a/README.md b/README.md
...
...
4
learning elegantly guides classification (bottom-top path) to learn better. In this study, our networks learn to 
4
learning elegantly guides classification (bottom-top path) to learn better. In this study, our networks learn to 
5
classify early-stage lung cancer from **CT** scans on **pathological** level. The deep learning models outperforms the 
5
classify early-stage lung cancer from **CT** scans on **pathological** level. The deep learning models outperforms the 
6
radiologists (2 senior and 2 junior) in our observer study, which indicates the potentials to facilitate
6
radiologists (2 senior and 2 junior) in our observer study, which indicates the potentials to facilitate
7
precision medicine.
7
precision medicine.
8
8
9
![Graphical Abstract](GraphicalAbstract.png)
10
11
More details, please refer to our paper:
9
More details, please refer to our paper:
12
10
13
**3D Deep Learning from CT Scans Predicts Tumor Invasiveness of Subcentimeter Pulmonary Adenocarcinomas**
11
**3D Deep Learning from CT Scans Predicts Tumor Invasiveness of Subcentimeter Pulmonary Adenocarcinomas**
14
12
15
Wei Zhao<sup>†</sup>, Jiancheng Yang<sup>†</sup>, Yingli Sun, Cheng Li, Weilan Wu, Liang Jin, Zhiming Yang, Bingbing Ni, Pan Gao, Peijun Wang, Yanqing Hua and Ming Li (<sup>†</sup>indicates equal contribution)
13
Wei Zhao<sup>†</sup>, Jiancheng Yang<sup>†</sup>, Yingli Sun, Cheng Li, Weilan Wu, Liang Jin, Zhiming Yang, Bingbing Ni, Pan Gao, Peijun Wang, Yanqing Hua and Ming Li (<sup>†</sup>indicates equal contribution)