Multi-Object Tracking (MOT) aims to maintain stable and uninterrupted trajectories for each target. Most state-of-the-art approaches first detect objects in each frame and then implement data association between new detections and existing tracks using motion models and appearance similarities. Despite achieving satisfactory results, occlusion and crowds can easily lead to missing and distorted detections, followed by missing and false associations. In this paper, we first revisit the classic tracker DeepSORT, enhancing its robustness over crowds and occlusion significantly by placing greater trust in predictions when detections are unavailable or of low quality in crowded and occluded scenes. Specifically, we propose a new framework comprising of three lightweight and plug-and-play algorithms: the probability map, the prediction map, and the covariance adaptive Kalman filter. The probability map identifies whether undetected objects have genuinely disappeared from view (e.g., out of the image or entered a building) or are only temporarily undetected due to occlusion or other reasons. Trajectories of undetected targets that are still within the probability map are extended by state estimations directly. The prediction map determines whether an object is in a crowd, and we prioritize state estimations over observations when severe deformation of observations occurs, accomplished through the covariance adaptive Kalman filter. The proposed method, named MapTrack, achieves state-of-the-art results on popular multi-object tracking benchmarks such as MOT17 and MOT20. Despite its superior performance, our method remains simple, online, and real-time. The code will be open-sourced later.

To evaluate the impact of a new whole-heart motion-correction algorithm (SnapShot Freeze 2, SSF2) on non-contrast cardiac CT images and the coronary artery calcium scores (CACS). 100 consecutive patients were prospectively included to undergo ECG-gated CT CACS scanning. Images were reconstructed with both standard (STD) algorithm and SSF2. CACS was calculated by a semi-automatic software. Image noise and signal-to-noise ratio (SNR) of the aortic root were measured. Two radiologists evaluated the subjective image quality of coronary arteries and calcified plaques using a 5-point scale and 3-point scale, respectively. CACS of 278 vessels with calcified plaques in 100 patients were calculated. The overall median (range) Agatston score, volume score, and mass score of SSF2 were 172.3 (307.9), 138 (252.9) and 38.3 (65.6), respectively, lower than 193.4 (342.0), 160.0 (284.8) and 39.3 (74.4) of STD (all P < 0.001). SSF2 provided lower noise (17.4 ± 2.7 VS. 18.5 ± 2.9 HU, P < 0.001) and higher SNR (2.90 ± 0.58 VS. 2.70 ± 0.55, P < 0.001) than those of STD. The subjective image quality scores of SSF2 group were significantly higher than that of STD group (all P < 0.001). The new whole-heart motion-correction algorithm improves non-contrast cardiac CT image quality, lowers CACS, and shows a potential for increasing CACS accuracy.

To assess the association between epicardial adipose tissue (EAT) index derived from cardiac computed tomography and atrial fibrillation (AF) recurrence after ablation by comparing with a propensity score matched non-recurrence AF patients. A total of 506 patients with AF recurrence and 174 patients without AF recurrence were enrolled in this retrospective study. Density and volume of total EAT surrounding the heart (Total-EAT) and EAT surrounding the left atrium (LA-EAT) were measured, propensity score matching(PSM) analyses were used to compare the outcomes of the two groups while controlling for confounders. Total-EAT density (HU) value (-81.27 ± 4.67 vs -84.05 ± 3.84, P < 0.001) and LA-EAT density (HU) value (-76.16 ± 4.11 vs -78.83 ± 3.81, P < 0.001) were significantly higher in the patients with AF recurrence than in those without recurrence. LA-EAT density (HU) value was significantly higher than Total-EAT (− 77.50 ± 4.18 vs -82.66 ± 4.49, P = 0.000). In a multiple logistic regression analysis, a higher LA-EAT density (odds ratio: 1.12; 95% CI: 1.02–1.22, p = 0.015) was significantly associated with the AF recurrence after adjusting for other risk factors. The LA-EAT density plays an important role in the AF recurrence after ablation. Assessment of LA-EAT density can improve ablation outcomes by refining patient selection.

The hypoxia microenvironment is highly associated with GBM’s malignant phenotypes. CircRNAs were reported involved in GBM’s biological characteristics and regulated by HIF-1α. However, the differential expression profile and role of circRNAs in GBM cells under hypoxia are still unclear. The expression profiles of circRNAs in LN229 and T98G under hypoxia were explored via circRNA sequencing analysis. Those circRNAs significantly dysregulated both in LN229 and T98G and could be found in circBase were selected and validated by qRT-PCR, RNase R digestion reaction, and Sanger sequencing. Normal cell line and fresh GBM tissues were also used for qRT-PCR validation. The roles of differentially expressed circRNAs were evaluated by bioinformatics analyses. There were 672 dysregulated circRNAs in LN229 and 698 dysregulated circRNAs in T98G. GO analysis indicated that the alteration of circRNA expression related to GBM cell’s biogenesis and metabolism. KEGG analysis demonstrated that TGF-β signaling pathway, HIF-1 signaling pathway, and metabolism-related signaling pathway were closely associated with differentially expressed circRNAs under hypoxia. These results were confirmed by GSEA analysis. The 6 selected and dysregulated circRNAs both in LN229 and T98G including hsa_circ_0000745, hsa_circ_0020093, hsa_circ_0020094, hsa_circ_0000943, hsa_circ_0004874, and hsa_circ_0002359 were validated by qRT-PCR. Inhibition of hsa_circ_0000745 inhibited GBM cell’s proliferation, migration, and invasion. HIF-1α centered circRNA-miRNA-mRNA networks analysis showed that the 6 validated circRNAs could cross-talk with 11 related miRNAs. The circRNA expressions are dysregulated in GBM cell under hypoxia. The 6 validated circRNAs could participate in GBM’s development and progression when hypoxia occurs. They might be the candidates for prognostic markers and adjuvant therapeutics of GBM in the future.

PurposeThis study aimed to evaluate the feasibility of differentiating the atrial fibrillation (AF) subtype and preliminary explore the prognostic value of AF recurrence after ablation using radiomics models based on epicardial adipose tissue around the left atrium (LA-EAT) of cardiac CT images. MethodThe cardiac CT images of 314 patients were collected wherein 251 and 63 cases were randomly enrolled in the training and validation cohorts, respectively. Mutual information and the random forest algorithm were used to screen for the radiomic features and construct the radiomics signature. Radiomics models reflecting the features of LA-EAT were built to differentiate the AF subtype, and the multivariable logistic regression model was adopted to integrate the radiomics signature and volume information. The same methodology and algorithm were applied to the radiomic features to explore the ability for predicting AF recurrence. ResultsThe predictive model constructed by integrating the radiomic features and volume information using a radiomics nomogram showed the best ability in differentiating AF subtype in the training [AUC, 0.915; 95% confidence interval (CI), 0.880-0.951] and validation (AUC, 0.853; 95% CI, 0.755-0.951) cohorts. The radiomic features have shown convincible predictive ability of AF recurrence in both training (AUC, 0.808; 95% CI, 0.750-0.866) and validation (AUC, 0.793; 95% CI, 0.654-0.931) cohorts. ConclusionsThe LA-EAT radiomic signatures are a promising tool in the differentiation of AF subtype and prediction of AF recurrence, which may have clinical implications in the early diagnosis of AF subtype and disease management.

Objectives: This study aimed to evaluate coronary inflammation by measuring the perivascular fat attenuation index (FAI) and quantify the atherosclerosis burden in patients with psoriasis and control individuals without psoriasis based on coronary computed tomography angiography (CCTA) images. Methods: A total of 98 consecutive patients with psoriasis (76 male [77.6%], aged 56.5 years, range 45.5–65.0) were recruited, and 196 patients (157 male [80.1%]; aged 54.6 ± 14.1 years) without established cardiovascular disease (CVD) who underwent CCTA within the same period were enrolled in the control group. Coronary plaque burden was quantified using the computed tomography-adapted Leaman score (CT-LeSc), and the FAI surrounding the proximal of three main epicardial vessels was measured to represent coronary inflammation. Results: Patients with psoriasis and the control subjects were well matched in CVD risk factors (all p > 0.05). Psoriasis patients had a greater overall CT-LeSc (5.86 vs. 4.69, p = 0.030) and lower perivascular FAI (−80.19 ± 7.48 vs. −78.14 ± 7.81 HU, p < 0.001). A similar result was found upon comparing psoriasis patients without biological or statin therapy with non-psoriasis individuals without statin treatments. Furthermore, the psoriasis group had a higher prevalence of non-calcified plaques (30.3% in the psoriasis group vs. 20.1% in the control subjects, p = 0.001). No difference in perivascular FAI on either calcified and mixed plaques or non-calcified plaques between the two groups was found. Conclusion: Patients with psoriasis have a higher atherosclerotic burden as quantified by CT-LeSc and less coronary inflammation as detected by perivascular FAI around the most proximal of the three major epicardial vessels. The usefulness of perivascular FAI for evaluating coronary inflammation in patients with chronic low-grade inflammatory disease such as psoriasis should be verified.

Background: Glioblastoma (GBM) represents the most aggressive glioma with high invasive potential. Recent studies proved the involvement of epithelial-mesenchymal transition (EMT) process in increasing the malignancy and invasiveness of GBM. LncRNAs have been verified to play pivotal roles in human disease including GBM. However, the molecular mechanisms of lncRNA-mediated EMT in GBM remain largely unknown. LINC-PINT, a LncRNA which has never been studied in GBM before, was predicted to be negatively associated with EMT in GBM. This study aimed to explore the biological function and the EMT relevance of LINC-PINT in GBM and further explore the molecular mechanism. Methods: The bioinformatic prediction data of LINC-PINT in GBM was derived from The Cancer Genome Atlas (TCGA) database by R software and GEPIA website. qRT-PCR assay was performed to detect the expression level of LINC-PINT in GBM cell lines. Cell counting kit-8 (CCK8), clone formation, transwell, and wound healing assays were performed to determine the biological function of LINC-PINT in vivo. Tumor xenograft experiment and tumor peritoneal metastasis experiments were performed to verify the in vivo function. Western blot and immunofluorescence staining assays were carried out to detect the relevance of LINC-PINT with EMT and Wnt/β-catenin signaling. Rescue assays were performed to check the regulation mechanism of LINC-PINT/Wnt signaling/EMT axis in GBM. Results: LINC-PINT was downregulated in GBM cell lines. LINC-PINT suppressed cell progression, invasion, and EMT in GBM. LINC-PINT blocked Wnt/β-catenin signaling in GBM. Conclusion: LINC-PINT suppressed cell proliferation, invasion, and EMT by blocking Wnt/β-catenin signaling in GBM.

NADH dehydrogenase [ubiquinone] 1 alpha subcomplex, 4-like 2 (NDUFA4L2) is a subunit of Complex I of the mitochondrial respiratory chain, which is important in metabolic reprogramming and oxidative stress in multiple cancers. However, the biological role and molecular regulation of NDUFA4L2 in glioblastoma (GBM) are poorly understood. Here, we found that NDUFA4L2 was significantly upregulated in GBM; the elevated levels were correlated with reduced patient survival. Gene knockdown of NDUFA4L2 inhibited tumor cell proliferation and enhanced apoptosis, while tumor cells initiated protective mitophagy in vitro and in vivo. We used lentivirus to reduce expression levels of NDUFA4L2 protein in GBM cells exposed to mitophagy blockers, which led to a significant enhancement of tumor cell apoptosis in vitro and inhibited the development of xenografted tumors in vivo. In contrast to other tumor types, NDUFA4L2 expression in GBM may not be directly regulated by hypoxia-inducible factor (HIF)-1α, because HIF-1α inhibitors failed to inhibit NDUFA4L2 in GBM. Apatinib was able to effectively target NDUFA4L2 in GBM, presenting an alternative to the use of lentiviruses, which currently cannot be used in humans. Taken together, our data suggest the use of NDUFA4L2 as a potential therapeutic target in GBM and demonstrate a practical treatment approach.

BACKGROUND: Glycolysis is an important metabolic manner in glioblastoma multiforme (GBM)'s rapid growth. It has been reported that glutamate-oxaloacetate transaminase 1 (GOT1) is low-expressed in GBM and patients with high-expressed GOT1 have better prognosis. However, the effect and mechanism of GOT1 on glycolysis and malignant phenotypes of GBM cells are still unclear. METHODS: The expression differences of GOT1 between GBM parenchyma and adjacent tissues were detected. The prognosis and clinical data with different levels of GOT1 were also analyzed. The glucose consumption, production of lactate and pyruvate were measured after GOT1 was knocked down or overexpressed. The effects of GOT1 on GBM cell's malignant phenotypes were analyzed by Western blot, CCK-8 assay, and flow cytometry. The relationship between GOT1 and pyruvate carboxylase (PC) was examined by immunoprecipitation and immunofluorescence. RESULTS: GOT1 was expressed little in GBM, and patients with highly expressed GOT1 had longer survival periods. Overexpressed GOT1 inhibited the glycolysis and malignant phenotypes of GBM cells. 2-DG treatment could partially reverse the enhancement of malignant phenotypes caused by knockdown of GOT1. The expression of GOT1 was positively correlated with PC. The inhibitory effect of GOT1 on glycolysis could be partially reversed by PC's knockdown.-BACKGROUND: Glycolysis is an important metabolic manner in glioblastoma multiforme (GBM)'s rapid growth. It has been reported that glutamate-oxaloacetate transaminase 1 (GOT1) is low-expressed in GBM and patients with high-expressed GOT1 have better prognosis. However, the effect and mechanism of GOT1 on glycolysis and malignant phenotypes of GBM cells are still unclear. -METHODS: The expression differences of GOT1 between GBM parenchyma and adjacent tissues were detected. The prognosis and clinical data with different levels of GOT1 were also analyzed. The glucose consumption, production of lactate and pyruvate were measured after GOT1 was knocked down or overexpressed. The effects of GOT1 on GBM cell's malignant phenotypes were analyzed by Western blot, CCK-8 assay, and flow cytometry. The relationship between GOT1 and pyruvate carboxylase (PC) was examined by immunoprecipitation and immunofluorescence. -RESULTS: GOT1 was expressed little in GBM, and patients with highly expressed GOT1 had longer survival periods. Overexpressed GOT1 inhibited the glycolysis and malignant phenotypes of GBM cells. 2-DG treatment could partially reverse the enhancement of malignant phenotypes caused by knockdown of GOT1. The expression of GOT1 was positively correlated with PC. The inhibitory effect of GOT1 on glycolysis could be partially reversed by PC's knockdown. CONCLUSIONS: GOT1 could impair glycolysis by interacting with PC and further inhibit the malignant phenotypes of GBM cells.

Traffic flow prediction is one of the core technologies in Intelligent Transportation System (ITS) to improve traffic management. However, in metropolitan circumstances, the complex traffic road networks and numerous unpredictable traffic anomalies are still tough problems, which bring challenges of leveraging topological and anomalies information to accurate traffic flow prediction. In this paper, we propose a Dynamic Hidden Markov Model (DHMM) based on global PageRank algorithm to overcome these challenges. The global PageRank algorithm is more applicable than traditional algorithm for traffic scenarios, through which the PageRank metric is calculated to measure the accumulation of traffic anomalies at intersections. By incorporating the PageRank metric, DHMM leverages topological and anomalies information to dynamically model the traffic variations. Experiments on real-world dataset demonstrate that the PageRank metric can describe the degree of traffic anomalies intuitively, and the proposed model has superior traffic flow prediction performance both under normal and abnormal traffic conditions.

BACKGROUND: Giant presacral Tarlov cysts (TCs) with pelvic extension are extremely rare and have many special features that differ from normal TCs in examination, diagnosis, symptoms, and treatment. We report 3 rare cases of giant presacral TCs with pelvic extension and review the pertinent literature. CASE DESCRIPTION: We report 3 cases of giant presacral TCs with rare pelvic extension and analyzed the symptoms, diagnoses, and surgical procedures. Operations with the key point of blocking the inlet of the fistula from inside the dural sac were performed in all 3 cases. All 3 patients revealed alleviation of previous symptoms with no serious complications. Postoperative magnetic resonance imaging showed all the cysts were well blocked with no cyst recurrence. CONCLUSIONS: Giant TC with pelvic extension is extremely rare and often is discovered on gynecological ultrasound, where it might be misdiagnosed as adnexal mass. Different from patients with normal TCs, these patients also may present with abdominal symptoms like hydronephrosis, abdominal, or pelvic pain due to the cyst's ventral mass effect. Thus, patients with abdominal and back symptoms at the same time should be paid particular attention for lumbosacral magnetic resonance imaging examination to avoid misdiagnosis. Surgical procedures are recommended for symptomatic cases. However, cyst resection by laparotomy is doomed to postoperative recurrence because the fistula still exists. We describe a simple procedure with the key point of blocking the inlet of cyst fistula, which is more applicable and minimizes the probability of cyst recurrence.

In various malignant tumors, NF-kappa B interacting long noncoding RNA (NKILA) displays antitumor activity by inhibiting the NF-kappa B pathway. However, the role of NKILA in gliomas remains unclear. Surprisingly, this study showed that NKILA is significantly upregulated in gliomas, and the increased levels of NKILA were correlated with a decrease in patient survival time. NKILA increased the expression level of hypoxia-inducible factor-1α, and the activity of the hypoxia pathway in gliomas. Furthermore, we demonstrated that NKILA enhances the Warburg effect and angiogenesis in gliomas both in vitro and in vivo. Therefore, NKILA is a potential therapeutic target in gliomas. In addition, we showed that a 20(S)-Rg3 monomer suppresses NKILA accumulation and reverses its stimulation of the Warburg effect and angiogenesis in gliomas, both in vitro and in vivo. Therefore, this study not only identified NKILA as a potential therapeutic target in gliomas, but also demonstrated a practical approach to treatment.

Background: The present study aimed to compare the feasibility and safety of early oral feeding (EOF) with traditional oral feeding (TOF) after radical total gastrectomy for gastric cancer.Methods: This retrospective study included consecutive patients who underwent total gastrectomy from April 2016 and November 2018. These patients were divided into two groups, according to their postoperative feeding protocol: EOF group (n = 314) and TOF group (n = 433). Propensity score matching was used to balance the potential confounders, and 276 patients were selected from each group. The EOF group received oral diet on postoperative day one, while the TOF group were started on oral feeding after the passage of flatus.Results: No significant differences were found in the postoperative complications (P = 0.426) and tolerance to oral feeding (P > 0.056) between the two groups. The changes in perioperative nutritional markers were also similar between the two groups (P > 0.05). The time to first passage of flatus or defecation (47.19 ± 12.00 h vs. 58.19 ± 9.89 h, P < 0.0001) and length of postoperative hospital stay (6.84 ± 2.31 days vs. 7.72 ± 2.86 days, P < 0.0001) were significantly lower in the EOF group compared to the TOF group.Conclusion: EOF may be safe and feasible after radical total gastrectomy with faster recovery and no increased risk of postoperative complications.

Traffic prediction is a fundamental and vital task in Intelligence Transportation System (ITS), but it is very challenging to get high accuracy while containing low computational complexity due to the spatiotemporal characteristics of traffic flow, especially under the metropolitan circumstances. In this work, a new topological framework, called Linkage Network, is proposed to model the road networks and present the propagation patterns of traffic flow. Based on the Linkage Network model, a novel online predictor, named Graph Recurrent Neural Network (GRNN), is designed to learn the propagation patterns in the graph. It could simultaneously predict traffic flow for all road segments based on the information gathered from the whole graph, which thus reduces the computational complexity significantly from O(nm) to O(n+m), while keeping the high accuracy. Moreover, it can also predict the variations of traffic trends. Experiments based on real-world data demonstrate that the proposed method outperforms the existing prediction methods.

In Vehicular Transportation System (VTS), a high-efficiency monitoring system is significant, which provides assistance in relieving the traffic congestion and the risk of accidents. Existent approaches generally use completion algorithms to estimate the traffic condition with the single source or homogeneous data, which are not sufficient. In this paper, we design a heterogeneous data fusion system to combine different kinds of traffic information. A robust regression based Granger causality test is introduced to search the strong connection between sensing reports from taxis and buses. Due to the complementary characteristics of driving patterns of taxis and buses, the proposed system can restore the traffic monitoring matrix with high accuracy and fix the sparsity problem. Real world data-driven evaluations show that the proposed system performs better on accuracy and tracking performance than other completion algorithms when the sampling rate is lower than 15%. Moreover, the proposed restoration method works well for the cases of very low sampling rate under 5%, while other approaches lose the efficiency.