Interpretable travel distance on the county-wise COVID-19 by sequence to sequence with attention

Background: Travel restrictions as a means of intervention in the COVID-19 epidemic have reduced the spread of outbreaks using epidemiological models. We introduce the attention module in the sequencing model to assess the effects of the different classes of travel distances. Objective: To establish a direct relationship between the number of travelers for various travel distances and the COVID-19 trajectories. To improve the prediction performance of sequencing model. Setting: Counties from all over the United States. Participants: New confirmed cases and deaths have been reported in 3158 counties across the United States. Measurements: Outcomes included new confirmed cases and deaths in the 30 days preceding November 13, 2021. The daily number of trips taken by the population for various classes of travel distances and the geographical information of infected counties are assessed. Results: There is a spatial pattern of various classes of travel distances across the country. The varying geographical effects of the number of people travelling for different distances on the epidemic spread are demonstrated. Limitation: We examined data up to November 13, 2021, and the weights of each class of travel distances may change accordingly as the data evolves. Conclusion: Given the weights of people taking trips for various classes of travel distances, the epidemics could be mitigated by reducing the corresponding class of travellers.

Computational analysis of pathological image enables interpretable prediction for microsatellite instability

Microsatellite instability (MSI) is associated with several tumor types and its status has become increasingly vital in guiding patient treatment decisions. However, in clinical practice, distinguishing MSI from its counterpart is challenging since the diagnosis of MSI requires additional genetic or immunohistochemical tests. In this study, interpretable pathological image analysis strategies are established to help medical experts to automatically identify MSI. The strategies only require ubiquitous Haematoxylin and eosin-stained whole-slide images and can achieve decent performance in the three cohorts collected from The Cancer Genome Atlas. The strategies provide interpretability in two aspects. On the one hand, the image-level interpretability is achieved by generating localization heat maps of important regions based on the deep learning network; on the other hand, the feature-level interpretability is attained through feature importance and pathological feature interaction analysis. More interestingly, both from the image-level and feature-level interpretability, color features and texture characteristics are shown to contribute the most to the MSI predictions. Therefore, the classification models under the proposed strategies can not only serve as an efficient tool for predicting the MSI status of patients, but also provide more insights to pathologists with clinical understanding.

A Fast and Effective Method of Macula Automatic Detection for Retina Images

Retina image processing is one of the crucial and popular topics of medical image processing. The macula fovea is responsible for sharp central vision, which is necessary for human behaviors where visual detail is of primary importance, such as reading, writing, driving, etc. This paper proposes a novel method to locate the macula through a series of morphological processing. On the premise of maintaining high accuracy, our approach is simpler and faster than others. Furthermore, for the hospital's real images, our method is also able to detect the macula robustly.