Diabetic retinopathy is a main cause of blindness in diabetic patients; therefore, detection and treatment of diabetic retinopathy at an early stage has an important effect on delaying and avoiding vision loss. In this paper, we propose a feasible solution for diabetic retinopathy classification using ResNet as the backbone network. By modifying the structure of the residual blocks and improving the downsampling level, we can increase the feature information of the hidden layer feature maps. In addition, attention mechanism is utilized to enhance the feature extraction effect. The experimental results show that the proposed model can effectively detect and classify diabetic retinopathy and achieve better results than the original model.

No abstract

Functional near-infrared spectroscopy (fNIRS) is a valuable non-invasive tool for monitoring brain activity. The classification of fNIRS data in relation to conscious activity holds significance for advancing our understanding of the brain and facilitating the development of brain-computer interfaces (BCI). Many researchers have turned to deep learning to tackle the classification challenges inherent in fNIRS data due to its strong generalization and robustness. In the application of fNIRS, reliability is really important, and one mathematical formulation of the reliability of confidence is calibration. However, many researchers overlook the important issue of calibration. To address this gap, we propose integrating calibration into fNIRS field and assess the reliability of existing models. Surprisingly, our results indicate poor calibration performance in many proposed models. To advance calibration development in the fNIRS field, we summarize three practical tips. Through this letter, we hope to emphasize the critical role of calibration in fNIRS research and argue for enhancing the reliability of deep learning-based predictions in fNIRS classification tasks. All data from our experimental process are openly available on GitHub § .