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 Experiment Video

Updated: Mar 17, 2026

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

8.9K

Predicting lithium-associated thyroid dysfunction through dimension reduction to establish a practical machine

Shu-Yu Jheng1, Fan-Ying Chan2, Fang-Yung Chang2

  • 1Department of Clinical Pharmacy, School of Pharmacy, College of Pharmacy, Taipei Medical University, Taipei, 11031, Taiwan; Department of Pharmacy, Taipei Medical University Hospital, Taipei Medical University, Taipei, 11031, Taiwan.

Journal of Affective Disorders
|March 15, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Catalpol reverses doxorubicin-induced cognitive decline, neuroinflammation, and restores synaptic integrity.

Phytomedicine : international journal of phytotherapy and phytopharmacology·2026
Same author

Incidence of Clinically Significant Neutropenia and Complications Related to Antibody-Drug Conjugates: A Real-World Study at the University of California.

Cancers·2026
Same author

Machine learning studies of drug-induced nephrotoxicity: a scoping review.

Therapeutic advances in drug safety·2026
Same author

Temporal Profile of Histological Changes During the Ovarian Cycle, Secretory Cycle of the Colleteric Gland, and Molt Cycle of the Mantle Cavity in the Mature Externa of the Parasitic Barnacle <i>Polyascus planus</i>.

Zoological studies·2025
Same author

Web-Based Explainable Machine Learning-Based Drug Surveillance for Predicting Sunitinib- and Sorafenib-Associated Thyroid Dysfunction: Model Development and Validation Study.

JMIR formative research·2025
Same author

[Development of an Education and Training Program for New Intensive Care Unit Staff: Ladders to Success].

Hu li za zhi The journal of nursing·2024

A simplified machine learning model effectively predicts thyroid dysfunction in lithium patients. This model uses only 7 features, maintaining accuracy for clinical use and aiding early intervention strategies.

Area of Science:

  • Clinical Informatics
  • Machine Learning in Medicine
  • Pharmacovigilance

Background:

  • Lithium treatment for mood disorders frequently causes thyroid dysfunction, complicating patient management.
  • Current predictive models for lithium-induced thyroid issues are often too complex for routine clinical application.
  • Developing a simplified yet accurate predictive tool is crucial for proactive patient care.

Purpose of the Study:

  • To develop and validate a machine learning model for predicting lithium-associated thyroid dysfunction.
  • To reduce the dimensionality of predictive models for enhanced clinical utility.
  • To leverage SHAP (SHapley Additive exPlanations) for model interpretability and feature selection.

Main Methods:

  • A multicenter retrospective study analyzed data from 1595 patients treated with lithium carbonate (2010-2021).
Keywords:
Adverse drug reactionDimensionality reductionLithiumMachine learningSHapley Additive exPlanationsThyroid dysfunction

More Related Videos

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.4K

Related Experiment Videos

Last Updated: Mar 17, 2026

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

8.9K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.4K
  • XGBoost, Support Vector Machine, and Logistic Regression models were evaluated.
  • A SHAP-based feature selection strategy was employed to create a simplified 7-feature model.

    • Main Results:

    • The XGBoost model showed strong predictive performance (AUROC 0.773, AUPRC 0.444).
    • The simplified 7-feature model achieved comparable performance (AUROC 0.802, AUPRC 0.460) without significant statistical differences.
    • SHAP analysis confirmed similar feature importance between original and simplified models, enhancing interpretability.

    Conclusions:

    • A simplified, clinically applicable XGBoost model using 7 features accurately predicts lithium-associated thyroid dysfunction.
    • SHAP analysis facilitates understanding of feature contributions, enabling targeted interventions.
    • Further validation in diverse populations is recommended for broader clinical implementation.