Search

Bidirectional Encoder Representations from Transformers (BERT) and BERT-based approaches are the current state-of-the-art in many natural language processing (NLP) tasks; however, their application to document classification on long clinical texts is limited. In this work, we introduce four methods to scale BERT, which by default can only handle input sequences up to approximately 400 words long, to perform document classification on clinical texts several thousand words long. We compare these methods against two much simpler architectures - a word-level convolutional neural network and a hierarchical self-attention network - and show that BERT often cannot beat these simpler baselines when classifying MIMIC-III discharge summaries and SEER cancer pathology reports. In our analysis, we show that two key components of BERT - pretraining and WordPiece tokenization - may actually be inhibiting BERT's performance on clinical text classification tasks where the input document is several thousand words long and where correctly identifying labels may depend more on identifying a few key words or phrases rather than understanding the contextual meaning of sequences of text.

Evaluating the knowledge of patients attending radiology departments regarding ionizing radiation used in medical imaging and its associated hazards can provide knowledge of the patient's awareness level of the associated risk of the radiation used in medical imaging. The aims of this study were to evaluate the awareness of patients regarding medical radiation types used in medical diagnostic imaging and its influence on their decision to proceed with that procedure. Over an 8-months period, a total of 418 patients, 48% Men and 52% Women, presenting for diagnostic imaging in the department of radiology, completed a 15-point questionnaire. The questionnaire included demographic and radiation awareness sections. Less than 32% of the participants had a potential risk of radiation explained by the doctor before the procedure. 59% of the participants expressed that the potential risk of radiation makes them anxious; less than about 25% of the participants showed that the potential risk of radiation affects their decision to have the procedure. Overall, the data collected from this survey indicate that there is a lack of information about radiation risk provided to the patients prior to the diagnostic procedure. Efforts should be made to ensure that patients receiving multiple medical imaging tests are aware of the radiation they are receiving.

Kidney cancer, with 4% of all malignancies, is one of the most common malignancies occurring among in adults. In Saudi Arabia, kidney cancer comprises 2.3% of all cancers, and its incidence has increased by 33%. Partial nephrectomy (PN) is considered as the gold standard for T1 renal masses.In this retrospective study, we did a chart review for all patients who underwent PNs between April 2013 and February 2019. Data comprised pre-sentation, tumor size, type of procedure (open vs. laparoscopic vs. robotic), and intra- and post-operative complications. Chi-square, ANOVA, and cross-tabulation were done using SPSS software. P > 0.05 was considered significant. Approval was obtained from the institutional review board of King Abdullah International Medical Research Center.In all, 69 patients were identified: 26 (37.7%) males and 43 (62.3%) females, with mean age = 54.53 ± 13.21 years; mean body mass index = 32.36 ± 7.03, and mean tumor size = 3.7 ± 1.72 cm. In terms of presentation, most patients (50, 72.4%) presented incidentally as opposed to symptomatic presentation. Of these patients, 18 (26.1%) underwent open partial nephrectomy (OPN), 29 (42%) laparoscopic partial nephrec-tomy (LPN), and 22 (31.9%) robotic partial nephrectomy (RPN). On comparing minimally invasive surgery (MIS) PN with OPN, we found that OPN had more blood loss and a longer hospital stay but a shorter operating room (OR) time.Results of PN irrespective of the procedure type, whether it was OPN, LPN, or RPN, were similar if performed by experienced surgeons. However, open procedures involved a higher blood loss, more operative time, and longer hospital stay when compared with minimally invasive techniques.

Early in the SARS-CoV-2 pandemic, convalescent plasma (CP) therapy was proposed as a treatment for severely ill patients. We conducted a CP treatment protocol under the Mayo Clinic Extended Access Program at University Hospital Brooklyn (UHB). Potential donors were screened with a lateral flow assay (LFA) for IgM and IgG antibodies against the SARS-CoV-2 S1 receptor-binding domain (RBD). Volunteers that were LFA positive were tested with an ELISA to measure IgG titers against the RBD. Subjects with titers of at least 1:1024 were selected to donate. Most donors with positive LFA had acceptable titers and were eligible to donate. Out of 171 volunteers, only 65 tested positive in the LFA (38.0%), and 55 (32.2%) had titers of at least 1:1024. Before our donation program started, 31 CP units were procured from the New York Blood Center (NYBC). Among the 31 CP units that were obtained from the NYBC, 25 units (80.6%) were positive in the LFA but only 12 units (38.7%) had titers of at least 1:1024. CP was administered to 28 hospitalized COVID-19 patients. Patients who received low titer CP, high titer CP and patients who did not receive CP were followed for 45 days after presentation. Severe adverse events were not associated with CP transfusion. Death was a less frequent outcome for patients that received high titer CP (>1:1024) 38.6% mortality, than patients that received low titer CP (≤1:1024) 77.8% mortality.

Background The incidence of ectopic kidneys is 1:12,000 clinically and 1:900 postmortem. Patients with pelvic mal-rotated kidneys are more susceptible to recurrent urinary tract infections, recurrent renal stones, and renal injury. Fusion of the kidney lower poles is relatively common compared to other types of renal anomalies. Case presentation We present the case of a 36-year-old Sudanese female patient who presented with a long history of recurrent urinary tract infections unresponsive to antibiotics. Ultrasound scan revealed bilateral pelvic kidneys. Computed tomography (CT) urography confirmed bilateral ectopic fused kidneys, with the left kidney mal-rotated (renal pelvis facing upwards and laterally). Kidney infection secondary to vesicoureteral reflux was diagnosed. Antibiotics were prescribed according to culture and sensitivity. The patient responded well to ciprofloxacin. Conclusion A history of recurrent urinary tract infections without an apparent cause is highly suggestive of renal anomaly and should be investigated expediently. Ultrasonography or CT imaging may be utilized to aid in diagnosis. Early recognition may help prevent the high risk of end-stage renal failure associated with anomalies.

BackgroundAutomated text classification has many important applications in the clinical setting; however, obtaining labelled data for training machine learning and deep learning models is often difficult and expensive. Active learning techniques may mitigate this challenge by reducing the amount of labelled data required to effectively train a model. In this study, we analyze the effectiveness of 11 active learning algorithms on classifying subsite and histology from cancer pathology reports using a Convolutional Neural Network as the text classification model.ResultsWe compare the performance of each active learning strategy using two differently sized datasets and two different classification tasks. Our results show that on all tasks and dataset sizes, all active learning strategies except diversity-sampling strategies outperformed random sampling, i.e., no active learning. On our large dataset (15K initial labelled samples, adding 15K additional labelled samples each iteration of active learning), there was no clear winner between the different active learning strategies. On our small dataset (1K initial labelled samples, adding 1K additional labelled samples each iteration of active learning), marginal and ratio uncertainty sampling performed better than all other active learning techniques. We found that compared to random sampling, active learning strongly helps performance on rare classes by focusing on underrepresented classes.ConclusionsActive learning can save annotation cost by helping human annotators efficiently and intelligently select which samples to label. Our results show that a dataset constructed using effective active learning techniques requires less than half the amount of labelled data to achieve the same performance as a dataset constructed using random sampling.

A best evidence topic has been constructed using a described protocol. The three-part question addressed was: In patients who underwent appendicectomy for uncomplicated appendicitis is the use of postoperative antibiotics associated with lower rates of surgical site infections? The search has been devised and 6 studies were deemed to be suitable to answer the question. The outcome assessed was the efficiency of postoperative antibiotic therapy in decreasing the rate of surgical site infections in uncomplicated appendicitis. Authors recommend against the use of postoperative antibiotics based on the supported evidence. Hence, its usage was not associated with lower rates of surgical site infections. On the contrary, it might increase the cost, postoperative morbidity and length of stay.

A best evidence topic has been constructed using a described protocol. The three-part question addressed was: does the anastomotic technique after ileocolic resection affects the recurrence rate in patients with Crohn's disease? Using OVID interface and PubMed interface, 16 articles were found; out of this 6 studies were deemed to be suitable to answer the question. The outcomes assessed were anastomotic recurrence rate. The best evidence showed that Kono-S ileocolic anastomotic technique is associated with significantly low recurrence rate in patient with Crohn's disease., Highlights • Kono-S ileocolic anastomotic technique. • Low recurrence rate. • Crohn's disease.

Objectives The decline in diarrhoeal disease-related mortality globally has been attributed to the use of oral rehydration solution (ORS) and zinc supplementation. However, data on ORS and zinc supplementation in Sudan are scarce. We aimed to investigate the access to ORS and zinc treatments and the associated factors, through the analysis of the latest available data from Sudan-Multiple Indicator Cluster Survey (MICS)-2014 obtained from the United Nations Children’s Fund (UNICEF). Results A total of 14,081 children were included in this analysis. During the 2 weeks preceding the survey, 29.3% of these children had a diarrhoeal disease. Only 18.9% and 14.8% of these children had received ORS and zinc supplements, respectively. Whereas children from the higher wealth index groups were more likely to receive ORS treatment (fourth group: AOR = 1.301; 95% CI 1.006–1.682), children from rural areas were less likely to receive ORS treatment (AOR = 0.666; 95% CI 0.552–0.803) and zinc supplements (AOR = 0.603; 95% CI 0.500–0.728). The results indicate the existence of unequal access to treatment of childhood diarrhoeal diseases among children under 5 years in Sudan.

Background COVID-19 pandemic is presenting serious challenges to the world’s healthcare systems. The high communicability of the COVID-19 necessitates robust medical preparedness and vigilance. Objective To report on the simulation-based training and test preparedness activities to prepare healthcare workers (HCWs) for effective and safe handling of patients with COVID-19. Methodology Two activities were conducted: simulation-based training to all HCWs and a full-scale unannounced simulation-based disaster exercise at King Fahad Medical City (KFMC). The online module was designed to enhance the knowledge on COVID-19. This module was available to all KFMC staff. The five hands-on practical part of the course was available to frontliner HCWs. The unannounced undercover simulated patients’ full-scale COVID-19 simulation-based disaster exercise took place in the emergency department over 3 hours. Six scenarios were executed to test the existing plan in providing care of suspected COVID-19 cases. Results 2620 HCWs took the online module, 17 courses were conducted and 337 frontliner HCWs were trained. 94% of learners were satisfied and recommended the activity to others. The overall compliance rate of the full-scale COVID-19 disaster drill with infection control guidelines was 90%. Post-drill debriefing sessions recommended reinforcing PPE training, ensuring availability of different sizes of PPEs and developing an algorithm to transfer patients to designated quarantine areas. Conclusion Simulation-based training and preparedness testing activities are vital in identifying gaps to apply corrective actions immediately. In the presence of a highly hazardous contagious disease like COVID-19, such exercises are a necessity to any healthcare institution.

Objective:The aim of this study was to evaluate the readiness of a tertiary medical cityʼs response to a disaster by assessing the hospital resources and knowledge, attitudes, practices, and familiarity of health care providers toward disaster and emergency preparedness.Methods:All KFMC (King Fahad Medical City) staff with > 1 year of clinical experience were eligible to participate in a cross-sectional study. Participants responded to the Emergency Preparedness Information Questionnaire (EPIQ), knowledge and practice questionnaires, and a disaster planning attitude checklist. Data about resources were collected using the hospital disaster preparedness self-assessment tool.Results:The overall mean knowledge score for disaster and emergency preparedness was 4.4 ± 1.1, and the mean overall familiarity score was 3.43 ± 0.97. Most participants knew that disaster drills (90.2%) and training (74.6%) are ongoing. Sixty-six (21.0%) agreed that KFMC is unlikely to experience a disaster. The highest and lowest EPIQ familiarity scores were for decontamination (83.0%) and accessing critical resources and reporting (64.3%), respectively. Most participants (99.4%) have access to work computers; however, only 53.0% used the Internet to access information on bioterrorism and/or emergency preparedness. The hospital is ready to respond in case of a disaster according to the used tool.Conclusions:The participants’ levels of knowledge, practices, and overall familiarity toward emergency and disaster preparedness were satisfactory; however, participant attitudes and familiarity with where and how to access critical resources in the event of an emergency or disaster situations require reinforcement.

Highlights • Laparoscopic adjustable gastric banding (LAGB) was considered as one of the most effective management for morbid obesity, with outstanding long-term results on weight loss, comorbidities, and quality of life. • Reported complications of LAGB include erosion, infection, migration, obstruction, and rarely ischemia. • We report a case of pregnant woman who underwent LAGB 12 years ago diagnosed as gastric obstruction and perforation that was managed by central gastrectomy with gastro-gastrostomy. • We encourage the bariatric surgeons to be attentive LAGB complications among women planning for pregnancy., Background The laparoscopic adjustable gastric band is the least invasive bariatric surgery, which has the advantage of preserving the anatomy of the gastrointestinal tract. In the last two decades, the laparoscopic adjustable gastric band (LAGB) has largely been replaced by other surgical techniques for weight loss because of its high inseparable complications. Although a popular surgical technique at the time of its introduction in 1993, LAGB nowadays reports for not more than 5.5% of all bariatric procedures. The estimated overall long term complication rates of LAGB are ranging from 0.1% to 28%. Objective In this report, we review one of the cases that are considered as an anecdotic event that might follow one of the bariatric procedures. Methods We reported a pregnant lady who undergoes gastric banding 12 years ago. presented with progressively increasing colicky epigastric pain, nausea, and vomiting. The investigations show gastric obstruction and perforation secondary to eroded gastric band and migration. Results The management was completed by the elimination of the gastric band proceed with central gastrectomy plus gastro-gastrostomy with an excellent outcome for the patient and her baby. Conclusion This case highlights an absolutely rare serial complication, that may happen even when the patient presents with a vague complaint. Intraoperative findings can determine the way of management to achieve suitable results. Lastly; we encourage the young Women who underwent or planning for bariatric surgery to do a careful follow-up.

Neuroendocrine neoplasms of the gallbladder are rare, comprising 0.5% of all neuroendocrine cancers and about 2% of gallbladder cancers. These neoplasms can also be found along with other malignant neoplasms of epithelial origin, mostly adenocarcinomas. Herein, we describe an unusual finding of a three-component mixed neuroendocrine non-neuroendocrine neoplasm (MiNEN) of the gallbladder. We also review the literature on 29 similar cases and summarize key features. We report on a 62-year-old woman who presented with right upper quadrant pain with a positive Murphy’s sign. A clinical diagnosis of neoplasia was entertained and she underwent cholecystectomy. Gross examination of the specimen revealed a 5-cm exophytic mass at the gallbladder fundus. Histopathologic examination of the mass showed an infiltrating squamous cell carcinoma, an adjacent neuroendocrine carcinoma (each of these two components composed more than 30% of the neoplasm), and a superficial adenocarcinoma (composing 10% of the neoplasm). Gallbladder MiNENs present with similar symptoms and in the same age group as do carcinomas; however, their prognosis is often poor. Specific management and treatment guidelines have not been established since MiNENs are very rare.

A best evidence topic has been constructed using a described protocol. The three-part question addressed was: in patient with Whipple's procedure which anastomotic technique has lower leak rate pancreaticogastostomy (PG) or pancreatojejunostomy (PJ)? Using the reported search, 38 articles were found; out of this six studies were deemed to be suitable to answer the question. The outcomes assessed were incidence of anastomotic leaks (pancreatic fistula) in both techniques PG and PJ. In conclusion, the best evidence showed that PG anastomosis has lower incidence of pancreatic fistula in comparison to PJ anastomosis., Highlights • Aim: Comparison of pancreatic fistula rate among pancreaticogastostomy (PG) vs pancreatojejunostomy (PJ) anastomosis. • Result: Six randomized controlled trials were included. • Conclusion: The best evidence showed that PG anastomosis has lower incidence of pancreatic fistula than PJ anastomosis.

Background: Sclerosing hemangiomas of the liver are rare, benign tumors with degenerative changes. These degenerative changes, however, often obscure the true, benign nature of the tumor and give them features indistinguishable from other malignant processes, thus making the diagnosis difficult. Case presentation: A 70-year-old male without any previously diagnosed liver disease or malignant process presented with incidental right hepatic mass in ultrasonography and weight loss. Physical exam was unremarkable. The labs were significant for mild pancytopenia, elevated total bilirubin and slightly decreased transferrin. Follow-up triple phase-contrast CT scan of the abdomen revealed a lobulated, poorly demarcated lesion measuring 4.8 x 4.5 cm, located in segment V of the liver with encasement of the left portal branch. The overall picture was indeterminate but highly suspicious for malignancy. A decision was made to perform a CT-guided biopsy which revealed sclerosing hemangioma of the liver. Conclusion: It is challenging to differentiate sclerosing hepatic hemangioma from atypical hepatocellular carcinoma, intrahepatic cholangiocarcinoma, and metastatic tumors utilizing only imaging modalities. The diagnostic workup should include biopsy of the atypical liver lesion which unveils the final diagnosis and avoid subjecting the patient to an extensive, and invasive surgical resection.

Objective We implement 2 different multitask learning (MTL) techniques, hard parameter sharing and cross-stitch, to train a word-level convolutional neural network (CNN) specifically designed for automatic extraction of cancer data from unstructured text in pathology reports. We show the importance of learning related information extraction (IE) tasks leveraging shared representations across the tasks to achieve state-of-the-art performance in classification accuracy and computational efficiency. Materials and Methods Multitask CNN (MTCNN) attempts to tackle document information extraction by learning to extract multiple key cancer characteristics simultaneously. We trained our MTCNN to perform 5 information extraction tasks: (1) primary cancer site (65 classes), (2) laterality (4 classes), (3) behavior (3 classes), (4) histological type (63 classes), and (5) histological grade (5 classes). We evaluated the performance on a corpus of 95 231 pathology documents (71 223 unique tumors) obtained from the Louisiana Tumor Registry. We compared the performance of the MTCNN models against single-task CNN models and 2 traditional machine learning approaches, namely support vector machine (SVM) and random forest classifier (RFC). Results MTCNNs offered superior performance across all 5 tasks in terms of classification accuracy as compared with the other machine learning models. Based on retrospective evaluation, the hard parameter sharing and cross-stitch MTCNN models correctly classified 59.04% and 57.93% of the pathology reports respectively across all 5 tasks. The baseline models achieved 53.68% (CNN), 46.37% (RFC), and 36.75% (SVM). Based on prospective evaluation, the percentages of correctly classified cases across the 5 tasks were 60.11% (hard parameter sharing), 58.13% (cross-stitch), 51.30% (single-task CNN), 42.07% (RFC), and 35.16% (SVM). Moreover, hard parameter sharing MTCNNs outperformed the other models in computational efficiency by using about the same number of trainable parameters as a single-task CNN. Conclusions The hard parameter sharing MTCNN offers superior classification accuracy for automated coding support of pathology documents across a wide range of cancers and multiple information extraction tasks while maintaining similar training and inference time as those of a single task–specific model.

Building a data-driven model to localize the origin of ventricular activation from 12-lead electrocardiograms (ECG) requires addressing the challenge of large anatomical and physiological variations across individuals. The alternative of a patient-specific model is, however, difficult to implement in clinical practice because the training data must be obtained through invasive procedures. In this paper, we present a novel approach that overcomes this problem of the scarcity of clinical data by transferring the knowledge from a large set of patient-specific simulation data while utilizing domain adaptation to address the discrepancy between the simulation and clinical data. The method that we have developed quantifies non-uniformly distributed simulation errors, which are then incorporated into the process of domain adaptation in the context of both classification and regression. This yields a quantitative model that, with the addition of 12-lead ECG data from each patient, provides progressively improved patient-specific localizations of the origin of ventricular activation. We evaluated the performance of the presented method in localizing 75 pacing sites on three in-vivo premature ventricular contraction (PVC) patients. We found that the presented model showed an improvement in localization accuracy relative to a model trained on clinical ECG data alone or a model trained on combined simulation and clinical data without considering domain shift. Furthermore, we demonstrated the ability of the presented model to improve the real-time prediction of the origin of ventricular activation with each added clinical ECG data, progressively guiding the clinician towards the target site.

Highlights • Complete surgical resection of liposarcoma is central in treatment based on grade. • Radiation reduces risk of local recurrence or minimize surgical morbidity in the highly radiosensitive myxoid/round cell liposarcoma group. • Chemotherapy is utilized in chemosensitive histologies with metastatic potential. • Understanding of the genetic aberrations allows development of targeted therapies for liposarcoma., Introduction Multifocal soft tissue sarcoma is a rare clinical entity occurring in 1 % of patients with extremity soft tissue sarcoma and in 4.5 % of patients with liposarcoma. Multifocal disease may arise either synchronously or metachronously and has been associated with poor prognosis. Case presentation Herein, we report a rare case of a Syrian woman patient with six different foci of liposarcoma in five sites at the time of diagnosis. Discussion Liposarcomas are currently classified into four different subtypes based on histologic or genetic analysis according to the World Health Organization (WHO), including cell-differentiated, dedifferentiated, myxoid, and pleomorphic. In the present work, we report a neglected patient with a multicentric round cell liposarcoma in five different sites. Conclusion Medicine cannot provide much care for these advanced cases of multicentric liposarcoma so far. Therefore, more research should be conducted to improve the ability to manage these entities and to identify potential novel therapies.

Objective To address the opioid overdose epidemic, it is important to understand the broad scope of efforts under way in states, particularly states in which the rate of opioid-involved overdose deaths is declining. The primary objective of this study was to examine core elements of overdose prevention activities in 4 states with a high rate of opioid-involved overdose deaths that experienced a decrease in opioid-involved overdose deaths from 2016 to 2017. Methods We identified 5 states experiencing decreases in age-adjusted mortality rates for opioid-involved overdoses from 2016 to 2017 and examined their overdose prevention programs via program narratives developed with collaborators from each state’s overdose prevention program. These program narratives used 10 predetermined categories to organize activities: legislative policies; strategic planning; data access, capacity, and dissemination; capacity building; public-facing resources (eg, web-based dashboards); training resources; enhancements and improvements to prescription drug monitoring programs; linkage to care; treatment; and community-focused initiatives. Using qualitative thematic analysis techniques, core elements and context-specific activities emerged. Results In the predetermined categories of programmatic activities, we identified the following core elements of overdose prevention and response: comprehensive state policies; strategic planning; local engagement; data access, capacity, and dissemination; training of professional audiences (eg, prescribers); treatment infrastructure; and harm reduction. Conclusions The identification of core elements and context-specific activities underscores the importance of implementation and adaptation of evidence-based prevention strategies, interdisciplinary partnerships, and collaborations to address opioid overdose. Further evaluation of these state programs and other overdose prevention efforts in states where mortality rates for opioid-involved overdoses declined should focus on impact, optimal timing, and combinations of program activities during the life span of an overdose prevention program.

More than 3.5 million people have died globally from COVID-19, yet an effective therapy is not available. It is, therefore, important to understand the signaling pathways that mediate disease progression in order to identify new molecular targets for therapeutic development. Here, we report that the blood serum levels of ephrin-A1 and the sheddase ADAM12 were significantly elevated in COVID-19 patients treated at SUNY Downstate Hospital of Brooklyn, New York. Both ephrin-A1 and ADAM12 are known to be involved in inflammation and regulate endothelial cell permeability, thus providing a gateway to lung injury. The clinical outcome correlated with the ephrin-A1 and ADAM12 serum levels during the first week of hospitalization. In contrast, the serum levels of TNFα were elevated in only a small subset of the patients, and these same patients also had highly elevated levels of the sheddase ADAM17. These data indicate that ephrin-A1-mediated inflammatory signaling may contribute to COVID-19 disease progression more so than TNFα-mediated inflammatory signaling. They also support the notion that, in COVID-19 inflammation, ADAM12 sheds ephrin-A1, while ADAM17 sheds TNFα. Furthermore, the results suggest that elevated serum levels and activity of cytokines, such as TNFα, and other secreted inflammatory molecules, such as ephrin-A1, are not simply due to overexpression, but also to upregulation of sheddases that release them into the blood circulation. Our results identify ephrin-A1, ADAM12, and other molecules in the ephrin-A1 signaling pathway as potential pharmacological targets for treating COVID-19 inflammation., COVID-19, Inflammation, Ephrin, Eph receptors, ADAM proteases.

Sialic acid that presents on the surface of lung epithelial cells is considered as one of the main binding targets for many respiratory viruses, including influenza and the current coronavirus (SARS-CoV-2) through the viral surface protein hemagglutinin. Gold nanoparticles (Au NPs) are extensively used in the diagnostic field owing to a phenomenon known as ‘surface plasmonic resonance’ in which the scattered light is absorbed by these NPs and can be detected via UV-Vis spectrophotometry. Consequently, sialic acid conjugated Au NPs (SA-Au NPs) were utilized for their plasmonic effect against SARS-CoV-2, influenza B virus, and Middle-East respiratory syndrome-related coronavirus (MERS) in patients’ swab samples. The SA-Au NPs system was prepared by a one-pot synthesis method, through which the NPs solution color changed from pale yellow to dark red wine color, indicting its successful preparation. In addition, the SA-Au NPs had an average particle size of 30 ± 1 nm, negative zeta potential (−30 ± 0.3 mV), and a UV absorbance of 525 nm. These NPs have proven their ability to change the color of the NPs solutions and patients’ swabs that contain SARS-CoV-2, influenza B, and MERS viruses, suggesting a rapid and straightforward detection tool that would reduce the spread of these viral infections and accelerate the therapeutic intervention.

Since the launch of the Sustainable Development Goals (SDGs) in 2015, Africa has made some progress towards its achievements, particularly in the area of gender equality- SDG 5, climate change -SDG 13 and preserving life on land- SDG 15. Although, some African countries have made tangible impact on the other goals such as health, food production and economic growth; these efforts experienced some setbacks with attention shifting to curbing the spread of the novel coronavirus. As much as mitigating the spread of the spread of COVID-19 is important, so is ensuring efforts made on other goals are not lost, as the SDGs are much interconnected, and failure in one impacts others. For the African continent to achieve sustainable development beyond COVID-19, strategic actions which will involve innovations, evaluations and strong political will towards implementations must be taken by relevant stakeholders, so the continent is not left behind in the global goals achievement by 2030.

Breast cancer is one of the most common life-threatening diseases that affects women globally. Saudi Arabia is also one of the countries that suffer from a serious number of this disease among women. In terms of diagnosis modalities, a mammogram is the first line for detecting breast cancer. In addition, breast cancer can be screened by real-time ultrasound images, which are of relatively less quality and have more impact (noninvasive) images. Therefore, the purpose of this study is to develop image enhancement techniques using deep learning and image processing techniques. The main goal is to improve the ultrasound images in order to help radiologists screen the disease more accurately. For this study, ninety female patients of ages between 15 – 77 years are considered. These patients were already diagnosed using ultrasound with breast lesions. The images are visually graded and evaluated by two trained radiologists, both pre- and post-enhancement. In particular, two parameters were considered; 1) BI-RAD categories, 2) Breast cancer classification. The agreement between radiologists and post-enhancement was assessed using simple kappa and weighted kappa statistics. Moreover, sensitivity and specificity are also calculated.

Choledocholithiasis is a common finding in clinical practice, with presentation varying from asymptomatic to life-threatening complications. In symptomatic patients, there is no doubt that treatment to clear the bile duct is indicated, but there is still a debate regarding the treatment of patients with silent common bile duct stones (CBDS). The question addressed by this best evidence topic is whether patients with asymptomatic CBDS should be managed in the same way as patients with symptoms or complications. The search strategy yielded 609 articles, from which 8 articles found to be relevant to this topic. We also summarised the most notable societal guidelines recommendations, regarding this topic. We tabulated the article title, author, year, country, study type, outcomes, results, and comments. We concluded that patients with asymptomatic CBD stones should be offered endoscopic treatment If they are fit, after discussion of the potential risks and benefits of both options of conservative and interventional treatment with the patients.

Graphical abstract, Highlights • Topiramate, an antiepileptic drug is also effective for treating alcohol abuse. • There is a high possibility of its use in alcohol-induced gastric injury. • Effect of Topiramate studied in alcohol induced gastric injury. • Topiramate had a significant gastro-protective effect against gastric injury. • Mediated through anti-secretory, anti-inflammatory, and mucosal strengthening activity., Background Topiramate (TPM), an antiepileptic drug, is also effective against alcohol dependency, a crucial factor in forming gastric ulcers. There is an increased possibility of patients with compromised gastric conditions getting exposed to TPM, but its effect on gastric ulcers is unknown. This study investigates the implication of acute TPM in ethanol-produced gastric ulceration in rats. Material and methods The effect of TPM studied in male 200−225 g Sprague Dawley rats against ethanol-induced gastric ulcers and for gastric secretion and acidity. The factors assessed include gastric secretion and acidity, gastric ulcer score, biochemical and histological changes, NF-kB, and p53 expression. The analysis of data performed by using the Kruskal Wallis test and Dunnett's multiple comparison tests. Results TPM pretreatment showed gastroprotective effects. It significantly reduced ethanol-induced increased gastric secretion, acidity, and gastric ulcer index and prevented gastric mucus depletion. The ethanol-induced inflammation and apoptosis were also significantly decreased by reducing the increased gastric myeloperoxidase activity and the expression of NF-kB and p53. TPM pretreatment also reduced the ethanol-induced damage to the gastric histology in rats. Conclusion TPM exerted a gastro-protective effect against ethanol-induced gastric ulcers mediated by reducing the gastric ulcer index, preventing a decrease of the mucus levels, reduction in inflammation, damage to gastric histology, and a decrease in the enhanced expression of NF-kB and TPM. Further detailed investigations are essential to understand the chronic influence of TPM on gastric ulcers.

Visceral Leishmaniasis (VL) due to Leishmania donovani is a neglected protozoan parasitic disease in humans, which is usually fatal if untreated. Phlebotomus orientalis, the predominant VL vector in East Africa, is a highly exophilic/exophagic species that poses a major challenge to current Integrated Vector Management (IVM). Here we report results of pilot studies conducted in rural villages in Gedarif state, Sudan, to evaluate outdoor residual spraying of 20mg active ingredient (a.i.) /m2 deltamethrin insecticide applied to the characteristic household compound boundary reed fence and to the outside of household buildings (Outdoor Residual Insecticide Spraying, ODRS), and as an alternative, spraying restricted to the boundary fence only (Restricted Outdoor Residual Insecticide Spraying, RODRS). Four to six clusters of 20 households were assigned to insecticide treatments or control in three experiments. Changes in sand fly numbers were monitored over 2,033 trap-nights over 43–76 days follow-up in four sentinel houses per cluster relative to unsprayed control clusters. Sand fly numbers were monitored by sticky traps placed on the ground on the inside (“outdoor”) and the outside (“peridomestic”) of the boundary fence, and by CDC light traps suspended outdoors in the household compound. The effects of ODRS on sand fly numbers inside sleeping huts were monitored by insecticide knockdown. After a single application, ODRS reduced P. orientalis abundance by 83%-99% in outdoor and peridomestic trap locations. ODRS also reduced numbers of P. orientalis found resting inside sleeping huts. RODRS reduced outdoor and peridomestic P. orientalis by 60%-88%. By direct comparison, RODRS was 58%-100% as effective as ODRS depending on the trapping method. These impacts were immediate on intervention and persisted during follow-up, representing a large fraction of the P. orientalis activity season. Relative costs of ODRS and RODRS delivery were $5.76 and $3.48 per household, respectively. The study demonstrates the feasibility and high entomological efficacy of ODRS and RODRS, and the expected low costs relative to current IVM practises. These methods represent novel sand fly vector control tools against predominantly exophilic/exophagic sand fly vectors, aimed to lower VL burdens in Sudan, with potential application in other endemic regions in East Africa., Author summary Phlebotomus orientalis is the predominant vector of visceral leishmaniasis (VL, kala azar) in Sudan and other countries of East Africa, where the disease causes high morbidity and mortality. This sylvatic sand fly species is abundant in wild habitats characterized by presence of black cotton soil and vegetation dominated by Balanites aegyptiaca and/or Acacia seyal trees. In villages, the vector bites people in the household yard and in nearby peri-domestic locations, exhibiting limited indoor resting behaviour. The marked exophagic and exophilic behaviours of P. orientalis represent a profound challenge for VL control by excluding indoor residual spraying of insecticides (IRS) and compromising the efficacy of insecticide-impregnated bednets (ITNs). In this study, we evaluated the entomological efficacy of residual pyrethroid applied outdoors to household boundary fences and the exterior walls of household huts (outdoor residual insecticide spraying, ODRS), to reduce the abundance of P. orientalis inside and outside houses. We also evaluated the entomological impact of a restricted outdoor residual insecticide spraying (RODRS), whereby insecticide was applied only to the boundary fence. The study was carried out in June 2016-June 2017 in Jebel-Algana and Umsalala villages, Gedarif state, eastern Sudan, which are highly endemic for VL. The results showed that a single ODRS application of 20mg a.i. /m2 2.8% deltamethrin provided average reductions of 83%-99% in outdoor and peridomestic P. orientalis sand fly numbers relative to unsprayed control clusters. RODRS reduced outdoor and peridomestic P. orientalis by 60%-88%. The average cost of ODRS and RODRS per household were $5.76 and $3.48, respectively. The costs of these community-based control measures were substantially lower than the costs of LLINs, which is the only evidence-based tool used to protect against VL in the area. Future studies should evaluate the impact of ODRS/RODRS transmission of VL incidence in endemic villages and in seasonal agricultural farms.

OBJECTIVES: Chronic hepatitis B virus infection is a major cause of morbidity and mortality. The aim of the study is to describe the hepatic histology in children chronically infected with hepatitis B virus living in the United States and Canada. METHODS: Liver biopsies of 134 treatment-naïve children with chronic hepatitis B virus infection were scored for inflammation, fibrosis, and other histological features, and correlated with clinical and laboratory data. RESULTS: Sixty percentage of subjects acquired the infection vertically, 51% were male, and 69% were hepatitis B e antigen-positive at the time of the biopsy. Hepatitis B DNA levels were generally high (mean 7.70 log IU/mL), as was serum alanine aminotransferase (median 120 U/L). Using the Ishak-modified histology activity index scoring system, interface hepatitis was mild in 31%, moderate in 61%, and severe in 6%. Lobular inflammation was mild in 54%, moderate in 29%, and marked in 7%. Portal inflammation was mild in 38% and moderate in 62% of subjects. Eighteen percentage had no fibrosis, 59% had portal expansion without bridging fibrosis, 19% had bridging fibrosis, and 4% had cirrhosis. Alanine aminotransferase positively correlated with inflammation and fibrosis. Neither age, duration of infection, nor Hepatitis B virus DNA levels correlated with fibrosis. Fibrosis-4 index did not correlate with fibrosis but correlated with inflammation. Aspartate aminotransferase/platelet ratio index correlated with both inflammation and fibrosis. CONCLUSIONS: Chronic hepatitis B virus infection results in significant inflammation and fibrosis during childhood. Serum alanine aminotransferase is a strong indicator of the severity and extent of hepatic inflammation and fibrosis.

Brain hemorrhage is a type of stroke which is caused by a ruptured artery, resulting in localized bleeding in or around the brain tissues. Among a variety of imaging tests, a computerized tomography (CT) scan of the brain enables the accurate detection and diagnosis of a brain hemorrhage. In this work, we developed a practical approach to detect the existence and type of brain hemorrhage in a CT scan image of the brain, called Accurate Identification of Brain Hemorrhage, abbreviated as AIBH. The steps of the proposed method consist of image preprocessing, image segmentation, feature extraction, feature selection, and design of an advanced classification framework. The image preprocessing and segmentation steps involve removing the skull region from the image and finding out the region of interest (ROI) using Otsu&rsquo, s method, respectively. Subsequently, feature extraction includes the collection of a comprehensive set of features from the ROI, such as the size of the ROI, centroid of the ROI, perimeter of the ROI, the distance between the ROI and the skull, and more. Furthermore, a genetic algorithm (GA)-based feature selection algorithm is utilized to select relevant features for improved performance. These features are then used to train the stacking-based machine learning framework to predict different types of a brain hemorrhage. Finally, the evaluation results indicate that the proposed predictor achieves a 10-fold cross-validation (CV) accuracy (ACC), precision (PR), Recall, F1-score, and Matthews correlation coefficient (MCC) of 99.5%, 99%, 98.9%, 0.989, and 0.986, respectively, on the benchmark CT scan dataset. While comparing AIBH with the existing state-of-the-art classification method of the brain hemorrhage type, AIBH provides an improvement of 7.03%, 7.27%, and 7.38% based on PR, Recall, and F1-score, respectively. Therefore, the proposed approach considerably outperforms the existing brain hemorrhage classification approach and can be useful for the effective prediction of brain hemorrhage types from CT scan images (The code and data can be found here: http://cs.uno.edu/~tamjid/Software/AIBH/code_data.zip).

BackgroundPublished training recommendations exist for youth athletes aimed at reducing injury risk. No studies have assessed the impact of counselling interventions using training recommendations on risk of injury in young athletes.ObjectivesTo determine if online training counselling regarding safe sport participation can reduce injury risk in youth athletes and to assess recommendation compliance, including barriers to compliance.MethodsA multicentre randomised intervention trial was performed at two Midwestern academic institutions with expertise in treating young athletes. Enrolled subjects ages 8–17 completed a baseline risk assessment survey and were randomised to a control or intervention group. Both groups completed exposure surveys every 3 months for 1 year. The intervention group also received online training counselling on safe sport participation every 3 months. Training characteristics including training volume, degree of specialisation and adherence to recommendations were captured. Differences in self-reported injury between groups, compliance to recommendations and barriers to compliance were evaluated.ResultsAt baseline, n=357 subjects were enrolled (n=172 control and n=185 intervention). Controls were nearly twice as likely to be injured during the intervention period after controlling for age, sex, baseline injury and level of specialisation. No improvement in recommendation compliance was detected among intervention subjects. Primary barriers to compliance were no prior knowledge of recommendations, personal choice and following coaches’ recommendations.ConclusionsIn this convenience sample of youth athletes, electronic training counselling surrounding safe sports participation was not determined to affect injury risk. Lack of knowledge and adherence to appropriate training recommendations is evident and barriers to compliance exist.

Background Human decidua basalis mesenchymal stem/multipotent stromal cells (DBMSCs) inhibit endothelial cell activation by inflammation induced by monocytes. This property makes them a promising candidate for cell-based therapy to treat inflammatory diseases, such as atherosclerosis. This study was performed to examine the ability of DBMSCs to protect endothelial cell functions from the damaging effects resulting from exposure to oxidatively stress environment induced by H2O2 and monocytes. Methods DBMSCs were co-cultured with endothelial cells isolated from human umbilical cord veins in the presence of H2O2 and monocytes, and various functions of endothelial cell were then determined. The effect of DBMSCs on monocyte adhesion to endothelial cells in the presence of H2O2 was also examined. In addition, the effect of DBMSCs on HUVEC gene expression under the influence of H2O2 was also determined. Results DBMSCs reversed the effect of H2O2 on endothelial cell functions. In addition, DBMSCs reduced monocyte adhesion to endothelial cells and also reduced the stimulatory effect of monocytes on endothelial cell proliferation in the presence of H2O2. Moreover, DBMSCs modified the expression of many genes mediating important endothelial cell functions. Finally, DBMSCs increased the activities of glutathione and thioredoxin reductases in H2O2-treated endothelial cells. Conclusions We conclude that DBMSCs have potential for therapeutic application in inflammatory diseases, such as atherosclerosis by protecting endothelial cells from oxidative stress damage. However, more studies are needed to elucidate this further.

Objective: The aim of the study is to evaluate serum level of Anti-Mullerian Hormone in patients with primary breast cancer proposed to have chemotherapy and to study the effect age on Anti-Mullerian Hormone and menstrual cycle pattern after chemotherapy. Background: In reproductive-aged breast cancer survivors, there is a need for “real-time” biomarkers of post-chemotherapy ovarian function. Method: This prospective cohort study had been conducted upon one hundred twenty women, sixty women as a patient group (n = 60), who are newly diagnosed with primary breast cancer and Sixty women as a control group (n = 60). Their ages between (21 and 41) years. Patients were subgrouped according to age as ≤35 and >35 years. Results: There were significant differences between patients and control groups regarding BMI,regularity of menstrual cycle (P

BACKGROUND: Exposure to stressful working conditions without adequate stress-coping strategies may lead to stress and even psychiatric morbidity. There are a dearth of studies on stress-coping strategies among consultant physicians in the Arabian Gulf. OBJECTIVES: Determine stress-coping strategies among consultants in Saudi Arabia and the relationship of strategies to level of stress. DESIGN: Analytical cross-sectional study. SETTINGS: Conducted between November 2014 and March 2015 among physician consultants registered at the Saudi Commission for Health Specialties. SUBJECTS AND METHODS: Text messages were used to directly ask consultants to complete an online questionnaire. MAIN OUTCOME MEASURES: The 28-item Brief COPE inventory and the Perceived Stress Scale. SAMPLE SIZE: 582. RESULTS: The consultants were largely males (71%) and Saudi (56%), and their mean age was 46.9 (7.9) years. Adaptive stress-coping strategies were more frequently used than maladaptive stress-coping strategies (68% versus 49%). Stress levels were positively correlated with maladaptive stress-coping strategies (r=0.41, P

Antibiotic resistance (AR) is a growing public health concern worldwide, and it is a top health challenge in the 21st century. AR among Enterobacteriaceae is rapidly increasing, especially in third-generation cephalosporins and carbapenems. Further, strains carrying mobilized colistin resistance ( mcr) genes 1 and 2 have been isolated from humans, food-producing animals, and the environment. The uncontrolled use of antibiotics in food-producing animals is a major factor in the generation and spread of AR. No studies have been done to evaluate AR in the veterinary sector of Qatar. This study aimed at establishing primary baseline data for the prevalence of AR among food-producing animals in Qatar. Fecal samples (172) were obtained from two broiler farms and one live bird market in Qatar, and 90 commensal Escherichia coli bacteria were isolated and subjected to susceptibility testing against 16 clinically relevant antibiotics by using the E-test method. The results found that 81 (90%) of 90 isolates were resistant to at least one antibiotic, 14 (15.5%) of 90 isolates were colistin resistant, 2 (2.2%) of 90 isolates were extended-spectrum β-lactamase producers, and 2 (2.2%) of 90 isolates were multidrug resistant to four antibiotic classes. Extended-spectrum β-lactamase-producing E. coli and colistin-resistant isolates were confirmed by using double-disc susceptibility testing and PCR, respectively. Such a high prevalence of antibiotic-resistant E. coli could be the result of a long application of antibiotic treatment, and it is an indicator of the antibiotic load in food-producing animals in Qatar. Pathogens carrying AR can be easily transmitted to humans through consumption of undercooked food or noncompliance with hygiene practices, mandating prompt development and implementation of a stewardship program to control and monitor the use of antibiotics in the community and agriculture.

This article provides a narrative review of journal articles on the use of fingerspelling to support the vocabulary and literacy development of deaf students. The role of fingerspelling in decoding written English and supporting vocabulary development has become an increasingly relevant topic in the bilingual education of deaf students. Search limiters included written English, full-text articles that were published in peer-reviewed journals after 2005. A total of eleven articles were reviewed. The findings from the review indicated that the use of fingerspelling could support vocabulary and literacy development among bilingual deaf students. The current review discusses the findings and offers recommendations for future research.

Background Rhazya stricta Decne. is a medicinal plant that is widespread in Saudi Arabia and desert areas of the Arabian Peninsula. Its extract contains alkaloids, tannins, and flavonoids that are involved in different biological activities. The study aim was to evaluate the effects of Rhazya stricta plant extracts on the proliferation and differentiation of NTERA-2 (NT2) pluripotent embryonal carcinoma cells. Methods Soxhlet extraction was carried out using different solvents to extract stems, leaves and fruit parts of this plant. Cytotoxicity was evaluated by an MTS cell viability assay. The ability of the plant extract to induce cell differentiation was examined phenotypically using an inverted light microscope. The expression of pluripotency markers was investigated by reverse transcriptase polymerase chain reaction (RT-PCR) and immunocytochemistry. Phytochemical screening of chloroform stem extracts was carried out and a chromatographic fingerprint was generated using gas chromatography – mass spectrometry (GC-MS). Results Chloroform stem extract induced differentiation of NT2 cells at 5 μg/ml, and the differentiated cells exhibited neurite formation. Following induction of differentiation, there was significant down-regulation of the pluripotency marker genes Oct4 and Sox2. In addition, the surface antigen pluripotency marker, TRA-1-60, was strongly down-regulated. Phytochemical analysis of the extract showed the presence of alkaloids and saponins. The chromatogram revealed the presence of fifteen compounds with different retention times. Conclusion Our results demonstrate for the first time that chloroform stem extract of R. stricta can induce neuronal differentiation of stem cells at an early stage and may contain potential therapeutic agent that can be used in neurodegenerative diseases.

Persistent left superior vena cava (PLSVC) is a common cardiac anomaly associated with congenital heart diseases. A diagnosis of PLSVC usually warrants a detailed fetal echocardiography. Lesser known associations are the extra cardiac anomalies notably the upper airway and the gastrointestinal tract anomalies. We highlight here the importance of detailed fetal assessment for extra cardiac anomalies in addition to fetal echocardiography in fetuses diagnosed with PLSVC. We hereby present a preterm infant who presented with a triad of PLSVC, laryngeal atresia, and esophageal atresia.

This study considers the importance of window styles known as Rowshan that is found in heritage buildings in the city of Jeddah. The primary objective of this study was to evaluate and compare the use of Rowshan with glass windows for accessing direct sunlight in interior spaces in order to avoid vitamin D deficiencies. It is clear that the use of Rowshan as a design strategy is a relatively simple way for improving vitamin D intake among occupants whilst also ensuring that a residence adheres to the privacy levels that are crucial in Arab and Middle-Eastern countries subjected to some considerations.

Collaboration among cancer registries is essential to develop accurate, robust, and generalizable deep learning models for automated information extraction from cancer pathology reports. Sharing data presents a serious privacy issue, especially in biomedical research and healthcare delivery domains. Distributing pretrained deep learning (DL) models has been proposed to avoid critical data sharing. However, there is growing recognition that collaboration among clinical institutes through DL model distribution exposes new security and privacy vulnerabilities. These vulnerabilities increase in natural language processing (NLP) applications, in which the dataset vocabulary with word vector representations needs to be associated with the other model parameters. In this paper, we propose a novel privacy-preserving DL model distribution across cancer registries for information extraction from cancer pathology reports with privacy and confidentiality considerations. The proposed approach exploits the adversarial training framework to distinguish private features from shared features among different datasets. It only shares registry-invariant model parameters, without sharing raw data nor registry-specific model parameters among cancer registries. Thus, it protects both the data and the trained model simultaneously. We compare our proposed approach to single-registry models, and a model trained on centrally hosted data from different cancer registries. The results show that the proposed approach significantly outperforms the single-registry models and achieves statistically indistinguishable micro and macro F1-score as compared to the centralized model.

Objectives: To explore the correlation between maternal and cord blood prolactin, the correlation between cord prolactin and birth weight, and to compare cord blood prolactin in new-borns of women with normal pregnancy and women with pregnancy complications namely; gestational hypertension, gestational diabetes and preterm labour.Methods: This study was performed from September to December 2018. Thirty-two women, delivered at Baghdad teaching hospital, and their newborns (32) were included. Maternal blood (5 ml) was taken before labour and cord blood (5 ml) was collected after placenta expulsion. Maternal and cord blood prolactin were analysed using fluorescence immunoassay. Results: Cord blood prolactin was higher in babies born to hypertensive women (405.28±77.52 ng/ml) than normal pregnancy women (244.80±60.80 ng/ml), P=0.000. Also, cord prolactin in gestational hypertension group was significantly higher than diabetic (P=0.006) and preterm labour (P=0.000) groups. No significant difference was noticed in cord blood prolactin in newborns of diabetic and normal pregnancy women (299.28±37.01, 244.80±60.80 ng/ml respectively, P=0.053). Preterm babies had lower cord prolactin (204.57±22.90 ng/ml) than normal pregnancy babies (244.80±60.80 ng/ml), however the difference was non-significant, P=0.118. Positive correlation was found between cord and maternal prolactin (P=0.000) and between cord prolactin and birth weight (P=0.018). Conclusion: Cord blood prolactin is high in newborns of hypertensive women, low in preterm neonates. Diabetes has no effect on cord prolactin level. doi: https://doi.org/10.12669/pjms.35.4.558 How to cite this:Alawad ZM, Al-Omary HL. Maternal and cord blood prolactin level and pregnancy complications. Pak J Med Sci. 2019;35(4):1122-1127. doi: https://doi.org/10.12669/pjms.35.4.558 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

We introduce a deep learning architecture, hierarchical self-attention networks (HiSANs), designed for classifying pathology reports and show how its unique architecture leads to a new state-of-the-art in accuracy, faster training, and clear interpretability. We evaluate performance on a corpus of 374,899 pathology reports obtained from the National Cancer Institute's (NCI) Surveillance, Epidemiology, and End Results (SEER) program. Each pathology report is associated with five clinical classification tasks – site, laterality, behavior, histology, and grade. We compare the performance of the HiSAN against other machine learning and deep learning approaches commonly used on medical text data – Naive Bayes, logistic regression, convolutional neural networks, and hierarchical attention networks (the previous state-of-the-art). We show that HiSANs are superior to other machine learning and deep learning text classifiers in both accuracy and macro F-score across all five classification tasks. Compared to the previous state-of-the-art, hierarchical attention networks, HiSANs not only are an order of magnitude faster to train, but also achieve about 1% better relative accuracy and 5% better relative macro F-score.

Finding optimal hyperparameters is necessary to identify the best performing deep learning models but the process is costly. In this paper, we applied model-based optimization, also known as Bayesian optimization, using the CANDLE framework implemented on a High-Performance Computing environment. As a use case we selected information extraction from cancer pathology reports using a multi-task convolutional neural network, and hierarchical convolutional attention network to be optimized. We utilized a synthesized text corpus of 8,000 training cases and 2,000 validation cases with four types of clinical task labels including primary cancer site, laterality, behavior, and histological grade. We demonstrated that hyperparameter optimization using the CANDLE framework is a feasible approach with respect to both scalability and clinical task performance.

Background & Aims Clearance of hepatitis B surface antigen (HBsAg) from serum is the most desirable end point and a proposed definition of functional cure for hepatitis B virus (HBV) infection. However, little is known about the long-term durability of HBsAg loss, and there is controversy over whether the development of antibodies against HBsAg (anti-HBs) is required for maintenance. We aimed to assess the durability of spontaneous or treatment-related (interferon or nucleos(t)ide analogue [NA]) loss of HBsAg. Methods We performed a retrospective study of patients with chronic HBV infection followed up at the National Institutes of Health from February 1980 through November 2017. We identified those with HBsAg loss, confirmed on 2 visits at least 24 weeks apart. Patients with hepatitis C virus, hepatitis D virus, human immunodeficiency virus, or human T lymphocyte virus co-infection or HBsAg loss after liver transplantation were excluded. Patients were assigned to the following groups: spontaneous clearance (cleared HBsAg without ever receiving treatment or those who received treatment with a NA or interferon and discontinued therapy >5 years before HBsAg loss), interferon-treated (cleared HBsAg either during treatment or ≤5 years after stopping interferon), and NA-treated (cleared HBsAg either during treatment or ≤5 years after stopping NA). Results Among the 787 HBsAg-positive patients, 89 achieved HBsAg loss; 65 of 89 had confirmed HBsAg loss, which was spontaneous in 19 of the patients (29%), after interferon in 22 (34%), and after NA in 24 (37%). Of the 65 patients with confirmed loss of HBsAg, 62 patients (95%) remained HBsAg negative after a mean time of 9.6 years from the first negative HBsAg test result. HBsAg seroreversion occurred in 3 of the 46 treated patients (7%) (1 interferon and 2 NA), 1 of whom was positive for anti-HBs. At the time of HBsAg loss, 33 of 65 (51%) were anti-HBs positive. At the last follow-up evaluation, anti-HBs was detectable in 50 of the 62 patients (81%) assessed. The rate of development of anti-HBs was proportionally higher among interferon-treated patients (19 of 21; 90%) than NA-treated patients (17 of 22; 77%) or patients with spontaneous loss of HBsAg (14 of 19; 74%). Conclusions In a retrospective study of 787 HBsAg-positive patients, loss of HBsAg (either spontaneous or after treatment) was confirmed in 8% and was durable. Seroconversion to anti-HBs increased over time and appeared to be more frequent after interferon treatment. HBsAg loss is therefore a robust end point for functional cure.

Spontaneous subcapsular hematoma of the liver is an extremely rare but potentially life-threatening condition. We report a case of subcapsular hematoma of the liver without any apparent lesion and in the absence of coagulopathy or trauma. A CT scan of the abdomen demonstrated a huge subcapsular hematoma around the liver. The patient was treated conservatively and was discharged home after one week. Up to our knowledge, this is one of the very few reported cases of a spontaneous subcapsular hepatic hematoma.

Kruppel-like factor 4 (KLF4) is a pluripotency transcription factor that helps in generating induced pluripotent stem cells (iPSCs). We sequenced for the first time the full coding sequence of Camelus dromedarius KLF4 (cKLF4), which is also known as the Arabian camel. Bioinformatics analysis revealed the molecular weight and the isoelectric point of cKLF4 protein to be 53.043 kDa and 8.74, respectively. The predicted cKLF4 protein sequence shows high identity with some other species as follows: 98% with Bactrian camel and 89% with alpaca KLF4 proteins. A three-dimensional (3D) structure was built based on the available crystal structure of the Mus musculus KLF4 (mKLF4) of 82 residues (PDB: 2 WBS) and by predicting 400 residues using bioinformatics software. The comparison confirms the presence of the zinc finger domains in cKLF4 protein. Phylogenetic analysis showed that KLF4 from the Arabian camel is grouped with the Bactrian camel, alpaca, cattle, and pig. This study will help in the annotation of KLF4 protein and in generating camel-induced pluripotent stem cells (CiPSCs).

To explore potential gender differences in the factors associated with asthma-specific quality of life (AQL).A cross-sectional study of consecutive series of adult patients attending primary care centers at three major hospitals in Riyadh, Saudi Arabia, was performed. AQL was measured using a standardized version of the AQL questionnaire (min = 1, max = 7), with higher scores indicating a better AQL. Multiple linear regression analysis was performed.The mean AQL was 4.3 (standard deviation [SD] = 1.5) for males and 4.0 (SD = 1.3) for females (p = 0.113). With each unit increase in asthma control, the AQL improved by 0.19 points (95% confidence interval [CI] = 0.14-0.23) in men and by 0.21 points (95% CI = 0.16-0.25) in women. Daily tobacco smoking was associated with a 0.72 point (95% CI = 0.14-1.30) decrease in the AQL among males. Women who had a household member who smoked inside the house had a significantly lower AQL (B = -0.59, 95% CI = -1.0 - -0.19). A monthly household income of 25,000 Saudi Riyals or more was associated with a better AQL among men (B = 0.51, 95% CI = 0.01-1.01), whereas being employed exhibited a protective effect in women (B = 0.48, 95% CI = 0.11-0.84). Higher levels of perceived asthma severity were associated with better AQL in women (B = 0.82, 95% CI = 0.36-1.28).Our findings revealed gender-specific differences in the correlates of AQL in Saudi Arabia, particularly in tobacco exposure, socio-economic factors and perceived asthma severity.

Although the sequencing information of Sox2 cDNA for many mammalian is available, the Sox2 cDNA of Camelus dromedaries has not yet been characterized. The objective of this study was to sequence and characterize Sox2 cDNA from the brain of C. dromedarius (also known as Arabian camel). A full coding sequence of the Sox2 gene from the brain of C. dromedarius was amplified by reverse transcription PCRjmc and then sequenced using the 3730XL series platform Sequencer (Applied Biosystem) for the first time. The cDNA sequence displayed an open reading frame of 822 nucleotides, encoding a protein of 273 amino acids. The molecular weight and the isoelectric point of the translated protein were calculated as 29.825 kDa and 10.11, respectively, using bioinformatics analysis. The predicted cSox2 protein sequence exhibited high identity: 99% for Homo sapiens, Mus musculus, Bos taurus , and Vicugna pacos ; 98% for Sus scrofa and 93% for Camelus ferus. A 3D structure was built based on the available crystal structure of the HMG-box domain of human stem cell transcription factor Sox2 (PDB: 2 LE4) with 81 residues and predicting bioinformatics software for 273 amino acid residues. The comparison confirms the presence of the HMG-box domain in the cSox2 protein. The orthologous phylogenetic analysis showed that the Sox2 isoform from C. dromedarius wa s grouped with humans, alpacas, cattle, and pigs. We believe that this genetic and structural information will be a helpful source for the annotation. Furthermore, Sox2 is one of the transcription factors that contributes to the generation-induced pluripotent stem cells (iPSCs), which in turn will probably help generate camel induced pluripotent stem cells (CiPSCs).

Objective: In machine learning, it is evident that the classification of the task performance increases if bootstrap aggregation (bagging) is applied. However, the bagging of deep neural networks takes tremendous amounts of computational resources and training time. The research question that we aimed to answer in this research is whether we could achieve higher task performance scores and accelerate the training by dividing a problem into sub-problems. Materials and Methods : The data used in this study consist of free text from electronic cancer pathology reports. We applied bagging and partitioned data training using Multi-Task Convolutional Neural Network (MT-CNN) and Multi-Task Hierarchical Convolutional Attention Network (MT-HCAN) classifiers. We split a big problem into 20 sub-problems, resampled the training cases 2,000 times, and trained the deep learning model for each bootstrap sample and each sub-problem—thus, generating up to 40,000 models. We performed the training of many models concurrently in a high-performance computing environment at Oak Ridge National Laboratory (ORNL). Results: We demonstrated that aggregation of the models improves task performance compared with the single-model approach, which is consistent with other research studies; and we demonstrated that the two proposed partitioned bagging methods achieved higher classification accuracy scores on four tasks. Notably, the improvements were significant for the extraction of cancer histology data, which had more than 500 class labels in the task; these results show that data partition may alleviate the complexity of the task. On the contrary, the methods did not achieve superior scores for the tasks of site and subsite classification. Intrinsically, since data partitioning was based on the primary cancer site, the accuracy depended on the determination of the partitions, which needs further investigation and improvement. Conclusion: Results in this research demonstrate that 1. The data partitioning and bagging strategy achieved higher performance scores. 2. We achieved faster training leveraged by the high-performance Summit supercomputer at ORNL.

Individual electronic health records (EHRs) and clinical reports are often part of a larger sequence-for example, a single patient may generate multiple reports over the trajectory of a disease. In applications such as cancer pathology reports, it is necessary not only to extract information from individual reports, but also to capture aggregate information regarding the entire cancer case based off case-level context from all reports in the sequence. In this paper, we introduce a simple modular add-on for capturing case-level context that is designed to be compatible with most existing deep learning architectures for text classification on individual reports. We test our approach on a corpus of 431,433 cancer pathology reports, and we show that incorporating case-level context significantly boosts classification accuracy across six classification tasks-site, subsite, laterality, histology, behavior, and grade. We expect that with minimal modifications, our add-on can be applied towards a wide range of other clinical text-based tasks.

We report the case of a 15-year-old girl presenting with distended abdomen, left flank pain, and a history of weight loss. Computed tomography showed a large tumor involving the left kidney that was initially diagnosed as renal cell carcinoma. She underwent exploratory open laparotomy and left radical nephrectomy followed by chemotherapy and showed good response. Histology of the resected tumor revealed features of Ewing's sarcoma of the kidney which was confirmed by molecular studies. This disease is rare, particularly in the pediatric population, and this report will help better understand the potential disease course and response to treatment. Keywords: Chemotherapy, Ewing's sarcoma of the kidney, Pediatric, Radical nephrectomy