Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
November 28, 2023PLoS ONEOpen Access

The proposed model predicted the need for advanced therapies with a mean AUC of 0.747 (SD 0.080), AUPRC of 0.642 (SD 0.080), and F1 score of 0.569 (SD 0.067).

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Timely referral for advanced therapies is critical for heart failure patients, but an interpretable clinical decision-making system using single-hospitalization EHR data to predict this need was needed.

Can an interpretable tropical geometry-based fuzzy neural network predict the need for advanced heart failure therapies in patients with LVEF ≤ 35%?

Comparison

Fuzzy logic and tropical geometry machine learning model vs other machine learning methods

Design

Retrospective EHR-based machine learning prediction model development and validation study

Authors

YZYufeng ZhangKAKeith D. AaronsonJGJonathan Gryak

Discussion

Loading...

Member takes

Overview

May aid risk stratification in HFrEF; hypothesis-generating for interpretable ML, requires prospective validation.

Structured PICO

Can an interpretable tropical geometry-based fuzzy neural network predict the need for advanced heart failure therapies in patients with LVEF ≤ 35%?

P
Population
300 heart failure patients (557 hospitalizations) from Michigan Medicine (2013-2021) with LVEF ≤ 35% and at least two heart failure hospitalizations within one year.
I
Intervention
Interpretable tropical geometry-based fuzzy neural network model
C
Comparator
Other machine learning methods
O
Outcome
Need for advanced therapies (heart transplantation, left ventricular assist device) at the subsequent hospitalizationhard clinical

An interpretable machine learning model using fuzzy logic and tropical geometry can predict the need for advanced heart failure therapies with good accuracy and transparent clinical rules.

Limitations

  • further research is needed to prospectively validate the risk factors identified by the model

Cite This Study

Zhang et al. (2023) studied this question.

synapsesocial.com/papers/6a1088f0d13714ec96ffff4dhttps://doi.org/10.1371/journal.pone.0295016
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Identifying potential candidates for advanced heart failure therapies using an interpretable machine learning algorithm2022 · 9 citations
  2. 2Enhancing heart failure treatment decisions: interpretable machine learning models for advanced therapy eligibility prediction using EHR data2024 · 4 citations
  3. 3COMPARISON OF MACHINE LEARNING MODELS IN HEART FAILURE PREDICTION AND THEIR INTEGRATION INTO CLINICAL DECISION SUPPORT SYSTEMS2025 · 2 citations
  4. 4Machine Learning–Driven Models to Predict Prognostic Outcomes in Patients Hospitalized With Heart Failure Using Electronic Health Records: Retrospective Study2021 · 36 citations
  5. 5Integrating machine learning techniques for enhanced prognostic modeling of heart failure risk in the american population2025 · 1 citations