ECG-Byte: A Tokenizer for End-to-End Generative Electrocardiogram Language Modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Han, William, Duan, Chaojing, Rosenberg, Michael A., Liu, Emerson, Zhao, Ding
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918107500511232
author Han, William
Duan, Chaojing
Rosenberg, Michael A.
Liu, Emerson
Zhao, Ding
author_facet Han, William
Duan, Chaojing
Rosenberg, Michael A.
Liu, Emerson
Zhao, Ding
contents Large Language Models (LLMs) have demonstrated exceptional versatility across domains, including applications to electrocardiograms (ECGs). A growing body of work focuses on generating text from multi-channeled ECG signals and corresponding textual prompts. Existing approaches often involve a two-stage process: pretraining an ECG-specific encoder with a self-supervised learning (SSL) objective, followed by finetuning an LLM for natural language generation (NLG) using encoder-derived features. However, these methods face two key limitations: inefficiency due to multi-stage training and challenges in interpreting encoder-generated features. To overcome these issues, we propose ECG-Byte, an adapted byte pair encoding (BPE) tokenizer pipeline for autoregressive language modeling of ECGs. ECG-Byte compresses and encodes ECG signals into tokens, enabling direct end-to-end LLM training by combining ECG and text tokens. This approach enhances interpretability, as ECG tokens can be directly mapped back to the original signals. Leveraging ECG-Byte, we achieve competitive NLG performance while training 3 times faster and using just 48\% of the data required by traditional two-stage methods.
format Preprint
id arxiv_https___arxiv_org_abs_2412_14373
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ECG-Byte: A Tokenizer for End-to-End Generative Electrocardiogram Language Modeling
Han, William
Duan, Chaojing
Rosenberg, Michael A.
Liu, Emerson
Zhao, Ding
Computation and Language
Signal Processing
I.2.7; J.3
Large Language Models (LLMs) have demonstrated exceptional versatility across domains, including applications to electrocardiograms (ECGs). A growing body of work focuses on generating text from multi-channeled ECG signals and corresponding textual prompts. Existing approaches often involve a two-stage process: pretraining an ECG-specific encoder with a self-supervised learning (SSL) objective, followed by finetuning an LLM for natural language generation (NLG) using encoder-derived features. However, these methods face two key limitations: inefficiency due to multi-stage training and challenges in interpreting encoder-generated features. To overcome these issues, we propose ECG-Byte, an adapted byte pair encoding (BPE) tokenizer pipeline for autoregressive language modeling of ECGs. ECG-Byte compresses and encodes ECG signals into tokens, enabling direct end-to-end LLM training by combining ECG and text tokens. This approach enhances interpretability, as ECG tokens can be directly mapped back to the original signals. Leveraging ECG-Byte, we achieve competitive NLG performance while training 3 times faster and using just 48\% of the data required by traditional two-stage methods.
title ECG-Byte: A Tokenizer for End-to-End Generative Electrocardiogram Language Modeling
topic Computation and Language
Signal Processing
I.2.7; J.3
url https://arxiv.org/abs/2412.14373