Search

Background There are conflicting results regarding the role of IC in T3-4N0-1 NPC patients from endemic area of China. Therefore, it is worth verifying the consistency of study between endemic area and non-endemic area for NPC and providing treatment guidance for non-endemic NPCs. We conducted a multicenter real-world study to investigate the value of adding IC to CCRT for T3-4N0-1 NPC in Northwest of China. Methods Data was extracted in three hospitals from Northwest China between May 1, 2010 and August 30, 2018. The Kaplan-Meier method was used to estimate the endpoints, survival curves were compared using the log-rank test. Initial propensity matching was conducted with a 1:1 match of IC + CCRT to CCRT. The primary endpoint of this study was overall survival (OS). Results A total of 108 patients with staging T3-4N0-1 were included in this study. The median follow-up time was 50 months (range: 6-118 months). IC followed by CCRT had trend to improve OS compared with CCRT (89.5% vs.77.6%, HR:0.41, 95%CI:0.16–1.04, P = 0.100). After analysis of well-balanced propensity score-matched cohort, IC followed by CCRT provided superior OS than CCRT alone. Adjusted 4-year OS was 89.5% for IC followed by CCRT vs. 71.1% for CCRT (HR:0.30, 95%CI:0.11–0.80, P = 0.027). No significant differences were detected in side effects between two groups. Conclusion This study implied IC followed by CCRT improved OS in patients with T3-4N0-1M0 NPC from Northwest China compared with CCRT. However, prospective studies with large sample are warranted to confirm the result.

Advances in moving object detection have been driven by the active application of deep learning methods. However, many existing models render superior detection accuracy at the cost of high computational complexity and slow inference speed. This fact has hindered the development of such models in mobile and embedded vision tasks, which need to be carried out in a timely fashion on a computationally limited platform. In this paper, we propose a super-fast (inference speed-154 fps) and lightweight (model size-1.45 MB) end-to-end 3D separable convolutional neural network with a multi-input multi-output (MIMO) strategy named “3DS_MM” for moving object detection. To improve detection accuracy, the proposed model adopts 3D convolution which is more suitable to extract both spatial and temporal information in video data than 2D convolution. To reduce model size and computational complexity, the standard 3D convolution is decomposed into depthwise and pointwise convolutions. Besides, we proposed a MIMO strategy to increase inference speed, which can take multiple frames as the network input and output multiple frames of detection results. Further, we conducted the scene dependent evaluation (SDE) and scene independent evaluation (SIE) on the benchmark CDnet2014 and DAVIS2016 datasets. Compared to state-of-the-art approaches, our proposed method significantly increases the inference speed, reduces the model size, meanwhile achieving the highest detection accuracy in the SDE setup and maintaining a competitive detection accuracy in the SIE setup.

Triple-negative breast cancer (TNBC) is a highly aggressive subtype and only some of patients could benefit from the immunotherapy. The present study aims to investigate the expression pattern and prognostic value of immune checkpoint genes (ICGs) in TNBC and develop a novel ICGs-signature to predict the prognosis and immune status in TNBC.ICGs expression profiles and clinical characteristics of TNBC samples were obtained from The Cancer Genome Atlas (TCGA) and the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) database. The least absolute shrinkage and selection operator (LASSO) Cox regression analysis was employed to construct a multi-gene signature for predicting the prognostic outcome. The risk scores were calculated based on the coefficients of each ICG in LASSO-Cox regression model. The median score was considered as the cut-off value to divide the TNBC patients into a high-risk group and a low-risk group. The Kaplan-Meier survival curves were generated to further explore the association between the risk scores and prognostic outcomes. Finally, single sample gene set enrichment analysis (ssGSEA) was conducted to evaluate the immune status and immunophenoscore (IPS) score was used for the quantitative evaluation of tumor immunogenicity.A novel ICGs-signature was developed and validated, which may be not only served as a robust prognostic marker, but also a potential indicator reflecting immunotherapy response.

Background Cranial nerve (CN) palsy due to cancer involvement has been considered as an unfavorable prognostic factor for patients with nasopharyngeal carcinoma (NPC). We assessed the role of IMRT based treatment on the recovery of CN palsy and investigated the prognostic value of complete recovery of CN palsy. Methods A total of 115 NPC patients with cancer-related CN palsy were included in the study. We referred CTCAE version 5.0 to evaluate the grade of CN palsy. Results All patients with grade 1 CN palsy recovered completely during the 2 years of follow-up after definite treatment. Most grade 2 palsy could change gradually to grade 1 palsy or complete recovery during 2 years of follow-up. Patients with more than 2 symptoms of CN palsy had poor 3-year disease-free survival (DFS) than these with 1 or 2 symptoms (60.3% vs. 84.9%, HR 0.25, 95% CI 0.07–0.89, P = 0.001). There were no significant differences for PFS, OS, DMFS and LRFS between patients with complete recovery and non-complete recovery from CN palsy after receiving IMRT based comprehensive treatment. Conclusions IMRT based comprehensive treatment could effectively promote the recovery of tumor-related CN palsy for NPC patient. More than 2 symptoms of CN palsy was a poor prognostic factor for DFS of NPC patients. The prognostic role of complete recovery of CN palsy was not identified in our study.

This paper reviews the second AIM learned ISP challenge and provides the description of the proposed solutions and results. The participating teams were solving a real-world RAW-to-RGB mapping problem, where to goal was to map the original low-quality RAW images captured by the Huawei P20 device to the same photos obtained with the Canon 5D DSLR camera. The considered task embraced a number of complex computer vision subtasks, such as image demosaicing, denoising, white balancing, color and contrast correction, demoireing, etc. The target metric used in this challenge combined fidelity scores (PSNR and SSIM) with solutions’ perceptual results measured in a user study. The proposed solutions significantly improved the baseline results, defining the state-of-the-art for practical image signal processing pipeline modeling.

This paper reviews the first AIM challenge on mapping camera RAW to RGB images with the focus on proposed solutions and results. The participating teams were solving a real-world photo enhancement problem, where the goal was to map the original low-quality RAW images from the Huawei P20 device to the same photos captured with the Canon 5D DSLR camera. The considered problem embraced a number of computer vision subtasks, such as image demosaicing, denoising, gamma correction, image resolution and sharpness enhancement, etc. The target metric used in this challenge combined fidelity scores (PSNR and SSIM) with solutions' perceptual results measured in a user study. The proposed solutions significantly improved baseline results, defining the state-of-the-art for RAW to RGB image restoration.

The RIT-DuPont dataset has been used extensively for formula development and testing since its inception during the 1980's, for example, in the development of CIE94 and CIEDE2000. The dataset was published as 156 color-tolerances, T50, along specific vector directions about 19 color centers. Probit analysis was used to transform judgments of 958 color-difference pairs by 50 observers to these 156 tolerances. For most statistical significance testing, the number of samples determines the confidence limits. Thus, there was an interest in publishing the individual color-difference pair visual and colorimetric data to improve the precision of significance testing. From these 958 pairs, 828 pairs had determinable visual differences. The others had either excessive visual uncertainty or had unanimous visual judgments such that visual differences were undefined. In addition, a method was devised to assign visual uncertainty to each of these pairs using the principles of maximum likelihood and the T50 values. Comparisons were made between the T50 and individual color-difference pair data both including and omitting uncertainty weightings. The weighted dataset was found to be equivalent to the T50 tolerances. © 2009 Wiley Periodicals, Inc. Col Res Appl, 2010