Segmentation and Recognition of the Pathological Features of Squamous Cell Carcinoma of the Skin Based on Multispectral Imaging.

Cutaneous squamous cell carcinoma (cSCC) is one of the most common skin cancers, a definitive diagnosis of cSCC is crucial to prevent patients from missing out on treatment. The gold standard for the diagnosis of cSCC is still pathological biopsy. Currently, its diagnostic efficiency and accuracy largely depend on the experience of pathologists. Here, we present a simple, fast, and robust technique, a microscopic multispectral imaging system based on LED illumination, to diagnose cSCC qualitatively and quantitatively. The adaptive threshold segmentation method was used to segment the multispectral images into characteristic structures. There was a statistically significant difference between the average nucleocytoplasmic ratio of normal skin (4.239%) and cSCC tissues (15.607%) (p < 0.01), and the keratin pearls cSCC have well-defined qualitative features. These results show that the qualitative and quantitative features obtained from multispectral imaging can be used to comprehensively determine whether or not the tissue is cancerous. This work has significant implications for the development of a low-cost and easy-to-use device, which can not only reduce the complexity of pathological diagnosis but can also achieve the goal of convenient digital staining and access to critical histological information.

Robust wave-feature adaptive heartbeat classification based on self-attention mechanism using a transformer model.

Objective. Electrocardiography is a common method for screening cardiovascular diseases. Accurate heartbeat classification assists in diagnosis and has attracted great attention. In this paper, we proposed an automatic heartbeat classification method based on a transformer neural network using a self-attention mechanism.Approach.An adaptive heartbeat segmentation method was designed to selectively focus on the time-dependent representation of heartbeats. A one-dimensional convolution layer was used to embed wave characteristics into symbolic representations, and then, a transformer block using multi-head attention was applied to deal with the dependence of wave-embedding. The model was trained and evaluated using the MIT-BIH arrhythmia database (MIT-DB). To improve the model performance, the model pre-trained on MIT-BIH supraventricular arrhythmia database (MIT-SVDB) was used and fine-tuned on MIT-DB.Main results.The proposed method was verified using the MIT-DB for two groups. In the first group, our method attained F1 scores of 0.86 and 0.96 for the supraventricular ectopic beat class and ventricular ectopic beat class, respectively. In the second group, our method achieved an average F1 value of 99.83% and better results than other state-of-the-art methods.Significance.We proposed a novel heartbeat classification method based on a transformer model. This method provides a new solution for real-time electrocardiogram heartbeat classification, which can be applied to wearable devices.