Binary classification for arrythmia detection with ECG datasets.

Overview

HEART DISEASE AI DATATHON 2021

[Eng] / [Kor]


#English

This is an AI diagnosis modeling contest that uses the heart disease echocardiography and electrocardiogram datasets for artificial intelligence learning promoted as part of the "2021 AI Learning Data Construction Project" to discriminate echocardiography/electrocardiogram diseases.

Task II. Arrythmia on ECG datasets

0. Model

Resnet-based architecture.
Best AUC-ROC Score: 0.9986926250732517

1. Installation

1.1. Environment

Python >= 3.6

1.2. Requirements:

  • tensorflow >= 2.5
  • xmltodict
  • scikit-learn
  • matplotlib
  • numpy
pip install -r requirements.txt

2. Usage

2.1. Training

  1. Basic usage
python train.py -d electrocardiogram/data/train -s model.h5
  1. Training with 8 leads inputs, elevation adjustment, data augmentation and gqussian noises
python train.py -d electrocardiogram/data/train -s model.h5 -l 8 -v -a -n

To see more options:

python train.py -h
  • options:
    • -d, --data : File path of training data
    • -s, --save : File name for saving trained model (extension should be '.h5')
    • -b, --batch : Batch size (default=500)
    • -e, --epoch : Number of epochs (default=50)
    • -l, --lead : Number of leads to be trained (2/8/12) (default=2)
    • -v, --elevation : Option for adjusting elevation
    • -a, --augmentation : Option for data augmentation (stretching & amplifying)
    • -n, --noise : Option for adding noise on data

2.2. Evaluation

  1. Basic usage
python eval.py -d electrocardiogram/data/validation -m model.h5
  1. Evaluation with the best model
python eval.py -d electrocardiogram/data/validation -m best.h5
  1. Evaluation with 12 leads inputs and elevation adjustment
python eval.py -d electrocardiogram/data/validation -m model.h5 -l 12 -v

To see more options:

python eval.py -h
  • options:
    • -d, --data : File path of validation data
    • -m, --model : File name of saved model
    • -l, --lead : Number of leads being trained (default=2) (2/8/12)
    • -v, --elevation : Option for adjusting elevation

#Korean

심초음파/심전도 ai 모델 데이터톤 2021

이 경진대회는 "2021 인공지능 학습용 데이터 구축사업"의 일환으로 추진된 인공지능 학습용 심장질환 심초음파 및 심전도 데이터셋을 이용하여 심초음파/심전도 질환을 판별하는 AI 진단 모델링 경진대회입니다.

Task II. Arrythmia on ECG datasets

심전도 데이터셋을 활용한 부정맥 진단 AI 모델 공모(심전도 데이터셋을 활용한 부정맥 진단 AI 모델 개발)

0. 모델

Resnet 구조 기반의 Binary classification model.
Best AUC-ROC Score: 0.9986926250732517

1. 설치

1.1. 환경

Python >= 3.6

1.2. 필요한 패키지:

  • tensorflow >= 2.5
  • xmltodict
  • scikit-learn
  • matplotlib
  • numpy
pip install -r requirements.txt

2. 사용법

2.1. Training

  1. 기본 사용법 예시 (제출용)
python train.py -d electrocardiogram/data/train -s model.h5
  1. 8개 리드, 상하조정, 데이터 어그멘테이션, 노이즈 적용
python train.py -d electrocardiogram/data/train -s model.h5 -l 8 -v -a -n

To see more options:

python train.py -h
  • options:
    • -d, --data : 트레이닝 데이터 경로
    • -s, --save : 학습된 모델명 (확장자 .h5로 써줄 것)
    • -b, --batch : 배치 사이즈 (default=500)
    • -e, --epoch : 에포크 수 (default=50)
    • -l, --lead : 트레이닝에 쓸 리드 수 (2/8/12) (default=2)
    • -v, --elevation : 상하 조정 옵션
    • -a, --augmentation : 데이터 어그멘테이션 옵션 (stretching & amplifying)
    • -n, --noise : 가우시안 노이즈 적용 옵션

2.2. Evaluation

  1. 기본 사용법 예시
python eval.py -d electrocardiogram/data/validation -m model.h5
  1. 체출된 Best model 평가 (제출용)
python eval.py -d electrocardiogram/data/validation -m best.h5
  1. 12개 리드, 상하조정 적용
python eval.py -d electrocardiogram/data/validation -m model.h5 -l 12 -v

To see more options:

python eval.py -h
  • options:
    • -d, --data : 벨리데이션 데이터 경로
    • -m, --model : 불러올 모델 파일명
    • -l, --lead : 트레이닝된 리드 수 (2/8/12) (default=2)
    • -v, --elevation : 상하 조정 옵션
Owner
HY_Kim
CSer in SUNY Korea.
HY_Kim
Train SN-GAN with AdaBelief

SNGAN-AdaBelief Train a state-of-the-art spectral normalization GAN with AdaBelief https://github.com/juntang-zhuang/Adabelief-Optimizer Acknowledgeme

Juntang Zhuang 10 Jun 11, 2022
Synthetic Humans for Action Recognition, IJCV 2021

SURREACT: Synthetic Humans for Action Recognition from Unseen Viewpoints Gül Varol, Ivan Laptev and Cordelia Schmid, Andrew Zisserman, Synthetic Human

Gul Varol 59 Dec 14, 2022
Official PyTorch implementation of MX-Font (Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts)

Introduction Pytorch implementation of Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Expert. | paper Song Park1

Clova AI Research 97 Dec 23, 2022
A library to inspect itermediate layers of PyTorch models.

A library to inspect itermediate layers of PyTorch models. Why? It's often the case that we want to inspect intermediate layers of a model without mod

archinet.ai 380 Dec 28, 2022
DeepLab is a state-of-art deep learning system for semantic image segmentation built on top of Caffe.

DeepLab Introduction DeepLab is a state-of-art deep learning system for semantic image segmentation built on top of Caffe. It combines densely-compute

Ali 234 Nov 14, 2022
Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation Networks

CyGNet This repository reproduces the AAAI'21 paper “Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation Network

CunchaoZ 89 Jan 03, 2023
Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation (CVPR 2021)

Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation (CVPR 2021, official Pytorch implementatio

Microsoft 247 Dec 25, 2022
A package related to building quasi-fibration symmetries

qf A package related to building quasi-fibration symmetries. If you'd like to learn more about how it works, see the brief explanation and References

Paolo Boldi 1 Dec 01, 2021
Pytorch Implementation of Spiking Neural Networks Calibration, ICML 2021

SNN_Calibration Pytorch Implementation of Spiking Neural Networks Calibration, ICML 2021 Feature Comparison of SNN calibration: Features SNN Direct Tr

Yuhang Li 60 Dec 27, 2022
Evolving neural network parameters in JAX.

Evolving Neural Networks in JAX This repository holds code displaying techniques for applying evolutionary network training strategies in JAX. Each sc

Trevor Thackston 6 Feb 12, 2022
Implementation of Kaneko et al.'s MaskCycleGAN-VC model for non-parallel voice conversion.

MaskCycleGAN-VC Unofficial PyTorch implementation of Kaneko et al.'s MaskCycleGAN-VC (2021) for non-parallel voice conversion. MaskCycleGAN-VC is the

86 Dec 25, 2022
LTR_CrossEncoder: Legal Text Retrieval Zalo AI Challenge 2021

LTR_CrossEncoder: Legal Text Retrieval Zalo AI Challenge 2021 We propose a cross encoder model (LTR_CrossEncoder) for information retrieval, re-retrie

Hieu Duong 7 Jan 12, 2022
PyTorch implementation of deep GRAph Contrastive rEpresentation learning (GRACE).

GRACE The official PyTorch implementation of deep GRAph Contrastive rEpresentation learning (GRACE). For a thorough resource collection of self-superv

Big Data and Multi-modal Computing Group, CRIPAC 186 Dec 27, 2022
Split Variational AutoEncoder

Split-VAE Split Variational AutoEncoder Introduction This repository contains and implemementation of a Split Variational AutoEncoder (SVAE). In a SVA

Andrea Asperti 2 Sep 02, 2022
Codes accompanying the paper "Learning Nearly Decomposable Value Functions with Communication Minimization" (ICLR 2020)

NDQ: Learning Nearly Decomposable Value Functions with Communication Minimization Note This codebase accompanies paper Learning Nearly Decomposable Va

Tonghan Wang 69 Nov 26, 2022
DFM: A Performance Baseline for Deep Feature Matching

DFM: A Performance Baseline for Deep Feature Matching Python (Pytorch) and Matlab (MatConvNet) implementations of our paper DFM: A Performance Baselin

143 Jan 02, 2023
Lucid Sonic Dreams syncs GAN-generated visuals to music.

Lucid Sonic Dreams Lucid Sonic Dreams syncs GAN-generated visuals to music. By default, it uses NVLabs StyleGAN2, with pre-trained models lifted from

731 Jan 02, 2023
The toolkit to generate auto labeled datasets

Ozeu Ozeu is the toolkit to autolabal dataset for instance segmentation. You can generate datasets labaled with segmentation mask and bounding box fro

Xiong Jie 28 Mar 28, 2022
Implementation of trRosetta and trDesign for Pytorch, made into a convenient package

trRosetta - Pytorch (wip) Implementation of trRosetta and trDesign for Pytorch, made into a convenient package

Phil Wang 67 Dec 17, 2022