SynEHRgy: Synthesizing Mixed-Type Structured Electronic Health Records using Decoder-Only Transformers

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Karami, Hojjat, Atienza, David, Ionescu, Anisoara
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909397478801408
author Karami, Hojjat
Atienza, David
Ionescu, Anisoara
author_facet Karami, Hojjat
Atienza, David
Ionescu, Anisoara
contents Generating synthetic Electronic Health Records (EHRs) offers significant potential for data augmentation, privacy-preserving data sharing, and improving machine learning model training. We propose a novel tokenization strategy tailored for structured EHR data, which encompasses diverse data types such as covariates, ICD codes, and irregularly sampled time series. Using a GPT-like decoder-only transformer model, we demonstrate the generation of high-quality synthetic EHRs. Our approach is evaluated using the MIMIC-III dataset, and we benchmark the fidelity, utility, and privacy of the generated data against state-of-the-art models.
format Preprint
id arxiv_https___arxiv_org_abs_2411_13428
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SynEHRgy: Synthesizing Mixed-Type Structured Electronic Health Records using Decoder-Only Transformers
Karami, Hojjat
Atienza, David
Ionescu, Anisoara
Machine Learning
Artificial Intelligence
Generating synthetic Electronic Health Records (EHRs) offers significant potential for data augmentation, privacy-preserving data sharing, and improving machine learning model training. We propose a novel tokenization strategy tailored for structured EHR data, which encompasses diverse data types such as covariates, ICD codes, and irregularly sampled time series. Using a GPT-like decoder-only transformer model, we demonstrate the generation of high-quality synthetic EHRs. Our approach is evaluated using the MIMIC-III dataset, and we benchmark the fidelity, utility, and privacy of the generated data against state-of-the-art models.
title SynEHRgy: Synthesizing Mixed-Type Structured Electronic Health Records using Decoder-Only Transformers
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2411.13428