Oblique back-illumination capillaroscopy has recently been introduced as a method for highquality, non-invasive blood cell imaging in human capillaries. To make this technique practical for clinical blood cell counting, solutions for automatic processing of acquired videos are needed. Here, we take the first step towards this goal, by introducing a deep learning multi-cell tracking model, named CycleTrack, which achieves accurate blood cell counting from capillaroscopic videos. CycleTrack combines two simple online tracking models, SORT and CenterTrack, and is tailored to features of capillary blood cell flow. Blood cells are tracked by displacement vectors in two opposing temporal directions (forward-and backwardtracking) between consecutive frames. This approach yields accurate tracking despite rapidly moving and deforming blood cells. The proposed model outperforms other baseline trackers, achieving 𝟔𝟓. 𝟓𝟕% Multiple Object Tracking Accuracy and 𝟕𝟑. 𝟗𝟓% ID F1 score on test videos. Compared to manual blood cell counting, CycleTrack achieves 𝟗𝟔. 𝟓𝟖 ± 𝟐. 𝟒𝟑% cell counting accuracy among 8 test videos with 1000 frames each compared to 𝟗𝟑. 𝟒𝟓% and 𝟕𝟕. 𝟎𝟐% accuracy for independent CenterTrack and SORT almost without additional time expense. It takes 800s to track and count approximately 8000 blood cells from 9,600 frames captured in a typical one-minute video. Moreover, the blood cell velocity measured by CycleTrack demonstrates a consistent, pulsatile pattern within the physiological range of heart rate. Lastly, we discuss future improvements for the CycleTrack framework, which would enable clinical translation of the oblique back-illumination microscope towards a real-time and non-invasive point-of-care blood cell counting and analyzing technology.

No abstract

No abstract

Clustering patterns are ubiquitously present in a variety of networked systems, and may change with the evolution of network topology. Probing into the cluster structures can shed light on the change of the entire network, especially those sudden changes emerging in the process of network evolution. Though abundant researches have been done in detecting the changes of dynamic networks, more precisely, change points at which the network topology experiences abrupt changes, most of the existing methods focus on local changes (e.g. edges change) that are commonly mixed with noise, giving rise to high false positive reports. Different from the previous work, here we inspect the topological changes from mesoscale clusters of dynamic networks, which will reduce the perturbation of link variation to detection accuracy. Towards this end, we look for the invariant clusters of nodes during the observation window in dynamic networks and propose a new measure to quantify the stability of node clusters with respect to the invariant clustering patterns. Then the change of dynamic networks at mesoscale can be captured by comparing the variations of stability measures. In the light of the proposed measurement, we design a change-point detection algorithm and conduct extensive experiments on synthetic and real-life datasets to demonstrate the effectiveness of our method. The results show the outperformance of our method in identifying change points, compared to several baseline methods.

Small rodent cardiac magnetic resonance imaging (MRI) plays an important role in preclinical models of cardiac disease. Accurate myocardial boundaries delineation is crucial to most morphological and functional analysis in rodent cardiac MRIs. However, rodent cardiac MRIs, due to animal's small cardiac volume and high heart rate, are usually acquired with sub-optimal resolution and low signal-to-noise ratio (SNR). These rodent cardiac MRIs can also suffer from signal loss due to the intra-voxel dephasing. These factors make automatic myocardial segmentation challenging. Manual contouring could be applied to label myocardial boundaries but it is usually laborious, time consuming, and not systematically objective. In this study, we present a deep learning approach based on 3D attention M-net to perform automatic segmentation of left ventricular myocardium. In the deep learning architecture, we use dual spatial-channel attention gates between encoder and decoder along with multi-scale feature fusion path after decoder. Attention gates enable networks to focus on relevant spatial information and channel features to improve segmentation performance. A distance derived loss term, besides general dice loss and binary cross entropy loss, was also introduced to our hybrid loss functions to refine segmentation contours. The proposed model outperforms other generic models, like U-Net and FCN, in major segmentation metrics including the dice score (0.9072), Jaccard index (0.8307) and Hausdorff distance (3.1754 pixels), which are comparable to the results achieved by state-of-the-art models on human cardiac ACDC17 datasets.Clinical relevance Small rodent cardiac MRI is routinely used to probe the effect of individual genes or groups of genes on the etiology of a large number of cardiovascular diseases. An automatic myocardium segmentation algorithm specifically designed for these data can enhance accuracy and reproducibility of cardiac structure and function analysis.