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

Pharmacogenetics of Drug Metabolism: Overview01:27

Pharmacogenetics of Drug Metabolism: Overview

191
Genetic polymorphism in drug metabolism is crucial to the inter-individual variability observed in drug responses. Drug metabolism primarily involves the chemical modification of drugs and other xenobiotics to enhance their elimination by increasing their polarity. Two main classes of enzymes mediate this biotransformation process: Phase I enzymes, primarily cytochrome P450s, catalyze oxidation and reduction reactions, while other enzymes, such as esterases, mediate hydrolysis, and Phase II...
191

You might also read

Related Articles

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

Sort by
Same author

A survey on artificial intelligence use for myeloma and lymphoma management.

Current research in translational medicine·2026
Same author

Blood pressure patterns and cardio-renal complications in primary aldosteronism compared with essential hypertension in a tunisian population.

Journal of human hypertension·2026
Same author

Health-related quality of life in women with hypopituitarism secondary to Sheehan syndrome: assessment and development of a new clinical score.

Journal of diabetes and metabolic disorders·2026
Same author

Management of Adrenal Incidentalomas: A Survey of Physicians from the Middle East and Africa.

La Tunisie medicale·2026
Same author

[Engaging professionals in quality: Understanding and adopting the FACT-JACIE accreditation process (SFGM-TC)].

Bulletin du cancer·2026
Same author

Central line-associated thrombosis and infections in patients with acute leukemias: The chicken or the egg dilemma.

Thrombosis research·2026

Related Experiment Video

Updated: Apr 28, 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

5.8K

Interpretable Machine Learning to Predict Metformin-Induced Vitamin B12 Deficiency: Association with Glycemic Control

Yasmine Salhi1, Meriem Yazidi2, Amine Dhraief3

  • 1ISITCom, University of Sousse, Sousse 4011, Tunisia.

Metabolites
|April 27, 2026
PubMed
Summary

Machine learning accurately predicts vitamin B12 deficiency in type 2 diabetes patients on metformin. Key factors like HbA1c and metformin dose help identify at-risk individuals for targeted screening.

Keywords:
SHAPXGBoostartificial intelligencemetformintype 2 diabetesvitamin B12 deficiency

More Related Videos

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

8.0K

Related Experiment Videos

Last Updated: Apr 28, 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

5.8K
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

8.0K

Area of Science:

  • Endocrinology and Metabolism
  • Artificial Intelligence in Healthcare
  • Nutritional Biochemistry

Background:

  • Vitamin B12 deficiency is a frequent complication in type 2 diabetes (T2D) patients using metformin long-term.
  • Early detection is crucial for timely intervention and preventing adverse outcomes.
  • Current diagnostic approaches may miss subclinical cases, necessitating advanced predictive tools.

Purpose of the Study:

  • To develop and interpret a machine learning model for predicting vitamin B12 deficiency in metformin-treated T2D patients.
  • To identify key clinical predictors of vitamin B12 deficiency using explainable AI techniques.
  • To enable targeted screening strategies for this common complication.

Main Methods:

  • Retrospective study of 257 T2D patients on metformin for ≥3 years.
  • Utilized eXtreme Gradient Boosting (XGBoost) for model development, comparing it with other ML algorithms.
  • Optimized hyperparameters using Bayesian search and validated using SHapley Additive exPlanations (SHAP) for interpretability.

Main Results:

  • The optimized XGBoost model demonstrated moderate predictive performance (ROC-AUC 0.671) on an independent test set.
  • Vitamin B12 deficiency was present in 37.0% of the study cohort.
  • SHAP analysis highlighted HbA1c, microalbuminuria, autonomic neuropathy, BMI, DN4 score, and fasting glucose as significant predictors.

Conclusions:

  • The XGBoost-SHAP framework offers interpretable predictions for vitamin B12 deficiency in T2D patients on metformin.
  • Identified specific patient profiles (e.g., low HbA1c with high metformin dose) at higher risk.
  • Further multi-center validation is recommended prior to clinical implementation.