Skip to content

Latest commit

 

History

History
35 lines (21 loc) · 883 Bytes

README.md

File metadata and controls

35 lines (21 loc) · 883 Bytes

Code for fine-grained-ecg

The official code for "Accurate diagnoses of 12-lead electrocardiograms using artificial intelligence: A attention aggregation multi-label model for diagnostic assistance" by Zizhu Liu, Qing Cao, Qi Jin,Jingsheng Lin, Gang Lv and Kang Chen This project test in Tensorflow and contains model only.

Description

model overall

model

attention maps

model

aggregation layer

model

Example:

from my_utils.default_config import dict2config, update_and_save_cfg, CONFIG_DICT
from models import networks

# chose config file first
config_path = 'AA_multi_label.json'
cfg_dict = update_and_save_cfg(CONFIG_DICT, config_path, save_cfg=False, date=False)
cfg = dict2config(cfg_dict)

# call the model
model = networks.build_model(cfg)