Search

With the rapid development of medical imaging technology, 3D image segmentation technology has gradually become a mainstream method, especially in brain tumour detection and diagnosis showing its unique advantages. The technique makes full use of 3D spatial information to locate and analyze tumours more accurately, thus playing an important role in improving diagnostic accuracy, optimising treatment planning and promoting research. However, it also suffers from significant computational expenditure and delayed processing pace. In this paper, we propose an innovative optimisation scheme to address this problem. We thoroughly investigate the MedNeXt network and propose Fast‐MedNeXt, which aims to increase the processing speed while maintaining accuracy. First, we introduce the partial convolution (PConv) technique, which replaces the deep convolutional layers in the network. This improvement effectively reduces computation and memory requirements while maintaining efficient feature extraction. Second, based on PConv, we propose PConv‐Down and PConv‐Up modules, which are applied to the up‐sampling and down‐sampling modules to further optimise the network structure and improve efficiency. To confirm the efficacy of the approach, we carried out a sequence of tests in the multimodal brain tumour segmentation challenge 2021 (BraTS2021). By comparing with the MedNeXt series network, the Fast‐MedNeXt reduced the latency by 22.1%, 20.5%, 15.8%, and 11.4% respectively, while the average accuracy also increased by 0.475% and 0.2% respectively. These significant performance improvements demonstrate the effectiveness of Fast‐MedNeXt in 3D medical image segmentation tasks and provide a new and more efficient solution for the field. [ABSTRACT FROM AUTHOR]

Abstract Objective CT-Urography combined with 3D printing technology, digital design, construction of individualized PCNL puncture guides, and preliminary analyze their efficacy, safety puncture positioning for PCNL. Methods Twenty-two patients with renal calculi were randomly selected at the affiliated Hospital of Xuzhou Medical University during 2017–2018. We randomly divided the patients into two groups: in 10 experimental groups, we used our 3D printing personalized percutaneous puncture guide access plate for PCNL, and in the control group, 12 patients with standard USG guide PCNL. The accuracy of puncture position, puncture time, and intraoperative blood loss was compared. Results In the experimental group, 10 patients with 3D printing personalized percutaneous puncture guide access plate. The puncture needle was accessed through the guide plate and verified by the color Doppler. The single puncture, needle position, and depth success rate were 100.00% (10/10). The angles were consistent with the preoperative design. In the control group, 12 patients via USG guided PCNL success rate was 75.00% (9/12). The puncture time and amount of hemorrhage was (7.78 ± 0.94) min and (49.31 ± 6.43) mL, and (9.04 ± 1.09) min and (60.08 ± 12.18) mL, respectively. The above data of the two groups were statistically significant (P

Remote sensing (RS) images play an indispensable role in many key fields such as environmental monitoring, precision agriculture, and urban resource management. Traditional deep convolutional neural networks have the problem of limited receptive fields. To address this problem, this paper introduces a hybrid network model that combines the advantages of CNN and Transformer, called MBT-UNet. First, a multi-branch encoder design based on the pyramid vision transformer (PVT) is proposed to effectively capture multi-scale feature information; second, an efficient feature fusion module (FFM) is proposed to optimize the collaboration and integration of features at different scales; finally, in the decoder stage, a multi-scale upsampling module (MSUM) is proposed to further refine the segmentation results and enhance segmentation accuracy. We conduct experiments on the ISPRS Vaihingen dataset, the Potsdam dataset, the LoveDA dataset, and the UAVid dataset. Experimental results show that MBT-UNet surpasses state-of-the-art algorithms in key performance indicators, confirming its superior performance in high-precision remote sensing image segmentation tasks. [ABSTRACT FROM AUTHOR]

Purpose: Numerous indicators can be used to predict tumor patients' prognosis and tumor regression grade (TRG). The role of the neutrophil-lymphocyte ratio (NLR) among individuals with locally advanced rectal cancer (LARC) following neoadjuvant chemoradiotherapy (nCRT) hasn't been studied, nevertheless. This study aims to explore the predictive value of the NLR before nCRT (pre-NLR) in TRG and prognosis of LARC patients undergoing nCRT.. Methods: In this retrospective investigation, 326 LARC patients receiving nCRT in total were included. The link between the pre-NLR and TRG was examined using a logistic regression analysis. A Cox-based nomogram was created in the meanwhile to forecast overall survival (OS). With the use of calibration plots and receiver operating characteristic (ROC) curves, we evaluated the nomogram's predictive capabilities. Results: The median pre-NLR across 326 patients was 2.2 (interquartile range, IQR: 1.7-2.7). In the logistic regression analysis, only the pre-NLR for TRG in LARC patients receiving nCRT was statistically significant (odds ratio, OR = 0.62, 95% CI: 0.47-0.80, P < 0.001). Pre-NLR, nCRT with surgery interval, ypTNM stage, TRG, vascular invasion, adjuvant chemotherapy, and carbohydrate antigen 19–9 before nCRT were revealed to be OS predictors in the Cox multivariate analysis. According to calibration plots and ROC curves, the predictive nomogram demonstrated high statistical performance on internal validation. Conclusion: This study demonstrated that a lower pre-NLR was probably associated with a greater rate of TRG in LARC patients undergoing nCRT. Furthermore, the pre-NLR was credibly correlated with OS in LARC patients undergoing nCRT. Meanwhile, we constructed a nomogram for predicting the prognosis in LARC patients undergoing nCRT. [ABSTRACT FROM AUTHOR]