Search

ObjectiveThe Buffalo Concussion Physical Examination (BCPE) is a brief, but pertinent physical examination designed for the subacute, outpatient assessment of concussion. The purpose of this study was to perform the BCPE on a larger sample and derive a scoring system to identify children at risk for Persistent Post-Concussive Symptoms (PPCS, recovery ≥30 days).MethodsThis prospective, observational cohort study from September 2016 to March 2019 was performed at three university-affiliated concussion clinics. Male and female children (n=270, 14.92±1.86 years, range 8–18, 38% female) were diagnosed with a concussion within 14 days of injury and followed-up until recovery. Logistic regression was used with history and physical examination variables to predict PPCS and a weighted scoring metric was derived.ResultsOut of 15 predictor variables, the main effects of 1 preinjury variable (≥3 previous concussions), 2 injury characteristic variables (days-since-injury and type-of-injury), 3 physical examination variables (orthostatic intolerance (OI), vestibulo-ocular reflex (VOR) and tandem gait) and 2 interaction terms (OI/VOR and tandem gait/type-of-injury) produced a score that was 85% accurate for identifying children with low-risk, medium-risk and high-risk for PPCS on cross-validation.ConclusionThe Risk for Delayed Recovery (RDR)-Score allows physicians in an outpatient setting to more accurately predict which children are at greater risk for PPCS early after their injury, and who would benefit most from targeted therapies. The RDR-Score is intended to be used as part of a comprehensive assessment that should include validated symptom checklists, mental health history and adjunct testing (eg, cognitive or physical exertion) where clinically indicated.

The research area of celebrity endorsement and its impact on consumers’ purchase intention has always drawn a considerable attention to the domain. This study aims to add a relevant research output to the domain in the context of FMCG products which is a new addition to the existing literature. The study has used an empirical approach of 200 sample respondents and mostly respondents are from northern part of India. It uses celebrity endorsement ( CE ) as independent variable ( IDV) and purchase intention (PI) as dependent variable. This study is based on the respondents’ immediate attitude or thinking about the impact which has been measured through an adapted questionnaire of opinion-based scales. The finding has resulted into positive effect of CE on PI although the proportion of variance in the PI that can be explained by the CE has been found weak , although statistically significant. The study has further suggested to explore more variables which can better predict purchase intention than celebrity endorsement in the context of FMCG product purchase. Keyword : Celebrity endorsement ; purchase intention; linear regression ; statistically significant; FMCG products; consumer The research area of celebrity endorsement and its impact on consumers’ purchase intention has always drawn a considerable attention to the domain. This study aims to add a relevant research output to the domain in the context of FMCG products which is a new addition to the existing literature. The study has used an empirical approach of 200 sample respondents and mostly respondents are from northern part of India. It uses celebrity endorsement ( CE ) as independent variable ( IDV) and purchase intention (PI) as dependent variable. This study is based on the respondents’ immediate attitude or thinking about the impact which has been measured through an adapted questionnaire of opinion-based scales. The finding has resulted into positive effect of CE on PI although the proportion of variance in the PI that can be explained by the CE has been found weak , although statistically significant. The study has further suggested to explore more variables which can better predict purchase intention than celebrity endorsement in the context of FMCG product purchase. Keyword : Celebrity endorsement ; purchase intention; linear regression ; statistically significant; FMCG products; consumer

BACKGROUND: Epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) are increasingly used for advanced non-small cell lung cancer (NSCLC) as first-line therapy. The bioavailability and efficacy of oral EGFR-TKIs could be affected by acid suppression (AS) therapy such as PPIs and H2RAs which are reported to be over-prescribed. Hence, there is a need to investigate the effect of AS on the overall survival (OS), progression-free survival (PFS) and adverse effect profile in patients treated with EGFR TKIs. METHODS: An electronic database search of Medline and Embase was performed following PRISMA guidelines on 17 January 2021. Studies analyzing interactions between EGFR TKIs and PPIs/H2RAs in NSCLC patients were included. Abstracts, non-English or non-Japanese studies or studies using non-EGFR TKIs were excluded. Hazard ratios (HRs) were pooled using generic inverse variance random effects model. Effect sizes for dichotomous variables were pooled using Mantel-Haenszel random effects model. Significance was considered at P≤0.05. Heterogeneity was assessed with Cochran Q-test and I2 test. Publication bias was assessed with funnel plots. The assessment of quality and risk of bias of randomized and non-randomized studies were undertaken with RoB 2 and the ROBINS-I tool respectively. RESULTS: Out of 1,173 potentially relevant articles, 14 articles were included in the final analysis. The pooled prevalence of AS in patients taking EGFR TKI was 30.71% in 4,010 individuals. Patients who were treated only with EGFR TKI had significantly better OS (HR =1.46, 95% CI: 1.27–1.72; P

In this paper, a hybrid numerical technique is developed to solve the Benjamin-Bona-Mahony-Burgers (BBMB) equation. This technique involves the application of quintic Hermite collocation method (QHCM) and weighted finite difference scheme. The discretization of BBMB equation is done using collocation method with Hermite splines of order five along spatial direction and using weighted finite difference scheme along temporal direction. The nonlinear equation is discretized without using the Hopf-Cole transformation. Stability of the proposed technique using Von-Neumann method is checked. Convergence analysis of the technique is also done in detail. Validity of the technique is shown by solving four test examples with different boundary and initial conditions. The calculated results are found to be in good agreement with the exact solutions and are better than the results from other techniques given in the literature.

OBJECTIVES Colorectal cancer is one of the leading causes of cancer-related deaths globally. A multitude of screening methods has been devised for early diagnosis, including the faecal immunochemical test (FIT). This qualitative review aims to discover the barriers and facilitators to the utility of the FIT from the general population's perspective. METHODS Authors searched five electronic bases (Medline, Embase, CINAHL, PsycINFO and Web of Science) till December 2019. The search was conducted using key search terms and qualitative and mixed-method studies were included. Two independent reviewers screened articles and conducted quality appraisal. Data were synthesised thematically. RESULTS A total of 11 articles that reported users' views on FIT kits were included. Three themes were generated from the included articles: FIT kit factors, patients' perception of colorectal cancer screening, and social health support system. The nature of the test and the supplementary information was found to affect the utilisation of the test. User's awareness and perspectives towards cancer and screening were found to have impacted the adoption of the FIT kit. Social support and local healthcare systems were also found to have influenced the use of FIT. CONCLUSION This systematic review focuses on addressing and understanding the perception of FIT from first-hand accounts. Since its inception, FIT screening has increased colorectal cancer screening uptake due to its increased reliability and the simplicity of the test. However, there is a need to increase the uptake of FIT kits through increasing accessibility of the screening process and considering the holistic patient experience.

We exploit a leading FinTech peer-to-peer lending platform paired with an automated robo-advising lending tool to test for and quantify the effects of cultural biases in large-stake risky choices for which the scope for statistical discrimination is minimal. Comparing the choices lenders make with those made by the robo-advising tool on their behalf, we find that both in-group vs. out-group discrimination and stereotypical discrimination are pervasive and economically sizable. Discrimination makes lenders worse off in terms of consumption utility—discriminating lenders face 32% higher default rates and about 11% lower returns on the loans they issue to borrowers belonging to favored demographics, relative to borrowers in the discriminated groups. All the results are stronger for lenders who reside in regions where the salience of cultural biases is higher.

212 Background: The HER2 oncogene serves as a prognostic marker in breast cancer and is a therapeutic target in breast, lung, and GI malignancies, including cancers with low Her2 overexpression. A prior study by Minner et al ( Clin Cancer Res, 2010) evaluating Her2 overexpression in prostate cancer (PCa) showed a prevalence ~20%. The goal of this study is to evaluate the prevalence and clinical significance of Her2 overexpression in PCa in a predominantly African American (AA) cohort. Methods: 124 PCa patients managed at the Washington DC VA Medical Center between 2000-2021 were randomly selected: 35 indolent (AJCC stage I), 46 locally advanced (AJCC stages II&III), 19 locally advanced at diagnosis but progressed to metastatic (AJCC stage IV), and 24 de novo metastatic (AJCC stage IV). Immunohistochemistry (IHC) for Her2 was performed on one representative tissue section from core biopsies or a radical prostatectomy for each case. IHC intensity was scored independently by two experienced pathologists as negative (0, no staining), low positive (1+, faint membranous staining), moderate positive (2+, weak to moderate complete, basolateral or lateral membranous staining) or strong positive (3+, strong complete, basolateral or lateral membranous staining). Fisher’s exact test was used to test the association between Her2 expression and categorical variables. Results: 108 patients (87%) self-identified as AA. The mean age at diagnosis was 62.9 years. The prevalence of positive Her2 expression (1+, 2+,3+) was 48% (59/124) in the entire cohort, 29% (10/35) in the indolent group, 48% (22/46) in the locally advanced group, 68% (13/19) in the locally advanced group that progressed to metastatic disease, and 58% (14/24) in the de novo metastatic group ( p=0.0014). Prevalence of Her2 positive expression was 28% (10/36) in patients with a Gleason Score (GS) of 6, whereas in patients with a GS of 9 it was 78% (9/11) ( p 2.0, SD 0.78), followed by the locally advanced group that progressed to metastatic disease (Her2 1.54, SD 0.66), the locally advanced group without metastatic disease (Her2 1.36, SD 0.58), and lowest in the indolent group (Her2 1.0, SD 0.0) ( p=0.0016). Next generation sequencing data was available for 20 metastatic patients (12 positive for Her2 expression) and did not show the presence of HER2 mutations or amplifications. Conclusions: In this predominantly AA cohort, expression of Her2 was identified in almost 50% of patients with PCa. Her2 scores ≥2 were associated with higher GS and advanced disease. Moreover, a Her2 score of 3 was seen in 6 patients in our cohort. Historic studies on targeting Her2 in advanced PCa have failed to demonstrate clinical benefit. However, in the era of targeted therapy, Her2 inhibition addressing low-levels of Her2 overexpression should be considered in future trials.

211 Background: Receptor-Interacting Protein Kinase 2 (RIPK2) has been known to play a significant role in inflammation and anti-microbial response via the NF-KB pathway. Recently, RIPK2 was also shown to stabilize and activate c-Myc, a molecule implicated in the progression of various malignancies including prostate cancer (PCa). Our study aims to determine the prognostic value of RIPK2 expression in PCa using immunohistochemistry (IHC) analysis of biopsy specimens with special focus on a predominantly African American (AA) cohort. Methods: Eligible patients aged ≥18 years diagnosed with PCa at the Washington DC VA Medical Center were randomly selected and retrospectively grouped into indolent, locally advanced, and metastatic disease categories and stratified by age, ethnicity, AJCC stage, PSA and Gleason score at diagnosis. Representative tissue from prostate core biopsy, transurethral resection, or radical prostatectomy was analyzed for each patient. RIPK2 expression and intensity were assessed by IHC and graded by two independent pathologists by the following scale: 0 (no cytoplasmic staining), 1+ (weak cytoplasmic staining similar to benign prostatic tissue), 2+ (moderate cytoplasmic staining), and 3+ (strong cytoplasmic staining). The primary endpoint was RIPK2 expression by disease group, and key secondary outcomes were RIPK2 association with Gleason score, PSA at diagnosis, age and race. Fisher’s exact test was used to evaluate statistical significance of the association between RIPK2 expression level and disease category. Results: A total of 89 patients were selected for analysis, of which 75 (84.3%) were AA. All tumors had at least 1+ level RIPK2 expression and there was a highly significant difference in RIPK2 expression according to disease severity (p < 0.001). Stratified RIPK2 expression was 1.53, 1.83, and 2.28 in the indolent, locally advanced, and metastatic groups respectively. We observed a significant directly proportional correlation between RIPK2 expression and a higher PSA level (p= 0.0021), Gleason score (p< 0.001), higher age (p= 0.0276), and AJCC stage (p< 0.001) at the time of diagnosis. Mean RIPK2 expression was 1.81 in AA patients and 2.15 in White patients (p= 0.0332). Conclusions: PCa is the most prevalent cancer in US males and there is an urgent need for more accurate biomarkers that can predict disease severity and suggest actionable targeted therapy. In this retrospective analysis, we confirm that higher RIPK2 expression is associated with more advanced PCa. We extend these findings to a predominantly AA cohort, which may promote improvements in personalized medicine for this population at higher risk for PCa if RIPK2-targeting drugs such as Ponatinib can be utilized.

217 Background: Increased utilization of next-generation sequencing has revealed evidence of early TP53 alterations in localized prostate cancer (PCa), suggesting a growing clinical potential for p53 immunohistochemistry (IHC). Prior research by Harmon and Gesztes et al. (Radiology, 2021 & Scientific Reports, 2022) support the clinical utility of detecting p53 overexpression in core biopsies and radical prostatectomies (RP) to predict underlying TP53 alterations, biochemical recurrence (BCR), and metastasis. However, to date, no established consensus exists evaluating p53 “overexpression” in PCa. This study aimed to compare two fundamental approaches to assess p53 expression across various specimens with regard to PCa progression. A secondary study aim explored whether ethnic differences in p53 expression exist. Methods: This retrospective cohort study comprised 84 randomly selected PCa patients (46 metastatic; 38 non-metastatic) enrolled between 1996 – 2021 at the DC VA Medical Center. Representative sections of core biopsies (60%), RP (17%), prostate chips (7%), and metastatic deposits (16%) were examined, where (%) equals a share of the total cohort. IHC was performed on selected sections with a p53 monoclonal antibody (Biocare Medical; p53 Tumor Suppressor Protein; clone DO-7). p53 nuclear expression was manually scored according to the highest intensity (0 absent, 1+ weak, 2+ moderate, or 3+ marked) and percentage (0%, 5%) of tumor cell nuclei observed in one index block per patient. All slides were reviewed by two pathologists who were blinded to clinical data. Fisher’s exact test was used to test the association between p53 positivity and categorical variables. Results: Of 84 men included, 63 (75%) self-identified as African American (AA). The median follow-up time was 6.2 years. Thirty-four patients (40%) exhibited p53 nuclear expression, of which 18 patients (21%) showed marked (3+) nuclear intensity. Notably, 11 of 18 (61%) 3+ patients displayed 5% p53 positivity showed marked (3+) p53 staining. The presence of marked p53 staining, regardless of percentage, was found to be significantly associated with a higher Gleason score ( p=0.0002), higher PSA at the time of biopsy ( p

Small bowel adenocarcinoma (SBA) is a rare neoplasm that is associated with Crohn's disease (CD). This study aims to quantify the prevalence of CD-SBA, review the current evidence of histopathology and molecular analysis findings, and identify the clinical presentation and outcomes of CD-SBA.Electronic databases Medline and Embase were searched for articles describing SBA in inflammatory bowel disease patients. The histopathology, molecular analysis findings, clinical presentation, prevalence, and outcomes of CD-SBA were extracted, and results were pooled with random effects.In total, 33 articles were included in the analysis. Prevalence of SBA was 1.15 (CI: 0.31-2.33) per 1000 CD patients. Only 11% (CI: 0.04-0.21) of CD-SBA patients had observable radiological features. CD-SBA was most commonly found in the ileum (84%), diagnosed at stage 2 (36%), with main presenting complaints including obstruction, weight loss, and abdominal pain. Significant histopathological findings included adjacent epithelial dysplasia, and an equal distribution of well-differentiated (49%) and poorly differentiated subtypes (46%). Most prevalent genetic mutation was KRAS mutation (18%), followed by mismatch repair deficiency (9.7%). The 5-year overall survival for CD-SBA patients was 29% (CI: 0.18-0.41), and 33% (CI: 0.26-0.41) for de novo SBA. No statistically significant increase in risk for CD-SBA was noted for treatment with thiopurines, steroids, and 5-ASA.Our meta-analysis found the prevalence of CD-SBA to be 1.15 per 1000 CD patients. The 5-year overall survival for CD-SBA was poor. The presenting symptoms were non-specific, and therefore the diagnosis requires a high index of suspicion.

Readability is on the cusp of a revolution. Fixed text is becoming fluid as a proliferation of digital reading devices rewrite what a document can do. As past constraints make way for more flexible opportunities, there is great need to understand how reading formats can be tuned to the situation and the individual. We aim to provide a firm foundation for readability research, a comprehensive framework for modern, multi-disciplinary readability research. Readability refers to aspects of visual information design which impact information flow from the page to the reader. Readability can be enhanced by changes to the set of typographical characteristics of a text. These aspects can be modified on-demand, instantly improving the ease with which a reader can process and derive meaning from text. We call on a multi-disciplinary research community to take up these challenges to elevate reading outcomes and provide the tools to do so effectively., This paper was generated collaboratively over the course of a series of online workshops, the results of which were extensively edited by Dr. Zoya Bylinskii, Dr. Ben D. Sawyer, and Dr. Benjamin Wolfe. Original illustrations by Bernard Kerr. Corresponding Author: Dr. Ben D. Sawyer

We propose SelfDoc, a task-agnostic pre-training framework for document image understanding. Because documents are multimodal and are intended for sequential reading, our framework exploits the positional, textual, and visual information of every semantically meaningful component in a document, and it models the contextualization between each block of content. Unlike existing document pre-training models, our model is coarse-grained instead of treating individual words as input, therefore avoiding an overly fine-grained with excessive contextualization. Beyond that, we introduce cross-modal learning in the model pre-training phase to fully leverage multimodal information from unlabeled documents. For downstream usage, we propose a novel modality-adaptive attention mechanism for multimodal feature fusion by adaptively emphasizing language and vision signals. Our framework benefits from self-supervised pre-training on documents without requiring annotations by a feature masking training strategy. It achieves superior performance on multiple downstream tasks with significantly fewer document images used in the pre-training stage compared to previous works., To appear in CVPR'2021

Introduction: This article by means of data analysis obtained from our study have tried to evaluate feasibility and acceptability of laparoscopic appendicectomy in Indian scenario. Objective: Through this prospective study an effort have been made to find ideal technique for appendicectomy which can serve masses in low middle income group countries like India where patient load outnumber resources both in expertise and facilities. Methods: A prospective analysis of two groups of population which have undergone appendicectomy (laparoscopic versus open conventional) was done as per inclusion criteria. Results obtained were analysed on grounds of cost benefit, time taken for surgery, pain, post-operative complications, hospital stay, time to resume routine activities, subjective cosmesis and finally patients preference. Data obtained was statistically analysed. Results: The mean age of patients in laparoscopic group was 22.5 years and 27.6 years in open appendicectomy group. Out of 287 patients included as per inclusion criteria 69.69% (200) patients opted for open appendicectomy. Mean operative time for laparoscopic procedure was more than open conventional procedure (98 minutes compared to 51 minutes). Post-operative complications and median length of hospital stay were comparable in both groups. There was no significant statistical difference related to pain, time duration to resume routine activities, and subjective cosmesis. Cost of overall procedure in laparoscopic appendicectomy was significantly higher than open conventional group. Conclusion: conventional open appendicectomy is reliable, easy, safe and cost-effective technique and its use can dramatically reduce cost burden of surgery as compared to laparoscopic surgery.

As reading on mobile devices is becoming more ubiquitous, content is consumed in shorter intervals and is punctuated by frequent interruptions. In this work, we explore the best way to mitigate the effects of reading interruptions on longer text passages. Our hypothesis is that short summaries of either previously read content (reviews) or upcoming content (previews) will help the reader re-engage with the reading task. Our target use case is for students who study using electronic textbooks and who are frequently mobile. We present a series of pilot studies that examine the benefits of different types of summaries and their locations, with respect to variations in text content and participant cohorts. We find that users prefer reviews after an interruption, but that previews shown after interruptions have a larger positive influence on comprehension. Our work is a first step towards smart reading applications that proactively provide text summaries to mitigate interruptions on the go., Comment: 6 pages, 4 figures, to appear as a Late Breaking Work in CHI 2021, see "https://chi2021.acm.org/"

Introduction Learning research in medical school can be daunting and difficult with many lacking appropriate guidance. Particularly, research in surgery can be difficult with most clinicians tied to long hours from clinical duties, leaving medical students lost to learn the ropes for medical research. Method We started a surgical research group in January of 2020 under a shepherd sheep model. Shepards were senior medical students who had previous experience in research and were provided resources from consultants on statistics and research designs. Thereafter, each Shepard were paired with junior medical students (sheep) and a resident to provide clinical background. Research was then carried out small teams with minimal guidance from consultants. Results To date, the group has published 11 articles (median impact factor: 2.41, range: 1.89 - 3.42), 3 articles in revision and 9 articles in review. Conclusions We provide preliminary evidence of a successful model for building medical student research in surgery. While current analysis was limited to meta-analysis and systematic review due to the availability of data, current expansion is currently underway to expand to observational studies.

Introduction Post-operative recurrence after Crohn’s surgery poses a significant challenge to clinicians with an estimated of 70-90% recurrence rate. However, with the advent of biologics, reduction of recurrence after surgery has been observed in various landmark trials. Hence, we aim to quantify the rate of recurrence with post-operative biologics. Method Searches were done on Embase and Medline database for articles using biologics for post-operative management. A single arm meta-analysis with generalized linear mixed model and Clopper-Pearson method confidence interval was applied in the synthesis of the data on six months, one year, two year and five year. Results A total of 1,864 abstracts were identified, with a final total of 24 articles and 1042 patients were included in the meta-analysis. The one-year rate of recurrence for surgical, endoscopic and clinical was 3.08% (CI: 0.74% - 11.95%), 19.93% (15.81% - 24.81%) and 13.06% (CI: 8.18% - 18.92%) respectively. Correspondingly, the five-year recurrence 16.90% (CI: 9.57% - 28.11%), 84.21% (CI: 72.35% - 91.57%), 5.60% (CI: 9.92% -23.66%) respectively. Conclusions An observed reduction in surgical rates was reduced at five years with post-operative Crohn’s disease with biologics. Biologics therapy thus can be considered as a suitable alternative for reducing post-operative recurrence in individuals with higher risk.

Introduction Gender discrimination remains pervasive in surgery, significantly impacting current and future surgeons and population health. This study analyses the gender barriers and critical retention factors for female surgeons and trainees in surgery. Method Five electronic databases were searched till May 2020. Titles and abstracts were sieved, followed by a full text review. Data synthesis and inductive thematic analysis were conducted using the Thomas and Harden methodology. Result 14 articles were included, involving 528 participants. Four themes were generated–unfavourable working environment, male-dominated culture, societal pressures and progress towards gender equality. Females in surgery often faced harassment, disrespect and perceptions of incompetence, resulting in hostile work conditions, which were aggravated by the inadequate support and mentorship. The persistence of male-dominated cultures was observed, with females facing prejudice and exclusion from professional and social circles. Differential treatment and higher expectations of female surgeons also arose from entrenched societal pressures. Despite these, increased acceptance of motherhood and greater recognition of contributions by female surgeons were reported, indicating some progress in gendered culture. Conclusions There is a need to increase female surgical leadership and allocate resources to address the deep-rooted causes of biased surgical culture and ingrained perceptions, to achieve greater gender equality in surgery.

Background Prevention of post-operative recurrence of Crohn's disease poses a significant challenge to clinicians. With the advent of biologics, reduction of recurrence after surgery has been observed in various studies. Hence, we performed a systematic review and meta-analysis to identify the rate of post-operative recurrence at different timepoints in the era of anti-TNF-α therapy biologics use. Methods We performed a literature search using Medline and Embase databases for studies investigating biologics in preventing post-operative recurrence of Crohn's disease. Data were extracted, and a single arm meta-analysis with generalized linear mixed model and Clopper-Pearson method for confidence interval was performed to identify the endoscopic, clinical, and surgical recurrence rates at six-months, one-, two- and five-years post-operatively. Results 24 articles were included in the meta-analysis. The endoscopic, clinical, and surgical postoperative recurrence rate with the use of anti-TNF-α agents at one-year was 21.72% (95%CI: 16.28% - 28.37%), 13.06% (95%CI: 8.18% - 18.92%) and 3.76% (95%CI: 1.37% - 9.91%) respectively. The five-year recurrence rate was 84.21% (95%CI: 72.35% - 91.57%) and 17.49% (95%CI: 9.17% - 30.80%) for endoscopic and surgical recurrence respectively. Subgroup analysis at one-year for type of anti-TNF-α agent or timing of initiation after surgery showed no significant difference in endoscopic, clinical and surgical recurrence rate. Conclusion Anti-TNF-α agents are effective at preventing clinical, endoscopic, and surgical postoperative Crohn's disease recurrence. The timing of initiating biological therapy after surgery has no significant effect on the rate of postoperative recurrence. The efficacy of infliximab and adalimumab for postoperative recurrence prevention is similar.

Aims Upper gastrointestinal Crohn's disease (UGI-CD) is an important subclassification of Crohn's Disease (CD). We performed a systematic review and meta-analysis to evaluate the prevalence, risk factors, and clinical outcomes associated with UGI-CD. Methods We searched Embase and Medline for articles reporting the clinical information of UGI-CD in CD patients, through 27 October 2020. Disease location and phenotype were coded according to the Montreal classification, and results were pooled with random effects by DerSimonian and Laird model. Results 26 articles were included. The prevalence of UGI-CD was 13%. UGI-CD was most commonly found in the stomach (56%) and was associated with concurrent ileocolonic involvement (54%). Non-stricturing, non-penetrating UGI-CD was the most common behavioral phenotype (61%). L4-jejunal disease was associated with the highest rates of surgery. Region of origin did not significantly influence the location and phenotype of UGI-CD. Young, male patients presenting with erythema nodosum, aphthous ulcers and stricturing-phenotype are more likely to have UGI-CD, which in turn is linked to increased risk of hospitalization and surgery. Conclusion UGI-CD is present in 13% of patients with CD, and patients with L4-jejunal disease are more likely to require surgery. Further studies examining the effect of ethnicity and region on UGI-CD are needed.

Recent studies indicate that NLU models are prone to rely on shortcut features for prediction, without achieving true language understanding. As a result, these models fail to generalize to real-world out-of-distribution data. In this work, we show that the words in the NLU training set can be modeled as a long-tailed distribution. There are two findings: 1) NLU models have strong preference for features located at the head of the long-tailed distribution, and 2) Shortcut features are picked up during very early few iterations of the model training. These two observations are further employed to formulate a measurement which can quantify the shortcut degree of each training sample. Based on this shortcut measurement, we propose a shortcut mitigation framework LTGR, to suppress the model from making overconfident predictions for samples with large shortcut degree. Experimental results on three NLU benchmarks demonstrate that our long-tailed distribution explanation accurately reflects the shortcut learning behavior of NLU models. Experimental analysis further indicates that LTGR can improve the generalization accuracy on OOD data, while preserving the accuracy on in-distribution data., Accepted by NAACL 2021

ObjectiveThis study was conducted to better understand the pervasive gender barriers obstructing the progression of women in surgery by synthesising the perspectives of both female surgical trainees and surgeons.MethodsFive electronic databases, including Medline, Embase, PsycINFO, CINAHL and Web of Science Core Collection, were searched for relevant articles. Following a full-text review by three authors, qualitative data was synthesized thematically according to the Thomas and Harden methodology and quality assessment was conducted by two authors reaching a consensus.ResultsFourteen articles were included, with unfavorable work environments, male-dominated culture and societal pressures being major themes. Females in surgery lacked support, faced harassment, and had unequal opportunities, which were often exacerbated by sex-blindness by their male counterparts. Mothers were especially affected, struggling to achieve a work-life balance while facing strong criticism. However, with increasing recognition of the unique professional traits of female surgeons, there is progress towards gender quality which requires continued and sustained efforts.ConclusionThis systematic review sheds light on the numerous gender barriers that continue to stand in the way of female surgeons despite progress towards gender equality over the years. As the global agenda towards equality progresses, this review serves as a call-to-action to increase collective effort towards gender inclusivity which will significantly improve future health outcomes.

After almost three decades since the first description of colonic stents, the controversies of its safe application continue to impede the readiness of adoption by clinicians for malignant left bowel obstruction. This review seeks to address some of the controversial aspects of stenting and its impact on surgical and oncological outcomes. Medline, Embase, and CNKI were searched for articles employing SEMS for left colonic obstruction. Outcomes analyzed include success rates, complications, and long-term survival. Pooled risk ratio (RR) and 95% confidence interval (CI) were estimated. 36 studies were included with 2002 patients across seven randomized controlled trials and 29 observational studies. High technical (92%) and clinical (82%) success rates, and low rates of complications, including perforation (5%), were found. Those with > 8% perforation rates had poorer technical success rates than those with ≤ 8%, but there were no significant differences in 90-day in-hospital mortality and three and 5-year overall and disease-free survival. A significant increase was found in technical (RR = 1.094; CI, 1.041–1.149; p

We present TIMERS - a TIME, Rhetorical and Syntactic-aware model for document-level temporal relation classification in the English language. Our proposed method leverages rhetorical discourse features and temporal arguments from semantic role labels, in addition to traditional local syntactic features, trained through a Gated Relational-GCN. Extensive experiments show that the proposed model outperforms previous methods by 5-18% on the TDDiscourse, TimeBank-Dense, and MATRES datasets due to our discourse-level modeling.

Event Argument Extraction (EAE) is the task of identifying roles of entity mentions/arguments in events evoked by trigger words. Most existing works have focused on sentence-level EAE, leaving document-level EAE (i.e., event triggers and arguments belong to different sentences in documents) an under-studied problem in the literature. This paper introduces a new deep learning model for document-level EAE where document structures/graphs are utilized to represent input documents and aid the representation learning. Our model employs different types of interactions between important context words in documents (i.e., syntax, semantic, and discourse) to enhance document representations. Extensive experiments are conducted to demonstratethe effectiveness of the proposed model, leading to the state-of-the-art performance for document-level EAE.