Related Experiment Video
Updated: Jun 30, 2025
Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
Published on: August 9, 2024
Multiple instance learning framework can facilitate explainability in murmur detection
Maurice Rohr1, Benedikt Müller1, Sebastian Dill1
1KIS*MED - AI Systems in Medicine, Technische Universität Darmstadt, Darmstadt, Germany.
Insights
This study introduces a novel explainable multitask model using multiple instance learning (MIL) to detect heart murmurs from phonocardiograms (PCGs). The model accurately predicts murmurs and clinical outcomes, improving cardiovascular disease diagnosis.
Area of Science:
- Cardiology
- Biomedical Engineering
- Machine Learning
Background:
- Cardiovascular diseases (CVDs) are a leading cause of global mortality.
- Heart murmurs detected via phonocardiograms (PCGs) can indicate CVDs but often require expert interpretation.
- Current methods may overlook subtle murmur indicators or lack explainability.
Purpose of the Study:
- To develop an explainable multitask model for predicting heart murmurs and clinical outcomes from multiple PCG recordings.
- To leverage multiple instance learning (MIL) for improved murmur detection and localization.
- To integrate explainable features for enhanced clinical decision support.
Main Methods:
- A two-stage multitask model incorporating MIL in the first stage for murmur detection in single PCGs.
- Fusion of explainable hand-crafted features with features from a pooling-based artificial neural network (PANN) in the second stage.
- Prediction of patient-specific murmur presence and clinical outcome using multiple PCG recordings via a feed-forward neural network.
Main Results:
- The MIL approach effectively identifies murmur locations and provides useful features for PCG analysis.
- The PANN model achieved a weighted accuracy of 0.714 on the CirCor dataset.
- The model demonstrated competitive classification performance for murmur detection and clinical outcome prediction.
Conclusions:
- This work is the first to demonstrate the utility of MIL for phonocardiogram classification.
- The study highlights a method for quantitative analysis of model explainability, mitigating confirmation bias.
- The findings underscore the value of combining MIL with handcrafted features for explainable AI in cardiovascular diagnostics.
Objective:
Cardiovascular diseases (CVDs) account for a high fatality rate worldwide. Heart murmurs can be detected from phonocardiograms (PCGs) and may indicate CVDs. Still, they are often overlooked as their detection and correct clinical interpretation require expert skills. In this work, we aim to predict the presence of murmurs and clinical outcomes from multiple PCG recordings employing an explainable multitask model.
Approach:
Our approach consists of a two-stage multitask model. In the first stage, we predict the murmur presence in single PCGs using a multiple instance learning (MIL) framework. MIL also allows us to derive sample-wise classifications (i.e. murmur locations) while only needing one annotation per recording ("weak label") during training. In the second stage, we fuse explainable hand-crafted features with features from a pooling-based artificial neural network (PANN) derived from the MIL framework. Finally, we predict the presence of murmurs and the clinical outcome for a single patient based on multiple recordings using a simple feed-forward neural network.
Main Results:
We show qualitatively and quantitatively that the MIL approach yields useful features and can be used to detect murmurs on multiple time instances and may thus guide a practitioner through PCGs. We analyze the second stage of the model in terms of murmur classification and clinical outcome. We achieved a weighted accuracy of 0.714 and an outcome cost of 13612 when using the PANN model and demographic features on the CirCor dataset (hidden test set of the George B. Moody PhysioNet challenge 2022, team "Heart2Beat", rank 12 / 40).
Significance:
To the best of our knowledge, we are the first to demonstrate the usefulness of MIL in PCG classification. Also, we showcase how the explainability of the model can be analyzed quantitatively, thus avoiding confirmation bias inherent to many post-hoc methods. Finally, our overall results demonstrate the merit of employing MIL combined with handcrafted features for the generation of explainable features as well as for a competitive classification performance.
More Related Videos
08:05Measuring Statistical Learning Across Modalities and Domains in School-Aged Children Via an Online Platform and Neuroimaging Techniques
Published on: June 30, 2020
03:14Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
