Journals
2020 EN
Cook Geoffrey MW · Sousa Catia · Schaeffer Julia
+14 more
Contact repulsion of growing axons is an essential mechanism for spinal nerve patterning. In birds and mammals the embryonic somites generate a linear series of impenetrable barriers, forcing axon growth cones to traverse one half of each somite as they extend towards their body targets. This study shows that protein disulphide isomerase provides a key component of these barriers, mediating contact repulsion at the cell surface in chick half-somites. Repulsion is reduced both in vivo and in vitro by a range of methods that inhibit enzyme activity. The activity is critical in initiating a nitric oxide/S-nitrosylation-dependent signal transduction pathway that regulates the growth cone cytoskeleton. Rat forebrain grey matter extracts contain a similar activity, and the enzyme is expressed at the surface of cultured human astrocytic cells and rat cortical astrocytes. We suggest this system is co-opted in the brain to counteract and regulate aberrant nerve terminal growth.
eLife Sciences Publications
Journals
2020 EN
Heba H Al Ameer · Sajjad M AlKadhem · Fadi Busaleh
+2 more
The World Health Organization is still revising the epidemiology of multi-system inflammatory syndrome in children (MIS-C) and the preliminary case definition, although there is a dearth of robust evidence regarding the clinical presentations, severity, and outcomes. Researchers, epidemiologists, and clinicians are struggling to characterize and describe the disease phenomenon while taking care of the diseased persons at the forefronts. This report tackles the first case of a 13-year-old Saudi female with the MIS-C mimicking Kawasaki disease. Her main manifestations were fever, gastrointestinal symptoms, evidence of organ failure with an increase in inflammatory markers, and a history of coronavirus disease (COVID-19) infection. She had glucose-6-phosphate dehydrogenase (G6PD) deficiency and no significant previous history of any disease. She presented with signs of acute illness: high-grade fever (39.6°C) for five days accompanied by sore throat, malaise, reduced oral intake, abdominal pain, diarrhea, skin rash, bilateral non-suppurative conjunctivitis, and erythematous, cracked lips. Eventually, she died despite aggressive management based on the Centers for Disease Control and Prevention and the Saudi Ministry of Health guidelines for COVID-19 management. Based on this case, we suggest that pediatricians need to be aware of such atypical presentations and early referral to tertiary care is imperative for further early diagnosis and management. MIS-C is a rare yet severe and highly critical complication of COVID-19 infection in pediatrics, leading to serious and life-threatening illnesses. Knowledge about the wide spectrum of presenting signs and symptoms and disease severity, including early detection and treatment, is pivotal to prevent a tragic outcome.
Journals
2020 EN
Zainab Almoosa · Heba H Al Ameer · Sajjad M AlKadhem
+3 more
Fortunately, coronavirus disease 2019 (COVID-19) infection in pediatric populations exhibits a mild course of disease. However, a small number have recently been identified who develop a significant systemic inflammatory response, a new disease entity called multisystem inflammatory syndrome in children (MIS-C), especially after the peak of the wave in Al-Ahsa, Saudi Arabia, in early June to mid-July. In MIS-C children usually present a few days to a few weeks after recovery from COVID-19 with high grade fever, GI symptoms, Kawasaki-like picture or even toxic shock-like syndrome. Raising awareness about this disease entity is very fundamental to enable pediatricians and other health care providers to identify and manage these patients before it is too late. We describe 10 different cases of MIS-C with different risk factors and presentations.
Journals
2020 EN
Feroze Kaliyadan · Mohammed A. Al Ameer · Ali Al Ameer
+1 more
Objectives: The COVID-19 pandemic has led to an increased use of telemedicine. The primary objective of the study was to evaluate attitudes and behaviors of licensed physicians in the region to telemedicine. Methodology: A cross-sectional design using an electronic survey as the primary tool was done. The questionnaire had a demographic component of the respondent (first part), covering age, specialty, and experience with telemedicine during the COVID pandemic, and a second part, which was in the form of a Likert scale, covering perceptions related to telemedicine. The Likert scale itself had two main areas: (1) attitudes toward telemedicine and (2) perceived barriers. Results: There were 392 valid responses of which 228 (58.1%) had used some form of telemedicine (other than standard phone calls) during the COVID-19 pandemic. The most common platforms used for telemedicine include WhatsApp ® (211, 53.8%), Zoom ® (131, 33.4%), Microsoft Teams ® (27, 6.2%), Sehha App (65, 16.5%), Email (84, 21.4%). There was a strong agreement on the following statements: “Telemedicine can reduce unnecessary outpatient visits” (87.5%), “Effectiveness of telemedicine depends on the specialty” (89.5%), and “Telemedicine can be used to monitor chronic patients from home” (88.3%). Concerning the barriers to telemedicine, the ones having the most concordance were technological limitations (66.6%) and concerns of diagnostic reliability (66.1%). Conclusions: The responses from our study seem to suggest that while the attitudes toward telemedicine are positive, practicing physicians are concerned about a perceived lack of clarity regarding related legal frameworks and barriers such as technological issues, cultural factors, and diagnostic concordance.
Journals
2020 EN
Samiullah Shaikh · Asif Ameer · Shaikh S
Background: Diabetes is the leading cause of nontraumatic amputation. Foot screening which detects and stratification of diabetics which are at the risk of developing diabetic foot ulcer is the simple and useful part of this model of care. Aims: Primary Aim: To stratify patients with type II diabetes into different risk categories of diabetic foot as per International Diabetic Federation guidelines. Secondary Aim: To determine the relationship of various risk factors with risk categories of diabetic foot. Study Design: Cross sectional study. Place and Duration of Study: Department of Medicine, Liaquat University Hospital Jamshoro / Hyderabad from February 2019 to August 2020. Methodology: This study included 117 consecutive patients with confirmed diagnosis of Type-II diabetes of either sex ≥ 18 years of age. Original Research Article Shaikh et al.; JAMMR, 32(23): 177-186, 2020; Article no.JAMMR.63189 178 Patients fulfilling above criteria were included in study. Feet were thoroughly examined for neuropathy, peripheral vascular disease, infections, ulcers and osteoarthropathy. All the data was recorded on proforma. Patients having normal protective sensations were put in low risk (category 0), those having loss of protective sensations in moderate risk (category 1), those having loss of protective sensations with either high pressure or poor circulation or structural foot deformities or onychomycosis in high risk (category 2) and those having past history of ulceration, amputation or neuropathic fracture were put in very high risk (category 3). Data was analyzed by using SPSS version. 20. Results: Total 117 patients of diabetic foot ulcer were studied, their mean age was 52.28±9.26 years, diabetic duration 10.21±8.10 years and mean HbA1c level was 10.07±1.96 mmol/l. Male were in majority 52.1%. Ulceration history was in 18.8% cases, amputation history was in 7.7% cases, 46 patients (39.3%) had risk category 1. A strong relationship was found between risk categories and age, sex, duration of diabetes, HBA1c. Conclusion: This study revealed that 33 (28%) patients attending the diabetic clinic were at high risk of developing diabetic ulcer.
Sciencedomain International
Journals
2020 EN
Subrab Khan Talpur · Mukesh Kumar · Ameer Abbass
+3 more
Objective: To detect treatment outcomes in current smokers, ex smokers and nonsmokers, in newly diagnosed pulmonary TB patients. Methodology: This cohort prospective study was conducted in the department of Microbiology, Basic Medical Sciences Institute, Jinnah Postgraduate Medical Centre Karachi, with the collaboration of different (DOTS) centers of Pathology Karachi Medical and Dental College, Karachi. All newly diagnosed pulmonary TB patients registered for treatment either of gender were included. The patients were divided into three groups. Group-A: Current smokers, Group-B: Exsmokers, Group-C: Non-smokers. Patients were followed for 6 months. Outcome was assessed in Original Research Article Talpur et al.; JPRI, 32(32): 32-38, 2020; Article no.JPRI.60641 33 terms of cured or failure. All the information was enrolled in pre-designed proforma. Data was analyzed by using SPSS version 20. Results: In current study mean age of non-smokers was 44.20±17.73, ex-smokers 43.13±15.67 years mean age of smokers was 38.07±15.67 years. Males were in majority in all study groups as in ex-smokers group, males were 98.75%, non-smoker males were 90.0% and in smoker group males were 92.50%. At starting of treatment, mean weight of smoker patients was 55.50±5.41 kg, ex-smoker’s 48.91±9.00 kg and mean weight of non-smokers 48.71±7.04 kg. P-0.001. At starting of treatment, the mean ESR of non-smokers was 89.31±10.02, ex-smoker’s was 82.62±12.18 and smokers ESR average was 80.61±15.83. P-0.001. After 6 month treatment, cured rate was (96.25%) in non-smokers, (90%) in smokers and (93.75%) in ex–smokers. Conclusion: This study concluded that cured rate was high in non-smokers. Smoking status in individuals greatly affects the tuberculosis treatment outcome with enhance failure rate.
Sciencedomain International
Resource
2020 EN
Ameer Al-Abayechi · Ágota Figula
In this paper we deal with the class C of decomposable solvable Lie groupshaving dimension at most six. We determine those Lie groups in C and theirsubgroups which are the multiplication group Mult(L) and the inner mappinggroup Inn(L) for three-dimensional connected simply connected topological loopsL. These loops L have one- or two-dimensional centre and their group Mult(L)has two- or three-dimensional commutator subgroup. Together with this result weobtain that every at most 3-dimensional connected topological proper loophaving a solvable Lie group of dimension at most six as its multiplicationgroup is centrally nilpotent of class two.
Resource
2020 EN
Ameer Haj-Ali · Nesreen K. Ahmed · Ted Willke
+3 more
One of the key challenges arising when compilers vectorize loops for today'sSIMD-compatible architectures is to decide if vectorization or interleaving isbeneficial. Then, the compiler has to determine how many instructions to packtogether and how many loop iterations to interleave. Compilers are designedtoday to use fixed-cost models that are based on heuristics to makevectorization decisions on loops. However, these models are unable to capturethe data dependency, the computation graph, or the organization ofinstructions. Alternatively, software engineers often hand-write thevectorization factors of every loop. This, however, places a huge burden onthem, since it requires prior experience and significantly increases thedevelopment time. In this work, we explore a novel approach for handling loopvectorization and propose an end-to-end solution using deep reinforcementlearning (RL). We conjecture that deep RL can capture different instructions,dependencies, and data structures to enable learning a sophisticated model thatcan better predict the actual performance cost and determine the optimalvectorization factors. We develop an end-to-end framework, from code tovectorization, that integrates deep RL in the LLVM compiler. Our proposedframework takes benchmark codes as input and extracts the loop codes. Theseloop codes are then fed to a loop embedding generator that learns an embeddingfor these loops. Finally, the learned embeddings are used as input to a Deep RLagent, which determines the vectorization factors for all the loops. We furtherextend our framework to support multiple supervised learning methods. Weevaluate our approaches against the currently used LLVM vectorizer and looppolyhedral optimization techniques. Our experiments show 1.29X-4.73Xperformance speedup compared to baseline and only 3% worse than the brute-forcesearch on a wide range of benchmarks.
Resource
2020 EN
Keertana Settaluri · Ameer Haj-Ali · Qijing Huang
+2 more
Domain specialization under energy constraints in deeply-scaled CMOS has beendriving the need for agile development of Systems on a Chip (SoCs). Whiledigital subsystems have design flows that are conducive to rapid iterationsfrom specification to layout, analog and mixed-signal modules face thechallenge of a long human-in-the-middle iteration loop that requires expertintuition to verify that post-layout circuit parameters meet the originaldesign specification. Existing automated solutions that optimize circuitparameters for a given target design specification have limitations of beingschematic-only, inaccurate, sample-inefficient or not generalizable. This workpresents AutoCkt, a machine learning optimization framework trained using deepreinforcement learning that not only finds post-layout circuit parameters for agiven target specification, but also gains knowledge about the entire designspace through a sparse subsampling technique. Our results show that formultiple circuit topologies, AutoCkt is able to converge and meet all targetspecifications on at least 96.3% of tested design goals in schematicsimulation, on average 40X faster than a traditional genetic algorithm. Usingthe Berkeley Analog Generator, AutoCkt is able to design 40 LVS passedoperational amplifiers in 68 hours, 9.6X faster than the state-of-the-art whenconsidering layout parasitics.
Resource
2020 EN
Qijing Huang · Ameer Haj-Ali · William Moses
+4 more
The performance of the code a compiler generates depends on the order inwhich it applies the optimization passes. Choosing a good order--often referredto as the phase-ordering problem, is an NP-hard problem. As a result, existingsolutions rely on a variety of heuristics. In this paper, we evaluate a newtechnique to address the phase-ordering problem: deep reinforcement learning.To this end, we implement AutoPhase: a framework that takes a program and usesdeep reinforcement learning to find a sequence of compilation passes thatminimizes its execution time. Without loss of generality, we construct thisframework in the context of the LLVM compiler toolchain and target high-levelsynthesis programs. We use random forests to quantify the correlation betweenthe effectiveness of a given pass and the program's features. This helps usreduce the search space by avoiding phase orderings that are unlikely toimprove the performance of a given program. We compare the performance ofAutoPhase to state-of-the-art algorithms that address the phase-orderingproblem. In our evaluation, we show that AutoPhase improves circuit performanceby 28% when compared to using the -O3 compiler flag, and achieves competitiveresults compared to the state-of-the-art solutions, while requiring fewersamples. Furthermore, unlike existing state-of-the-art solutions, our deepreinforcement learning solution shows promising result in generalizing to realbenchmarks and 12,874 different randomly generated programs, after training ona hundred randomly generated programs.