[2020/07/08] :boom: (Updated by Xiaoxue Gao)
[2020/06/29] :boom: (Updated by Xiaoxue Gao)
./best_model/
.[2020/06/10] :boom: (Updated by Xiaoxue Gao)
./New_data/6.10/
.[2020/06/02] :boom: (Updated by Mengshuang He)
./New_data/6.2/
.[2020/05/27] :boom: (Updated by Xiaoxue Gao)
./New_data/5.27/
.Update the model of Online Diagnosis System, the performance is as follows:
Date | ACC | AUC | F1 | Recall |
---|---|---|---|---|
2020/05/27 | 90.4 | 96.2 | 90.1 | 95.1 |
[2020/05/18] :boom: (Updated by Xiaoxue Gao)
./New_data/5.18/
../New_data/New_COVIDCT_meta(update to 5.18).csv
and ./New_data/New_NonCOVIDCT_meta(update to 5.18).csv
.[2020/05/16] :boom: (Updated by Mengshuang He)
./New_data/5.16/
../New_data/New_COVIDCT_meta(update to 5.16).csv
and ./New_data/New_NonCOVIDCT_meta(update to 5.16).csv
.[2020/05/14] :boom: (Updated by Xiaoxue Gao)
./New_data/5.14/
../New_data/New_COVIDCT_meta(update to 5.14).csv
and ./New_data/New_NonCOVIDCT_meta(update to 5.14).csv
.Update the model of Online Diagnosis System, the performance is as follows:
Date | ACC | AUC | F1 | Recall |
---|---|---|---|---|
2020/05/14 | 88.1 | 92.9 | 87.9 | 86.2 |
[2020/05/13] :boom: (Updated by Xiaoxue Gao)
./New_data/5.13/
../New_data/New_NonCOVIDCT_meta(update to 5.13).csv
.[2020/05/11] :boom: (Uploaded by Mengshuang He)
./data_split/COVID-CT-MetaInfo_new.csv
. [2020/05/07] Create repository.
Chest (computed tomography) CT scanning is one of the most important technologies for COVID-19 diagnosis in the current clinical practice, which motivates more concerted efforts in developing AI-based diagnostic tools to alleviate the enormous burden on the medical system. We develop a lesion-attention deep neural network (LA-DNN) to predict COVID-19 positive or negative with a richly annotated chest CT image dataset. The CT image dataset contains 746 public chest CT images of COVID-19 patients collected from over 760 preprints, and the data annotations are accompanied with the textual radiology reports. We extract two types of important information from these annotations: One is the flag of whether an image indicates a positive or negative case of COVID-19, and the other is the description of five lesions on the CT images associated with the positive cases. The proposed data-driven LA-DNN model focuses on the primary task of binary classification for COVID-19 diagnosis, while an auxiliary multi-label learning task is implemented simultaneously to draw the model's attention to the five lesions of COVID-19 during the training. The joint task learning process makes it a highly sample-efficient deep model that can learn COVID-19 radiology features effectively with very limited samples. Our code is public in ./Code/
.
Figure 1. The architecture of the proposed lesion-attention deep neural networks.
We used this public dataset: "COVID-CT-Dataset: a CT scan dataset about COVID-19." arXiv, 2020.
./data_split/COVID-CT-MetaInfo_new.csv
../data_split/train_meta.csv
, ./data_split/val_meta.csv
, ./data_split/test_meta.csv
.We will keep collecting new CT images for both COVID-19 and NonCOVID-19.
./New_data/
../New_data/
. Suppose we add new CT images added on May 14, then the path will be:./New_data/5.14/
. The positive and negative samples are separately stored with two zip files with names ./New_data/5.14/5.14_covidct.zip
and ./New_data/5.14/5.14_nocovidct.zip
respectively../New_data/New_COVIDCT_meta(update to 5.14).csv
and ./New_data/New_NonCOVIDCT_meta(update to 5.14).csv
.An online system has been developed for fast online diagnoses using CT images at the web address https://www.covidct.cn/.
:satisfied: Welcome to visit!!! :satisfied:
Figure 2. Navigation bar of Online Diagnosis System.
Figure 3. Performance of our proposed LA-DNN model for COVID-19 diagnosis in comparison with the baseline (Left: ROC curves, Right: Precision-recall curves; This is the latest result after adding new data).
Figure 4. Grad-CAM++ visualization for the baseline and our LA-DNN model with the backbone net of DenseNet-169 (Column 1 represents the original CT scans; Columns 2 and 3 are the class activation maps of the baseline; Columns 4 and 5 are the class activation maps of our LA-DNN model).
Figure 5. Plots of the pairwise relationships among the five lesions on making the final binary classification of COVID-19.
We uploaded a video on YouTube to introduce our project. The web address is https://youtu.be/MripiZ-1pHU.
The details of our model can be found in this preprint: Online COVID-19 diagnosis with chest CT images: Lesion-attention deep neural networks.
Please cite our paper if you find this work useful:
@inproceedings{liu2020fast,
title={A fast online COVID-19 diagnostic system with chest CT scans},
author={Liu, Bin and Gao, Xiaoxue and He, Mengshuang and Liu, Lin and Yin, Guosheng},
booktitle={Proceedings of KDD},
volume={2020},
year={2020}
}
@article {Liu2020.05.11.20097907,
author = {Liu, Bin and Gao, Xiaoxue and He, Mengshuang and Lv, Fengmao and Yin, Guosheng},
title = {Online COVID-19 diagnosis with chest CT images: Lesion-attention deep neural networks},
year = {2020},
doi = {10.1101/2020.05.11.20097907},
publisher = {Cold Spring Harbor Laboratory Press},
URL = {https://www.medrxiv.org/content/early/2020/05/14/2020.05.11.20097907},
eprint = {https://www.medrxiv.org/content/early/2020/05/14/2020.05.11.20097907.full.pdf},
journal = {medRxiv},
}