Search

Artificial intelligence (AI) is transforming the field of medicine. However, expansion and adoption of AI applications in the intraoperative domain has been more challenging. To understand why, it is crucial to examine the current state of intraoperative AI (iAI) from multiple perspectives, including research, engineering, and industry. In this review, we provide a comprehensive analysis of iAI technology by conducting a systematic review of both clinical and technological literature, consolidating an up-to-date database of AI devices that are FDA and CE approved for use in surgical settings, and analysing the trend of globally filed patents and research from industry. Our analysis provides an accurate depiction of the latest abilities, limitations, and potential of iAI. We propose future directions and challenges that need to be addressed to expedite the adoption of iAI, given that the current level of implementation may not fully align with the expectations generated by the significant attention and interest in AI technologies. Thus, it is imperative to fill this critical gap in literature to enable the realisation of the vast potential of iAI in transforming surgical practice.

A systematic review protocol for maternal health outcomes associated with child custody loss. The aim of this review is to synthesize the evidence of health outcomes associated with child removal amongst women in high-income countries. To this end, the proposed systematic review will answer the following questions: 1. What are the physical health outcomes associated with child removal for women? 2. What are the mental health outcomes associated with child removal for women?

Evaluation of various interventions through network meta-analysis to look at their effectiveness on linear growth in children aged 0-6 months.

Background: Improving the health of pregnant women is important to prevent adverse birth outcomes, such as preterm birth and low birthweight. We evaluated the comparative effectiveness of interventions under the domains of micronutrient, balanced energy protein, deworming, maternal education, and water sanitation and hygiene (WASH) for their effects on these adverse birth outcomes. Methods: For this network meta-analysis, we searched for randomized clinical trials (RCTs) of interventions provided to pregnant women in low- and middle-income countries (LMICs). We searched for reports published until September 17, 2019 and hand-searched bibliographies of existing reviews. We extracted data from eligible studies for study characteristics, interventions, participants’ characteristics at baseline, and birth outcomes. We compared effects on preterm birth ( Results: Our network meta-analyses were based on 101 RCTs (132 papers) pertaining to 206,531 participants. Several micronutrients and balanced energy food supplement interventions demonstrated effectiveness over standard-of-care. For instance, versus standard-of-care, micronutrient supplements for pregnant women, such as iron and calcium, decreased risks of preterm birth (iron: RR=0.70, 95% credible interval [Crl] 0.47, 1.01; calcium: RR=0.76, 95%Crl 0.56, 0.99). Daily intake of 1500kcal of local food decreased the risks of preterm birth (RR=0.36, 95%Crl 0.16, 0.77) and LBW (RR=0.17, 95%Crl 0.09, 0.29), respectively when compared to standard-of-care. Educational and deworming interventions did not show improvements in birth outcomes, and no WASH intervention trials reported on these adverse birth outcomes. Conclusion: We found several pregnancy interventions that improve birth outcomes. However, most clinical trials have only evaluated interventions under a single domain (e.g. micronutrients) even though the causes of adverse birth outcomes are multi-faceted. There is a need to combine interventions that of different domains as packages and test for their effectiveness. Registration: PROSPERO CRD42018110446; registered on 17 October 2018.

Background:Optimizing linear growth in children during complementary feeding period (CFP) (6-24 months) are critical for their development. Several interventions, such as micronutrient and food supplements, deworming, maternal education, and water, sanitation and hygiene (WASH), could potentially be provided to prevent stunting, but their comparative effectiveness is currently unclear. In this study, we evaluated comparative effectiveness of interventions under these domains on child linear growth outcomes of height-for-age z-score (HAZ) and stunting (HAZ Methods:For this study, we searched for low- and middle-income country (LMIC)-based randomized clinical trials (RCTs) of aforementioned interventions provided to children during CFP. We searched for reports published until September 17, 2019 and hand-searched bibliographies of existing reviews. We performed random-effects network meta-analysis (NMA) for HAZ and stunting.Results:The evidence base for our NMA was based on 79 RCTs (96 papers) involving 81,786 children. Among the micronutrients, compared to standard-of-care, iron + folic acid (IFA) (mean difference =0.08; 95% credible interval [CrI]: 0.01, 0.15) and multiple micronutrients (MMN) (mean difference =0.06; 95%CrI: 0.01, 0.11) showed improvements for HAZ; MMN also reduced the risks for stunting (RR=0.86; 95%Crl: 0.73, 0.98), whereas IFA did not (RR=0.92; 95%Crl: 0.64, 1.23). For food supplements, flour in the caloric range of 270-340 kcal (RR=0.73; 95%Crl: 0.51, 1.00) and fortified lipid-based nutrient supplements (LNS) containing 220-285 kcal (RR=0.80; 95%Crl: 0.66, 0.97) decreased the risk of stunting compared to standard-of-care, but these interventions and other food supplements did not show improvements for HAZ. Deworming, maternal education, and WASH interventions did not show improvements for HAZ nor stunting.Conclusion:While we found micronutrient and food supplements to be effective for HAZ and/or stunting, the evidence base for other domains in this life stage was limited, highlighting the need for more investigation.Registration:PROSPEROCRD42018110449; registered on 17 October 2018.

Collaboration and successful teamworking are important components of clinical practise, and these skills should be cultivated early in medical school. The breadth of current medical school curricula means that students often have limited exposure to clinical neurosciences. Since its inception in 2009, the Neurology and Neurosurgery Interest Group (NANSIG) has become a national (UK and Republic of Ireland) example of student and junior doctor synergistic collaboration to deliver educational materials, research, conferences, seminars and workshops, as well as advocating for diversity in this field. Recently, it has expanded to incorporate an international audience and cater for a larger group of young medical professionals. The organisation has overcome numerous challenges and is constantly innovating new approaches to harness the necessary knowledge, skills and network to succeed in a career in neurosciences, neurology and neurosurgery. This article summarises the initiatives undertaken by the group over its first 10 years of existence and its organisational structure, as well as its future plans.

Deep Neural Network (DNN) models have continuously been growing in size in order to improve the accuracy and quality of the models. Moreover, for training of large DNN models, the use of heterogeneous GPUs is inevitable due to the short release cycle of new GPU architectures. In this paper, we investigate how to enable training of large DNN models on a heterogeneous GPU cluster that possibly includes whimpy GPUs that, as a standalone, could not be used for training. We present a DNN training system, HetPipe (Heterogeneous Pipeline), that integrates pipelined model parallelism (PMP) with data parallelism (DP). In HetPipe, a group of multiple GPUs, called a virtual worker, processes minibatches in a pipelined manner, and multiple such virtual workers employ data parallelism for higher performance. We also propose a novel parameter synchronization model, which we refer to as Wave Synchronous Parallel (WSP) to accommodate both PMP and DP for virtual workers, and provide convergence proof of WSP. Our experimental results on a given heterogeneous setting show that with HetPipe, DNN models converge up to 49% faster compared to the state-of-the-art DP technique.

EQUATOR Checklist. EQUATOR Checklist for â Systematic review of basket trials, umbrella trials, and platform trials: A landscape analysis of master protocolsâ . (DOC 63 kb)

The Convolutional Neural Network (CNN) model, often used for image classification, requires significant training time to obtain high accuracy. To this end, distributed training is performed with the parameter server (PS) architecture using multiple servers. Unfortunately, scalability has been found to be poor in existing architectures. We find that the PS network is the bottleneck as it communicates a large number of gradients and parameters with the many workers. This is because synchronization with the many workers has to occur at every step of training. Depending on the model, communication can be in the several hundred MBs per synchronization. In this paper, we propose a scheme to reduce network traffic through layer placement that considers the resources that each layer uses. Through analysis of the characteristics of CNN, we find that placement of layers can be done in an effective manner. We then incorporate this observation within the TensorFlow framework such that layers can be automatically placed for more efficient training. Our evaluation making use of this placement scheme show that training time can be significantly reduced without loss of accuracy for many CNN models.

Supplementary Appendix. Supplementary to â Systematic review of basket trials, umbrella trials, and platform trials: A landscape analysis of master protocolsâ . (DOCX 604 kb)

Background:Exclusive breastfeeding (EBF) during the first six months of life is critical for child’s linear growth. While there is strong evidence in favor of EBF, the evidence with regards to other interventions for linear growth is unclear. We evaluated intervention domains of micronutrients, food supplements, deworming, maternal education, water sanitation and hygiene (WASH), and kangaroo care, for their comparative effectiveness on linear growth.Methods:For this review, we searched for randomized clinical trials (RCTs) of the interventions provided to infants aged 0-6 months and/or their breastfeeding mothers in low- and middle-income countries reporting on length-for-age z-score (LAZ), stunting, length, and head circumference. We searched for reports published until September 17th, 2019 and hand-searched bibliographies of existing reviews. For LAZ and stunting, we used network meta-analysis (NMA) to compare the effects of all interventions except for kangaroo care, where we used pairwise meta-analysis to compare its effects versus standard-of-care. For length and head circumference, we qualitatively summarized our findings.Results:We found 29 RCTs (40 papers) involving 35,119 mother and infant pairs reporting on the effects of aforementioned interventions on linear growth outcomes. Our NMA on LAZ found that compared to standard-of-care, multiple micronutrients administered to infants (MMN-C) improved LAZ (mean difference: 0.20; 95% credible interval [CrI]: 0.03,0.35), whereas supplementing breastfeeding mothers with MMN did not (MMN-M, mean difference: -0.02, 95%CrI: -0.18,0.13). No interventions including MMN-C (relative risk: 0.74; 95%CrI: 0.36,1.44) reduced risk for stunting compared to standard-of-care. Kangaroo care, on the other hand, improved head circumference (mean difference: 0.20 cm/week; 95% confidence intervals [CI]: 0.09,0.31 cm/week) and length (mean difference: 0.23 cm/week; 95%CI: 0.10,0.35 cm/week) compared to standard-of-care. Conclusion:Our study found important improvements for kangaroo care, but we did not find sufficient evidence for other interventions.Registration:PROSPEROCRD42018110450; registered on 17 October 2018.

Faculty Advisor Study Survey. (PDF 115 kb)

This file contains detailed figures to explain the additional informaton of the process and materials. These include solution images, FTIR, and TEM.

Second oral malignancy (SOM) is a common occurrence for patients with oral cancers, contributing to low 5-year survival rates for this disease (~60%). Oral mucosal changes at previously treated cancer sites are common and can be defined clinically by white-light examination and toluidine blue staining, and molecularly by loss of heterozygosity (LOH) analysis. To determine the role of such changes in predicting SOM, this study analyzed data collected from 194 oral squamous cell carcinoma patients in follow-up after treatment of which 31 (16%) developed SOM. Two features were shown to be associated with elevated risk of SOM: the clinical presence of an oral premalignant lesion (OPL) (P

Stiff, strong and tough ultrafine polyethylene fibers that rival the best high performance fibers, but with diameters less than one micron, are fabricated for the first time by “gel-electrospinning.” In this process, solution concentration and process temperatures are chosen to induce the formation of gel filaments “in flight,” which are subsequently drawn at high rates as a consequence of the whipping instability. The resulting submicron-diameter fibers exhibited Young’s moduli of 73 ± 13 GPa, yield strengths of 3.5 ± 0.6 GPa, and toughnesses of 1.8 ± 0.3 GPa, on average. Among the smallest fibers examined, one with a diameter of 490 ± 50 nm showed a Young’s modulus of 110 ± 16 GPa, ultimate tensile strength of 6.3 ± 0.9 GPa, and toughness of 2.1 ± 0.3 GPa, a combination of mechanical properties that is unparalleled among polymer fibers to date. The correlation of stiffness, strength and toughness with fiber diameter is attributed to high crystallinity and crystallite orientation, combined with fewer defects and enhanced chain slip associated with small diameter and high specific surface area. Gel-electrospinning improves the prospects for production of such fibers at scale., United States. Army. Natick Soldier Research, Development and Engineering Center