Pre-training of Graph Augmented Transformers for Medication Recommendation

Last update: Dec 27, 2022

Related tags

Overview

G-Bert

Pre-training of Graph Augmented Transformers for Medication Recommendation

Intro

G-Bert combined the power of Graph Neural Networks and BERT (Bidirectional Encoder Representations from Transformers) for medical code representation and medication recommendation. We use the graph neural networks (GNNs) to represent the structure information of medical codes from a medical ontology. Then we integrate the GNN representation into a transformer-based visit encoder and pre-train it on single-visit EHR data. The pre-trained visit encoder and representation can be fine-tuned for downstream medical prediction tasks. Our model is the first to bring the language model pre-training schema into the healthcare domain and it achieved state-of-the-art performance on the medication recommendation task.

Requirements

pytorch>=0.4
python>=3.5
torch_geometric==1.0.3

Guide

We list the structure of this repo as follows:

.
├── [4.0K]  code/
│   ├── [ 13K]  bert_models.py % transformer models
│   ├── [5.9K]  build_tree.py % build ontology
│   ├── [4.3K]  config.py % hyperparameters for G-Bert
│   ├── [ 11K]  graph_models.py % GAT models
│   ├── [   0]  __init__.py
│   ├── [9.8K]  predictive_models.py % G-Bert models
│   ├── [ 721]  run_alternative.sh % script to train G-Bert
│   ├── [ 19K]  run_gbert.py % fine tune G-Bert
│   ├── [ 19K]  run_gbert_side.py
│   ├── [ 18K]  run_pretraining.py % pre-train G-Bert
│   ├── [4.4K]  run_tsne.py # output % save embedding for tsne visualization
│   └── [4.7K]  utils.py
├── [4.0K]  data/
│   ├── [4.9M]  data-multi-side.pkl 
│   ├── [3.6M]  data-multi-visit.pkl % patients data with multi-visit
│   ├── [4.3M]  data-single-visit.pkl % patients data with singe-visit
│   ├── [ 11K]  dx-vocab-multi.txt % diagnosis codes vocabulary in multi-visit data
│   ├── [ 11K]  dx-vocab.txt % diagnosis codes vocabulary in all data
│   ├── [ 29K]  EDA.ipynb % jupyter version to preprocess data
│   ├── [ 18K]  EDA.py % python version to preprocess data
│   ├── [6.2K]  eval-id.txt % validation data ids
│   ├── [6.9K]  px-vocab-multi.txt % procedure codes vocabulary in multi-visit data
│   ├── [ 725]  rx-vocab-multi.txt % medication codes vocabulary in multi-visit data
│   ├── [2.6K]  rx-vocab.txt % medication codes vocabulary in all data
│   ├── [6.2K]  test-id.txt % test data ids
│   └── [ 23K]  train-id.txt % train data ids
└── [4.0K]  saved/
    └── [4.0K]  GBert-predict/ % model files to reproduce our result
        ├── [ 371]  bert_config.json 
        └── [ 12M]  pytorch_model.bin

Preprocessing Data

We have released the preprocessing codes named data/EDA.ipynb to process data using raw files from MIMIC-III dataset. You can download data files from MIMIC and get necessary mapping files from GAMENet.

Quick Test

To validate the performance of G-Bert, you can run the following script since we have provided the trained model binary file and well-preprocessed data.

cd code/
python run_gbert.py --model_name GBert-predict --use_pretrain --pretrain_dir ../saved/GBert-predict --graph

Cite

Please cite our paper if you find this code helpful:

@article{shang2019pre,
  title={Pre-training of Graph Augmented Transformers for Medication Recommendation},
  author={Shang, Junyuan and Ma, Tengfei and Xiao, Cao and Sun, Jimeng},
  journal={arXiv preprint arXiv:1906.00346},
  year={2019}
}

Acknowledgement

Many thanks to the open source repositories and libraries to speed up our coding progress.

Pre-training of Graph Augmented Transformers for Medication Recommendation

Related tags

Overview

G-Bert

Intro

Requirements

Guide

Preprocessing Data

Quick Test

Cite

Acknowledgement

Owner

AI Toolkit for Healthcare Imaging

Geometric Vector Perceptrons --- a rotation-equivariant GNN for learning from biomolecular structure

ERISHA is a mulitilingual multispeaker expressive speech synthesis framework. It can transfer the expressivity to the speaker's voice for which no expressive speech corpus is available.

Fast and Simple Neural Vocoder, the Multiband RNNMS

Deep Learning Interviews book: Hundreds of fully solved job interview questions from a wide range of key topics in AI.

Trained on Simulated Data, Tested in the Real World

Registration Loss Learning for Deep Probabilistic Point Set Registration

Official TensorFlow code for the forthcoming paper

Tensorflow implementation of soft-attention mechanism for video caption generation.

CLIP (Contrastive Language–Image Pre-training) trained on Indonesian data

BlueFog Tutorials

Official implementation of ACTION-Net: Multipath Excitation for Action Recognition (CVPR'21).

A Pytorch Implementation of Domain adaptation of object detector using scissor-like networks

This is Unofficial Repo. Lips Don't Lie: A Generalisable and Robust Approach to Face Forgery Detection (CVPR 2021)

Yolov5 deepsort inference，使用YOLOv5+Deepsort实现车辆行人追踪和计数，代码封装成一个Detector类，更容易嵌入到自己的项目中

Iranian Cars Detection using Yolov5s, PyTorch

Exploiting Robust Unsupervised Video Person Re-identification

Code and datasets for TPAMI 2021

pytorch implementation of fast-neural-style

Resources for our AAAI 2022 paper: "LOREN: Logic-Regularized Reasoning for Interpretable Fact Verification".