Related Experiment Videos

Multimodal Machine Learning Integrating Clinical and Proteomic Data for Early Prediction of Hypertensive

Yuan Fei1, Siwei Liu2, Tianlang Tong3

  • 1Center for Single-Cell Omics, School of Public Health Shanghai Jiao Tong University School of Medicine Shanghai China.

Journal of the American Heart Association
|June 15, 2026
PubMed

Insights

Machine learning models integrating clinical and proteomic data can predict hypertensive complications early. Specific proteins like Growth/differentiation factor 15 show significant associations, aiding risk prediction and intervention.

Area of Science:

  • Cardiovascular Research
  • Proteomics
  • Machine Learning in Medicine

Background:

  • Hypertension is a major risk factor for cardiovascular, cerebrovascular, and renal diseases.
  • Early prediction of hypertensive complications is crucial for improving patient prognosis and quality of life.

Purpose of the Study:

  • To develop and validate multimodal machine learning models for early prediction of hypertensive complications.
  • To integrate clinical and proteomic features for enhanced predictive accuracy.

Main Methods:

  • Analysis of 502,166 participants from the UK Biobank.
  • Proteomic profiling using Olink Explore platform and extraction of clinical data from electronic health records.
  • Feature selection via Cox models and gradient boosting, with model construction using random forest and interpretation via Shapley Additive Explanations.

Main Results:

  • Multimodal models achieved AUCs of 0.73 for heart disease, 0.83 for brain disease, and 0.79 for kidney disease.
  • Identified key proteins including Growth/differentiation factor 15, adaptor protein 3 complex subunit σ-2, and tumor necrosis factor receptor superfamily member 10B associated with complications.
  • Demonstrated significant predictive power for clinical progression of hypertensive complications (all P<0.001).

Conclusions:

  • Multimodal machine learning models integrating proteomic and clinical data effectively identify hypertensive complications early.
  • Growth/differentiation factor 15, adaptor protein 3 complex subunit σ-2, and tumor necrosis factor receptor superfamily member 10B are potential biomarkers for risk prediction and early intervention.
Abstract

Keywords:
hypertensive complicationmachine learningmultimodalprognosisproteomic

Related Concept Videos

Hypertension III: Clinical Manifestations and Diagnostic Studies01:30

Hypertension III: Clinical Manifestations and Diagnostic Studies

Hypertension is asymptomatic and also referred to as the "silent killer" until it progresses to a severe stage or causes target organ disease. Patients may experience symptoms stemming from the strain on blood vessels and tissues in various organs or the heart's increased workload.Physical exams might show no abnormalities other than high blood pressure. Signs of vascular damage, when present, correspond to the organs supplied by the affected vessels, leading to target organ damage. For...
Hypertension and Regulation of Blood Pressure01:18

Hypertension and Regulation of Blood Pressure

Hypertension, the most common cardiovascular disease, is diagnosed through repeated measurements of elevated blood pressure. Its risks, including damage to the kidney, heart, and brain, are directly proportional to blood pressure levels. Starting from 115/75 mm Hg, the risk of cardiovascular disease doubles with each increment of 20/10 mm Hg. The diagnosis relies on blood pressure measurements, not on patient symptoms, as hypertension is often asymptomatic until end-organ damage is imminent or...