AUTHOR=Yang Meiyi , He Xiaopeng , Xu Lifeng , Liu Minghui , Deng Jiali , Cheng Xuan , Wei Yi , Li Qian , Wan Shang , Zhang Feng , Wu Lei , Wang Xiaomin , Song Bin , Liu Ming 

TITLE=CT-based transformer model for non-invasively predicting the Fuhrman nuclear grade of clear cell renal cell carcinoma

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.961779

DOI=10.3389/fonc.2022.961779

ISSN=2234-943X

ABSTRACT=Background:Clear cell Renal Cell Carcinoma (ccRCC) is the most common malignant tumor in the urinary system and the predominant subtype of malignant renal tumors with high mortality. Biopsy is the main examination to determine ccRCC grade, but it can lead to unavoidable complications and sampling bias. Therefore, non-invasive technology (e.g., CT examination) for ccRCC grading attracts more and more attention. However, noise labels on CT images containing multiple grades but only one label make prediction difficult.

Aim: We proposed a Transformer-based deep learning algorithm with CT images to improve the diagnostic accuracy of grading prediction.

Methods: We propose a transformer-based network architecture that efficiently employ convolutional neural networks(CNNs) and self-attention mechanisms to acquire a persuasive feature fed into a nonlinear classifier to predict the Fuhrman nuclear grade of ccRCC. We integrate different training models to improve robustness to predict Fuhrman nuclear grade. Then, we conducted experiments on a collected ccRCC dataset containing 759 patients and used average classification accuracy, sensitivity, specificity, and Area Under Curve as indicators to evaluate the quality of research. In the comparative experiments, we further performed various current deep learning algorithms to show the advantages of the proposed method.


Results:The mean accuracy, sensitivity, specificity, and Area Under Curve achieved by CNN were 82.3\%, 89.4\%, 83.2 \%, and 85.7\%, respectively. In contrast, the proposed Transformer-based model obtains mean accuracy of 87.1\% with sensitivity of 91.3\%, specificity of 85.3\%, and an Area Under Curve (AUC) of 90.3\%. The integrated model acquires a better performance (86.5\% ACC  and AUC 91.2\%).

Conclusion:Transformer-based network performs better than traditional deep learning algorithms in terms of the accuracy of ccRCC prediction. Meanwhile, the transformer has a certain advantage in dealing with noise labels existing in CT images of ccRCC. This method is promising to be applied to other medical tasks (e.g., the grade of neurogliomas and meningiomas).