Related Experiment Video
Updated: Sep 7, 2025

Cutoff Value of Phase Angle by Bioelectrical Impedance Analysis at Admission as a Prognostic Factor in Patients with Acute Heart Failure
Published on: June 10, 2025
Dynamic risk stratification using Markov chain modelling in patients with chronic heart failure
Syed Kazmi1,2, Chandrasekhar Kambhampati2, John G F Cleland3
1Department of Academic Cardiology, Hull University Teaching Hospital NHS Trust, Hull, UK.
This study introduces an AI-powered Markov chain model for predicting chronic heart failure (CHF) patient outcomes. The model accurately forecasts future events, showing CHF progression is more predictable than previously thought.
Area of Science:
- Cardiology
- Artificial Intelligence
- Health Informatics
Background:
- Chronic heart failure (CHF) patient risk changes dynamically with disease progression and treatment.
- Accurate risk stratification is crucial for managing CHF patients effectively.
Purpose of the Study:
- To develop and validate a dynamic risk stratification Markov chain model using artificial intelligence for patients with chronic heart failure (CHF).
- To assess the predictability of CHF patient outcomes over time.
Main Methods:
- A Markov chain model was developed using AI, analyzing data from 7496 patients assessed for heart failure (HF).
- Health states (death, hospitalization, outpatient visit, no event, leaving service) were assessed every 4 months.
- Model probabilities were derived from the first two transitions (0-8 months) and validated against observed figures.
Main Results:
- The AI-driven model demonstrated strong predictive accuracy for future events in CHF patients.
- Model predictions for cumulative death probability at cycle 4 (14%) closely matched observed figures (14%).
- Over two years, the model predicted a 0.19 probability of dying, closely aligning with the observed 0.18.
Conclusions:
- A Markov chain model, informed by the initial 8 months of follow-up data, reliably predicts future events in chronic heart failure patients.
- The study suggests that the clinical course of CHF is more linear and predictable than commonly assumed.
Related Concept Videos
Heart Failure IV: Classification and Diagnostic Evaluation
Heart Failure II: Pathophysiology
Pathophysiology of Heart Failure
Heart Failure III: Clinical Manifestations
Heart Failure V: Medical Management
Heart Failure I: Introduction

