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

You might also read

Related Articles

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

Sort by
Same author

Risk factors for adrenal insufficiency after adrenalectomy for mild autonomous cortisol secretion.

Surgery·2026
Same author

Gene- and domain-aware calibration increases the clinical utility of variant effect predictors.

Research square·2026
Same author

Using sensitivity analysis to evaluate the fracture benefit of parathyroidectomy.

Surgery·2026
Same author

A graph-based learning approach to predict the effects of gene perturbations on molecular phenotypes.

bioRxiv : the preprint server for biology·2026
Same author

Gene- and domain-aware calibration increases the clinical utility of variant effect predictors.

bioRxiv : the preprint server for biology·2026
Same author

A scalable approach to resolving variants of uncertain significance.

bioRxiv : the preprint server for biology·2026

Related Experiment Video

Updated: Mar 10, 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

7.6K

Improving diagnostic recognition of primary hyperparathyroidism with machine learning.

Yash R Somnay1, Mark Craven2, Kelly L McCoy3

  • 1Section of Endocrine Surgery, Department of Surgery, University of Wisconsin, Madison, WI.

Surgery
|December 20, 2016
PubMed
Summary

Machine learning accurately diagnoses primary hyperparathyroidism, even in mild cases, by analyzing clinical and lab data. This technology can improve recognition of this under-diagnosed endocrine disorder.

More Related Videos

Establishment of a Simple and Effective Rat Model for Intraoperative Parathyroid Gland Imaging
07:12

Establishment of a Simple and Effective Rat Model for Intraoperative Parathyroid Gland Imaging

Published on: August 17, 2022

5.1K
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

Related Experiment Videos

Last Updated: Mar 10, 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

7.6K
Establishment of a Simple and Effective Rat Model for Intraoperative Parathyroid Gland Imaging
07:12

Establishment of a Simple and Effective Rat Model for Intraoperative Parathyroid Gland Imaging

Published on: August 17, 2022

5.1K
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

Area of Science:

  • Endocrinology
  • Medical Informatics
  • Machine Learning

Background:

  • Primary hyperparathyroidism (PHP) is often under-recognized, leading to only 50% of patients being referred for parathyroidectomy, the only cure.
  • Diagnosis can be challenging, especially in mild cases with subtle biochemical indicators.
  • Machine learning (ML) offers a potential solution by building predictive algorithms from data.

Purpose of the Study:

  • To evaluate the ability of machine learning algorithms to distinguish primary hyperparparathyroidism from normal physiology.
  • To assess the accuracy of ML models in diagnosing PHP using clinical and laboratory data.
  • To determine if ML can aid in the recognition of under-diagnosed primary hyperparathyroidism.

Main Methods:

  • A retrospective cohort study involving 11,830 patients (6,777 with PHP, 5,053 controls) managed at endocrine surgery programs.
  • Utilized a labeled training set and 10-fold cross-validation to evaluate ML model accuracy.
  • Tested various algorithms, including Bayesian networks and ensembles, using the Weka platform, with predictors like age, sex, calcium, phosphate, parathyroid hormone, vitamin D, and creatinine.

Main Results:

  • Bayesian network models achieved 95.2% accuracy (AUC=0.989) in classifying PHP.
  • Omitting parathyroid hormone did not significantly reduce accuracy (AUC=0.985).
  • For mild PHP, Bayesian networks correctly identified 71.1% of patients with normal calcium and 92.1% with normal parathyroid hormone. Combined Bayesian networking and AdaBoost improved accuracy to 97.2% for all PHP cases and 91.9% for mild PHP.

Conclusions:

  • Machine learning models can accurately diagnose primary hyperparathyroidism without human input, even in mild presentations.
  • The integration of ML tools into electronic medical record systems could enhance the identification of this under-diagnosed condition.
  • ML demonstrates significant potential to improve the diagnostic pathway for primary hyperparathyroidism.