpoint of a potentiometric or acid-base titration without reaching the endpoint. Typically, titration data is collected as a function of the titrant's volume up to a point less than the equivalence volume and then transformed into a linear format. The straight line is extended to the x-axis, indicating the necessary titrant volume to achieve the equivalence point.
For potentiometric titration, the Gran plot is created by plotting...
881
Model Approaches for Pharmacokinetic Data: Distributed Parameter Models01:06

Model Approaches for Pharmacokinetic Data: Distributed Parameter Models

166
Pharmacokinetic models are mathematical constructs that represent and predict the time course of drug concentrations in the body, providing meaningful pharmacokinetic parameters. These models are categorized into compartment, physiological, and distributed parameter models.
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
166
One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation01:24

One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation

902
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...
902
Randomized Experiments01:13

Randomized Experiments

8.6K
The randomization process involves assigning study participants randomly to experimental or control groups based on their probability of being equally assigned. Randomization is meant to eliminate selection bias and balance known and unknown confounding factors so that the control group is similar to the treatment group as much as possible. A computer program and a random number generator can be used to assign participants to groups in a way that minimizes bias.
Simple randomization
Simple...
8.6K

You might also read

Related Articles

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

Sort by
Same author

Assessing the quality of electronic health record data and the claims linked data for target trial emulation studies.

JAMIA open·2026
Same author

Agentic Authoring of OMOP Concept Sets from Natural Language.

medRxiv : the preprint server for health sciences·2026
Same author

Glucagon-Like Peptide-1 Receptor Agonists and Cardiovascular Events in Adults With Obesity and Autoimmune Disease: A Target Trial Emulation.

Journal of the American Heart Association·2026
Same author

GatorDuo: Global-Consistency Dual-Graph Refinement With Pseudo-Label Agreement for Spatial Transcriptomics.

bioRxiv : the preprint server for biology·2026
Same author

Machine Learning Prediction of Pharmacogenetic Testing Uptake Among Opioid-Prescribed Patients Using Electronic Health Records: Retrospective Cohort Study.

JMIR medical informatics·2026
Same author

LADDER: Language-Driven Slice Discovery and Error Rectification in Vision Classifiers.

Findings of ACL. ACL·2026

Related Experiment Video

Updated: Nov 17, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.5K

Deep propensity network using a sparse autoencoder for estimation of treatment effects.

Shantanu Ghosh1, Jiang Bian2, Yi Guo2

  • 1Department of Computer and Information Science and Engineering, University of Florida, Gainesville, Florida, USA.

Journal of the American Medical Informatics Association : JAMIA
|February 17, 2021
PubMed
Summary
This summary is machine-generated.

Estimating causal effects from observational data is challenging due to bias. The novel Deep Propensity Network using a Sparse Autoencoder (DPN-SA) improves treatment effect estimation and counterfactual prediction in complex datasets.

Keywords:
big databiomedical informaticscausal AIcausal inferencedeep learningelectronic health recordpropensity scoretreatment effect

More Related Videos

Decoding Natural Behavior from Neuroethological Embedding
08:00

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

279
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

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

Published on: October 10, 2018

8.5K

Related Experiment Videos

Last Updated: Nov 17, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.5K
Decoding Natural Behavior from Neuroethological Embedding
08:00

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

279
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

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

Published on: October 10, 2018

8.5K

Area of Science:

  • Causal inference
  • Machine learning
  • Health data science

Background:

  • Estimating causal effects from observational data is difficult due to inherent biases, such as discrimination in treatment assignment.
  • Evaluating 'what-if' scenarios, or counterfactuals, is crucial for understanding true causal effects.
  • High dimensionality and complex treatment assignment patterns in datasets pose significant challenges for traditional methods.

Purpose of the Study:

  • To introduce a novel deep learning architecture, the Deep Propensity Network using a Sparse Autoencoder (DPN-SA), for propensity score matching and counterfactual prediction.
  • To address challenges in treatment effect estimation, including high dimensionality, nonlinear/nonparallel treatment assignment, and residual confounding.
  • To provide a robust method for estimating causal effects in the presence of complex data structures.

Main Methods:

  • The study utilized two randomized prospective datasets and one semi-synthetic dataset with simulated counterfactual outcomes.
  • A novel deep learning architecture, DPN-SA, was developed and compared against logistic regression, LASSO, and Deep Counterfactual Networks with Propensity Dropout (DCN-PD).
  • Model performance was evaluated using metrics such as average treatment effects, mean squared error for heterogeneity precision, and average treatment effect on the treated.

Main Results:

  • The DPN-SA demonstrated superior performance, outperforming logistic regression and LASSO by 36%-63% and DCN-PD by 6%-10% across all tested datasets.
  • All deep learning architectures consistently yielded average treatment effects close to true values with low variance.
  • The DPN-SA's performance remained robust even when subjected to noise injection and the addition of correlated variables.

Conclusions:

  • Deep sparse autoencoders, as implemented in DPN-SA, are highly suitable for treatment effect estimation, particularly with electronic health records.
  • The DPN-SA effectively handles high-dimensional covariate sets, large sample sizes, and complex heterogeneity in treatment assignments.
  • This novel architecture offers a significant advancement in accurately estimating causal effects from complex observational data.