MIEO: encoding clinical data to enhance cardiovascular event prediction
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909841524523008 |
|---|---|
| author | Borghini, Davide Marchi, Davide Nardone, Angelo Scerra, Giordano Galfrè, Silvia Giulia Pingitore, Alessandro Prencipe, Giuseppe Priami, Corrado Sîrbu, Alina |
| author_facet | Borghini, Davide Marchi, Davide Nardone, Angelo Scerra, Giordano Galfrè, Silvia Giulia Pingitore, Alessandro Prencipe, Giuseppe Priami, Corrado Sîrbu, Alina |
| contents | As clinical data are becoming increasingly available, machine learning methods have been employed to extract knowledge from them and predict clinical events. While promising, approaches suffer from at least two main issues: low availability of labelled data and data heterogeneity leading to missing values. This work proposes the use of self-supervised auto-encoders to efficiently address these challenges. We apply our methodology to a clinical dataset from patients with ischaemic heart disease. Patient data is embedded in a latent space, built using unlabelled data, which is then used to train a neural network classifier to predict cardiovascular death. Results show improved balanced accuracy compared to applying the classifier directly to the raw data, demonstrating that this solution is promising, especially in conditions where availability of unlabelled data could increase. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_11257 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MIEO: encoding clinical data to enhance cardiovascular event prediction Borghini, Davide Marchi, Davide Nardone, Angelo Scerra, Giordano Galfrè, Silvia Giulia Pingitore, Alessandro Prencipe, Giuseppe Priami, Corrado Sîrbu, Alina Machine Learning Quantitative Methods As clinical data are becoming increasingly available, machine learning methods have been employed to extract knowledge from them and predict clinical events. While promising, approaches suffer from at least two main issues: low availability of labelled data and data heterogeneity leading to missing values. This work proposes the use of self-supervised auto-encoders to efficiently address these challenges. We apply our methodology to a clinical dataset from patients with ischaemic heart disease. Patient data is embedded in a latent space, built using unlabelled data, which is then used to train a neural network classifier to predict cardiovascular death. Results show improved balanced accuracy compared to applying the classifier directly to the raw data, demonstrating that this solution is promising, especially in conditions where availability of unlabelled data could increase. |
| title | MIEO: encoding clinical data to enhance cardiovascular event prediction |
| topic | Machine Learning Quantitative Methods |
| url | https://arxiv.org/abs/2510.11257 |