Search

Learning to segmentation without large-scale samples is an inherent capability of human. Recently, Segment Anything Model (SAM) performs the significant zero-shot image segmentation, attracting considerable attention from the computer vision community. Here, we investigate the capability of SAM for medical image analysis, especially for multi-phase liver tumor segmentation (MPLiTS), in terms of prompts, data resolution, phases. Experimental results demonstrate that there might be a large gap between SAM and expected performance. Fortunately, the qualitative results show that SAM is a powerful annotation tool for the community of interactive medical image segmentation., Comment: Preliminary investigation

Multi-phase liver contrast-enhanced computed tomography (CECT) images convey the complementary multi-phase information for liver tumor segmentation (LiTS), which are crucial to assist the diagnosis of liver cancer clinically. However, the performances of existing multi-phase liver tumor segmentation (MPLiTS)-based methods suffer from redundancy and weak interpretability, % of the fused result, resulting in the implicit unreliability of clinical applications. In this paper, we propose a novel trustworthy multi-phase liver tumor segmentation (TMPLiTS), which is a unified framework jointly conducting segmentation and uncertainty estimation. The trustworthy results could assist the clinicians to make a reliable diagnosis. Specifically, Dempster-Shafer Evidence Theory (DST) is introduced to parameterize the segmentation and uncertainty as evidence following Dirichlet distribution. The reliability of segmentation results among multi-phase CECT images is quantified explicitly. Meanwhile, a multi-expert mixture scheme (MEMS) is proposed to fuse the multi-phase evidences, which can guarantee the effect of fusion procedure based on theoretical analysis. Experimental results demonstrate the superiority of TMPLiTS compared with the state-of-the-art methods. Meanwhile, the robustness of TMPLiTS is verified, where the reliable performance can be guaranteed against the perturbations.

BackgroundSerum carcinoembryonic antigen (CEA) is an important biomarker for diagnosis, prognosis, recurrence, metastasis monitoring, and the evaluation of the effect of chemotherapy in colorectal cancer (CRC). However, few studies have focused on the role of early postoperative CEA in the prognosis of stage II CRC.MethodsPatients with stage II CRC diagnosed between January 2007 and December 2015 were included. Receiver operating characteristic (ROC) curves were used to obtain the cutoff value of early postoperative CEA, CEA ratio and CEA absolute value. The areas under curves (AUCs) were used to estimate the predictive abilities of the CEA and T stage. The stepwise regression method was used to screen the factors included in the Cox regression analysis. Before and after propensity score (PS) - adjusted Cox regression and sensitivity analysis were used to identify the relationship between early postoperative CEA and prognosis. Meta-analysis was performed to verify the results. Kaplan-Meier survival curves were used to estimate the effects of CEA on prognosis.ResultsWe included 1081 eligible patients. ROC curves suggested that the cutoff value of early postoperative CEA was 3.66 ng/ml (P 0.05). Patients with a CEA ratio ≤0.55 or CEA absolute value ≤-0.98 had a worse prognosis (all P-values were < 0.001). Survival analysis suggested that adjuvant chemotherapy for stage II CRC patients with elevated early postoperative CEA may improve the CSS (P = 0.040).ConclusionsEarly postoperative CEA was a better biomarker for prognosis of stage II CRC patients than T stage and preoperative CEA, and has the potential to become a high-risk factor to guide the prognosis and treatment of stage II CRC patients.

Market risk models often deal with risk measurement and modeling in a specific capitalhorizon, while model developers have to select an estimation horizon for the parameterestimation. The use of overlapping data may be a solution for the trade-off between bettersignal-to-noise ratio and the lower number of observations in the long-horizon data. We focuson the beta estimate in one-factor linear regression model. Three data generating process areconsidered in simulation: (1) independent identically distributed (iid) model, (2) generalizedautoregressive conditional heteroskedasticity model, (3) decomposition of stock price into arandom walk and stationary components. If the daily data perfectly satisfy a linear modelwith iid error, estimate using daily data may be better than estimate using long-horizonoverlapping data. For the general linear regression model, generalized least squares withlonger horizon may produce better results since it takes into account the serial correlationin overlapping data.

Tianyi Xia, Sang Zhu, Yan Wen, Shouhong Gao, Mingming Li, Xia Tao, Feng Zhang, Wansheng Chen Department of Pharmacy, Changzheng Hospital, Second Military Medical University, Shanghai, People’s Republic of China Background: Nephrotoxicity of calcineurin inhibitors (CNIs) is the major concern for long-term allograft survival despite its predominant role in current immunosuppressive regime after renal transplantation. CNI nephrotoxicity is multifactorial with demographic, environmental, and pharmacogenetic flexibility, whereas studies indicating risk factors for CNI nephrotoxicity obtained incomplete or conflicting results.Methods: A systematic review and meta-analysis of risk factors for CNI nephrotoxicity was performed on all retrieved studies through a comprehensive research of network database. Data were analyzed by Review Manager 5.2 with heterogeneity assessed using the Cochrane Q and I2 tests. CNI nephrotoxicity was primarily indicated with protocol biopsy or index-based clinical diagnosis, and the secondary outcome was defined as delayed graft function.Results: Twelve observational studies containing a total of 2,849 cases were identified. Donor age (odds ratio [OR], 1.01; 95% CI, 1.01–1.03; p=0.02), recipient zero-time arteriosclerosis (OR, 1.44; 95% CI, 1.04–1.99; p=0.03), and CYP3A5*3/*3 genotype (OR, 2.80; 95% CI, 2.63–2.98; p=0.00) were confirmed as risk factors for CNI nephrotoxicity. Subgroup and sensitivity analysis claimed donor age as a significant contributor in Asian and Caucasian areas.Conclusion: Older donor age, recipient zero-time arteriosclerosis, and CYP3A5*3/*3 genotype might add up the risk for CNI nephrotoxicity, which could be interpreted into a robust biomarker system. Keywords: calcineurin inhibitor, transplantation, nephrotoxicity, risk factor, systematic review, meta-analysis