School of Computer Science and Engineering, Dalian Minzu University, Dalian, Liaoning 116600, China.
School of Computer Science and Engineering, Dalian Minzu University, Dalian, Liaoning 116600, China.
Comput Methods Programs Biomed. 2024 Jan;243:107876. doi: 10.1016/j.cmpb.2023.107876. Epub 2023 Oct 18.
Currently, COVID-19 is a highly infectious disease that can be clinically diagnosed based on diagnostic radiology. Deep learning is capable of mining the rich information implied in inpatient imaging data and accomplishing the classification of different stages of the disease process. However, a large amount of training data is essential to train an excellent deep-learning model. Unfortunately, due to factors such as privacy and labeling difficulties, annotated data for COVID-19 is extremely scarce, which encourages us to propose a more effective deep learning model that can effectively assist specialist physicians in COVID-19 diagnosis.
In this study,we introduce Masked Autoencoder (MAE) for pre-training and fine-tuning directly on small-scale target datasets. Based on this, we propose Self-Supervised Learning with Self-Distillation on COVID-19 medical image classification (SSSD-COVID). In addition to the reconstruction loss computation on the masked image patches, SSSD-COVID further performs self-distillation loss calculations on the latent representation of the encoder and decoder outputs. The additional loss calculation can transfer the knowledge from the global attention of the decoder to the encoder which acquires only local attention.
Our model achieves 97.78 % recognition accuracy on the SARS-COV-CT dataset containing 2481 images and is further validated on the COVID-CT dataset containing 746 images, which achieves 81.76 % recognition accuracy. Further introduction of external knowledge resulted in experimental accuracies of 99.6% and 95.27 % on these two datasets, respectively.
SSSD-COVID can obtain good results on the target dataset alone, and when external information is introduced, the performance of the model can be further improved to significantly outperform other models.Overall, the experimental results show that our method can effectively mine COVID-19 features from rare data and can assist professional physicians in decision-making to improve the efficiency of COVID-19 disease detection.
目前，COVID-19 是一种高度传染性疾病，可以基于诊断影像学进行临床诊断。深度学习能够挖掘住院影像数据中隐含的丰富信息，并实现疾病进程不同阶段的分类。然而，训练一个优秀的深度学习模型需要大量的训练数据。不幸的是，由于隐私和标记困难等因素，COVID-19 的注释数据极其匮乏，这促使我们提出一种更有效的深度学习模型，以有效协助专家医生进行 COVID-19 诊断。
在本研究中，我们引入掩蔽自动编码器（MAE）进行预训练和直接在小规模目标数据集上的微调。在此基础上，我们提出了 COVID-19 医学图像分类的自监督学习与自蒸馏（SSSD-COVID）。除了对掩蔽图像块进行重建损失计算外，SSSD-COVID 还对编码器和解码器输出的潜在表示进行自蒸馏损失计算。额外的损失计算可以将知识从解码器的全局注意力转移到仅获取局部注意力的编码器。
我们的模型在包含 2481 张图像的 SARS-COV-CT 数据集上实现了 97.78%的识别准确率，并在包含 746 张图像的 COVID-CT 数据集上进一步验证，识别准确率达到 81.76%。进一步引入外部知识后，在这两个数据集上的实验准确率分别达到了 99.6%和 95.27%。
SSSD-COVID 可以仅在目标数据集上获得良好的结果，当引入外部信息时，模型的性能可以进一步提高，显著优于其他模型。总体而言，实验结果表明，我们的方法可以从稀有数据中有效地挖掘 COVID-19 特征，并可以协助专业医生进行决策，提高 COVID-19 疾病检测的效率。
