Search

Charts are useful communication tools for the presentation of data in a visually appealing format that facilitates comprehension. There have been many studies dedicated to chart mining, which refers to the process of automatic detection, extraction and analysis of charts to reproduce the tabular data that was originally used to create them. By allowing access to data which might not be available in other formats, chart mining facilitates the creation of many downstream applications. This paper presents a comprehensive survey of approaches across all components of the automated chart mining pipeline, such as (i) automated extraction of charts from documents; (ii) processing of multi-panel charts; (iii) automatic image classifiers to collect chart images at scale; (iv) automated extraction of data from each chart image, for popular chart types as well as selected specialized classes; (v) applications of chart mining; and (vi) datasets for training and evaluation, and the methods that were used to build them. Finally, we summarize the main trends found in the literature and provide pointers to areas for further research in chart mining.

To study the clinical, laboratory characteristics and outcomes of multisystem inflammatory syndrome in children (MIS-C) temporally related to coronavirus disease 2019 (COVID-19) in a resource-limited setting. All children meeting the World Health Organization case definition of MIS-C were prospectively enrolled. Baseline clinical and laboratory parameters were compared between survivors and non-survivors. Enrolled subjects were followed up for 4–6 weeks for evaluation of cardiac outcomes using echocardiography. The statistical data were analyzed using the stata-12 software. Thirty-one children with MIS-C were enrolled in an 11-month period. Twelve children had preexisting chronic systemic comorbidity. Fever was a universal finding; gastrointestinal and respiratory manifestations were noted in 70.9% and 64.3%, respectively, while 57.1% had a skin rash. Fifty-eight percent of children presented with shock, and 22.5% required mechanical ventilation. HSP like rash, gangrene and arthritis were uncommon clinical observations.The median duration of hospital stay was 9 (6.5–18.5) days: four children with preexisting comorbidities succumbed to the illness. The serum ferritin levels (ng/ml) [median (IQR)] were significantly higher in non-survivors as compared to survivors [1061 (581, 2750) vs 309.5 (140, 720.08), p value = 0.045]. Six patients had coronary artery involvement; five recovered during follow-up, while one was still admitted. Twenty-six children received immunomodulatory drugs, and five improved without immunomodulation. The choice of immunomodulation (steroids or intravenous immunoglobulin) did not affect the outcome. Most children with MIS-C present with acute hemodynamic and respiratory symptoms.The outcome is favorable in children without preexisting comorbidities.Raised ferritin level may be a poor prognostic marker. The coronary outcomes at follow-up were reassuring. Supplementary Information The online version contains supplementary material available at 10.1007/s00296-021-05030-y.

An Intensive Care Unit (ICU) is an organized system for the provision of care to critically ill patients that provides intensive and specialized medical and nursing care, an enhanced capacity for monitoring, and multiple modalities of physiologic organ support to sustain life during a period of life-threatening organ system insufficiency. While this availability of trained manpower and specialized equipment makes it possible to care for critically ill patients, it also presents singular challenges in the form of man and material management, design concerns, budgetary concerns, and protocolization of treatment. Consequently, the establishment of an ICU requires rigorous design and planning, a process that can take months to years. However, the Coronavirus disease-19 (COVID-19) epidemic has required the significant capacity building to accommodate the increased number of critically ill patients. At the peak of the pandemic, many countries were forced to resort to the building of temporary structures to house critically ill patients, to help tide over the crisis. This narrative review describes the challenges and lessons learned while establishing a 250 bedded ICU in a temporary structure and achieving functionality within a period of a fortnight.

Introduction: Arterial blood gas (ABG) analysis is the gold standard method and frequently performed intervention to evaluate acid-base status along with adequacy of ventilation and oxygenation among patients with predominantly critical / acute diseases. Aims And Objectives: The aim of this study is to evaluate the correlation of VBG analysis and pulse oximetry (SpO2) with ABG analysis in critically ill patients. Materials And Methods:Intensive Care Unit (ICU), Command Hospital (Eastern Command), Kolkata, Adult patients requiring arterial blood gas analysis, JAN 2018 –JUNE 2019, 100 critically ill patients and Age – 18yrs and older, Sex – Either sex. Conclusion: In this study population of critically ill patients, pH and pCO2 on VBG analysis correlated with pH and pCO2 on ABG analysis. The SpO2 correlated well with pO2 on ABG analysis

Background Prediction of fluid responsiveness in hypotensive patients is a challenge. The correlation between a novel noninvasive dynamic indicator, Pleth Variability Index (PVI ®), and a gold-standard Systolic Pressure Variation (SPV) as a measure of fluid responsiveness was assessed in the Intensive Care Unit (ICU) or Operation Theatre (OT) in a tertiary care hospital. Methods A prospective experimental study was conducted over a span of one year on 100 mechanically ventilated patients with hypotension. Vital parameters along with SPV and PVI ® were recorded before and after a standard volume expansion protocol. A 10% SPV threshold was used to define fluid responders and nonresponders. Results Pearson’s correlation graph at baseline showed positive correlation between PVI ® and SPV (r = 0.59, p-value = 0.001). Strength of correlation was comparatively less but still showed positive correlation at 15 (r = 0.39, p-value = 0.009) and 30 (r = 0.404, p-value = 0.004) minutes of fluid bolus. The Bland Altman analysis of baseline values of PVI ® and SPV showed good agreement with a mean bias of 9.05. Percentage change of PVI ® and SPV over 30 min showed a statistically significant positive correlation in the responder group (r = 0.53, p Conclusion A positive correlation exists between SPV and PVI ®, justifying the use of noninvasive PVI ® in a clinical setting of hypotension.

The COVID-19 pandemic has gripped the world since January 2020 and has changed our lives in unprecedented ways. It has changed the way we work in the Operation Theatres and Intensive Care Units mainly because of the high risk of disease transmission to the healthcare workers. In order to reduce the risk of disease transmission, an understanding of the means of transmission of the virus is essential to develop a rational strategy that allows patients to receive treatment without placing either the patient or healthcare workers at risk. It should be cautioned that this is a rapidly changing field and there is a paucity of randomised controlled trials related to various aspects of the disease. It is therefore advisable to revise any recommendations in this article, as and when new evidence emerges.

Background MicroRNAs have recently emerged as promising circulating biomarkers in diverse cancer types, including ovarian cancer. We utilized conditional, doxycycline‐induced fallopian tube (FT)‐derived cancer models to identify changes in miRNA expression in tumors and plasma, and further validated the murine findings in high‐grade ovarian cancer patient samples. Methods We analyzed 566 biologically informative miRNAs in doxycycline‐induced FT and metastatic tumors as well as plasma samples derived from murine models bearing inactivation of Brca, Tp53, and Pten genes. We identified miRNAs that showed a consistent pattern of dysregulated expression and validated our results in human patient serum samples. Results We identified six miRNAs that were significantly dysregulated in doxycycline‐induced FTs (P, We have identified miRNAs that accompany fallopian tube‐derived ovarian cancer progression in murine models. Furthermore, we validated some of the miRNAs in patient serum samples.

Target-agnostic person tracking and re-identification across multiple non-overlapping cameras is an open vision problem. It is the task of maintaining the correct identity of people at different time instances and possibly different cameras. This study focuses on existing algorithms that facilitate online person tracking by using discriminative spatio-temporal features from video data, and presents the open issues and future research directions. The initial take on the problem introduces person tracking as a pure association problem, where the influence of human appearance, biometric and location information on re-identification are addressed explicitly. These constraints are modeled and used to understand and associate detections in real world environments. Next, a spatio-temporal model using LSTM networks for propagating associations and recovering from errors by taking advantage of the spatial and temporal information in videos is described. The spatio-temporal context indicates a way for discriminative appearance learning. The novelty of the mentioned approaches is that they do not require to learn target-specific appearance models and collect samples to distinguish different people from each other. The methods are evaluated on large-scale tracking datasets. State-of-the-art performance is achieved using motion metadata such as person bounding box and camera number, and shows better associations for the challenging exit-entry cases.

We propose a framework to extract and binarize handwritten content in lecture videos. The extracted content could potentially be used to index video collections powering content-based search and navigation within lecture videos helping students and educators across the world. A deep learning pipeline is used to detect handwritten text, formulae and sketches and then binarize the extracted content. We exploit the spatio-temporal structure of our binarized detections to compute associativity information of content across all video frames. This information is later used to segment the video. Experiments are conducted to compare the performance of key components of our framework in isolation, as well as the impact on overall performance, with respect to existing methods. We evaluate our framework on the publicly available AccessMath lecture video dataset obtaining an f-measure of $$94.32\%$$ for binary connected components. Code for the framework (including trained weights) and summarization will be released.

Multistage, or serial, fusion refers to the algorithms sequentially fusing an increased number of matching results at each step and making decisions about accepting or rejecting the match hypothesis, or going to the next step. Such fusion methods are beneficial in the situations where running additional matching algorithms needed for later stages is time consuming or expensive. The construction of multistage fusion methods is challenging, since it requires both learning fusion functions and finding optimal decision thresholds for each stage. In this paper, we propose the use of single neural network for learning the multistage fusion. In addition we discuss the choices for the performance measurements of the trained algorithms and for the selection of network training optimization criteria. We perform the experiments using three face matching algorithms and IJB-A and IJB-C databases.

Background: We studied the safety of percutaneous dilatational tracheostomy (PCDT) in severe acute respiratory syndrome novel coronavirus 2 (SARS-nCoV2). Patients & Methods: From 01 March 20 to 30 November 2020, 1635 required hospital admission of which 145 (9%) required intensive (ICU) care. The primary outcomes are mortality and secondary outcomes were duration of invasive mechanical ventilation (IMV), length of stay (LOS) in ICU and hospital, & days required for decannulation.Results: Out of the 145 (9%), 107 (73.7%) were males (mean 61.4 years, median body mass index (BMI) of 28.2 kg/m2 ), 38 (26.2%) were females (mean 58.10 years, median BMI of 31.2 kg/m2). In the cohort of 80 (55.17%) requiring IMV, 19 (23.7%) died within 72 hours and were not included in the study, 37 (group “NT”) and 24 (group “T”) had a median duration of ventilation of 9 d (IQR, 6-11) and 12 d (IQR, 11-17.25) respectively. Patients in group “T” underwent PCDT based on clinical criteria (fraction of inspired oxygen (FiO2 ) of ≤ 50% with positive end-expiratory pressure (PEEP) of ≤ 10 cms of H2O with stable hemodynamics), 16 (66.7%) had survived. The reverse transcription-polymerase chain reaction (RT-PCR) does not need to be negative, and none of the health care workers (HCW’s) were infected. The Cox-hazard ratio [HR] is 0.19, 95% Confidence interval [CI] (0.09, 0.41) with a P-value of Conclusions: PCDT is safe and effective in patients anticipated in need of prolonged mechanical ventilation.

OBJECTIVES: Misdiagnosis of acute and chronic otitis media in children can result in significant consequences from either undertreatment or overtreatment. Our objective was to develop and train an artificial intelligence algorithm to accurately predict the presence of middle ear effusion in pediatric patients presenting to the operating room for myringotomy and tube placement. METHODS: We trained a neural network to classify images as “ normal” (no effusion) or “abnormal” (effusion present) using tympanic membrane images from children taken to the operating room with the intent of performing myringotomy and possible tube placement for recurrent acute otitis media or otitis media with effusion. Model performance was tested on held-out cases and fivefold cross-validation. RESULTS: The mean training time for the neural network model was 76.0 (SD ± 0.01) seconds. Our model approach achieved a mean image classification accuracy of 83.8% (95% confidence interval [CI]: 82.7–84.8). In support of this classification accuracy, the model produced an area under the receiver operating characteristic curve performance of 0.93 (95% CI: 0.91–0.94) and F1-score of 0.80 (95% CI: 0.77–0.82). CONCLUSIONS: Artificial intelligence–assisted diagnosis of acute or chronic otitis media in children may generate value for patients, families, and the health care system by improving point-of-care diagnostic accuracy. With a small training data set composed of intraoperative images obtained at time of tympanostomy tube insertion, our neural network was accurate in predicting the presence of a middle ear effusion in pediatric ear cases. This diagnostic accuracy performance is considerably higher than human-expert otoscopy-based diagnostic performance reported in previous studies.

Visualization recommendation (VisRec) systems provide users with suggestions for potentially interesting and useful next steps during exploratory data analysis. These recommendations are typically organized into categories based on their analytical actions, i.e., operations employed to transition from the current exploration state to a recommended visualization. However, despite the emergence of a plethora of VisRec systems in recent work, the utility of the categories employed by these systems in analytical workflows has not been systematically investigated. Our paper explores the efficacy of recommendation categories by formalizing a taxonomy of common categories and developing a system, Frontier, that implements these categories. Using Frontier, we evaluate workflow strategies adopted by users and how categories influence those strategies. Participants found recommendations that add attributes to enhance the current visualization and recommendations that filter to sub-populations to be comparatively most useful during data exploration. Our findings pave the way for next-generation VisRec systems that are adaptive and personalized via carefully chosen, effective recommendation categories., 10 pages. This work has been submitted to IEEE TVCG. Copyright may be transferred without notice, after which this version may no longer be accessible

Digital Subtraction Angiography (DSA) is considered the gold standard for imaging and guiding treatment of neurovascular lesions, such as cerebral aneurysms and carotid stenoses. Though DSA can show high-resolution morphology, it remains difficult to extract temporal physiological information, because higher frame-rates are necessary to accurately quantify neurovascular flow details. Recent advances in photon-counting detector technology have led us to develop High-Speed Angiography (HSA), where X-ray images are acquired at 1000 fps for more accurate visualization and quantification of blood flow. Blood flow was imaged using HSA under constant flow conditions within various 3D printed patient-specific phantoms. Blood velocity was quantified using an open source Optical Flow algorithm, OpenOpticalFlow, to perform velocity estimation based on the spatio-temporal intensity changes of iodinated contrast wavefronts. The results of these algorithms are then compared with Computational Fluid Dynamics (CFD) simulations, using the same inlet boundary conditions and model geometries. The performance of these algorithms at lower temporal resolution was then also assessed by simulating lower frame rates from the acquired 1000 fps data. It is important to ascertain the hemodynamic effect of abnormal neurovascular conditions, as well as their effect on treatment of such conditions during the actual clinical interventional procedure. While theoretical CFD results requiring considerable computer capability are delayed for hours or more, it is expected that clinical results from multiple HSA sequences will be available almost immediately while the patient is still under treatment, and even right after flow conditions are changed beneficially by the intervention.

Lecture videos are rapidly becoming an invaluable source of information for students across the globe. Given the large number of online courses currently available, it is important to condense the information within these videos into a compact yet representative summary that can be used for search-based applications. We propose a framework to summarize whiteboard lecture videos by finding feature representations of detected handwritten content regions to determine unique content. We investigate multi-scale histogram of gradients and embeddings from deep metric learning for feature representation. We explicitly handle occluded, growing and disappearing handwritten content. Our method is capable of producing two kinds of lecture video summaries - the unique regions themselves or so-called key content and keyframes (which contain all unique content in a video segment). We use weighted spatio-temporal conflict minimization to segment the lecture and produce keyframes from detected regions and features. We evaluate both types of summaries and find that we obtain state-of-the-art peformance in terms of number of summary keyframes while our unique content recall and precision are comparable to state-of-the-art.

Representational Learning in the form of high dimensional embeddings have been used for multiple pattern recognition applications. There has been a significant interest in building embedding based systems for learning representations in the mathematical domain. At the same time, retrieval of structured information such as mathematical expressions is an important need for modern IR systems. In this work, our motivation is to introduce a robust framework for learning representations for similarity based retrieval of mathematical expressions. Given a query by example, the embedding can find the closest matching expression as a function of euclidean distance between them. We leverage recent advancements in image-based and graph-based deep learning algorithms to learn our similarity embeddings. We do this first, by using unimodal encoders in graph space and image space and then, a multi-modal combination of the same. To overcome the lack of training data, we force the networks to learn a deep metric using triplets generated with a heuristic scoring function. We also adopt a custom strategy for mining hard samples to train our neural networks. Our system produces rankings similar to those generated by the original scoring function, but using only a fraction of the time. Our results establish the viability of using such a multi-modal embedding for this task.

The volume of online lecture videos is growing at a frenetic pace. This has led to an increased focus on methods for automated lecture video analysis to make these resources more accessible. These methods consider multiple information channels including the actions of the lecture speaker. In this work, we analyze two methods that use spatio-temporal features of the speaker skeleton for action classification in lecture videos. The first method is the AM Pose model which is based on Random Forests with motion-based features. The second is a state-of-the-art action classifier based on a two-stream adaptive graph convolutional network (2S-AGCN) that uses features of both joints and bones of the speaker skeleton. Each video is divided into fixed-length temporal segments. Then, the speaker skeleton is estimated on every frame in order to build a representation for each segment for further classification. Our experiments used the AccessMath dataset and a novel extension which will be publicly released. We compared four state-of-the-art pose estimators: OpenPose, Deep High Resolution, AlphaPose and Detectron2. We found that AlphaPose is the most robust to the encoding noise found in online videos. We also observed that 2S-AGCN outperforms the AM Pose model by using the right domain adaptations.

Data produced from a particular source often exhibit correlations with those arising from other sources. Common data sources include (i) sensors—that gather raw data, (ii) feature extractors—which process the raw data from sensors to generate features representing the original data, and (iii) evaluators- –which produce a score or a measure that conveys the likelihood of the provided features belonging to an application specific hypothesis.

Well-designed data visualizations can lead to more powerful and intuitive processing by a viewer. To help a viewer intuitively compare values to quickly generate key takeaways, visualization designers can manipulate how data values are arranged in a chart to afford particular comparisons. Using simple bar charts as a case study, we empirically tested the comparison affordances of four common arrangements: vertically juxtaposed, horizontally juxtaposed, overlaid, and stacked. We asked participants to type out what patterns they perceived in a chart, and coded their takeaways into types of comparisons. In a second study, we asked data visualization design experts to predict which arrangement they would use to afford each type of comparison and found both alignments and mismatches with our findings. These results provide concrete guidelines for how both human designers and automatic chart recommendation systems can make visualizations that help viewers extract the 'right' takeaway., Comment: 12 pages (9 + 2 pages of references), 9 figures, for IEEE VIS conference (TVCG journal)

This work summarizes the results of the second Competition on Harvesting Raw Tables from Infographics (ICPR 2020 CHART-Infographics). Chart Recognition is difficult and multifaceted, so for this competition we divide the process into the following tasks: Chart Image Classification (Task 1), Text Detection and Recognition (Task 2), Text Role Classification (Task 3), Axis Analysis (Task 4), Legend Analysis (Task 5), Plot Element Detection and Classification (Task 6.a), Data Extraction (Task 6.b), and End-to-End Data Extraction (Task 7). We provided two sets of datasets for training and evaluation of the participant submissions. The first set is based on synthetic charts (Adobe Synth) generated from real data sources using matplotlib. The second one is based on manually annotated charts extracted from the Open Access section of the PubMed Central (UB PMC). More than 25 teams registered out of which 7 submitted results for different tasks of the competition. While results on synthetic data are near perfect at times, the same models still have room to improve when it comes to data extraction from real charts. The data, annotation tools, and evaluation scripts have been publicly released for academic use.

Natural language interfaces (NLIs) have become a prevalent medium for conducting visual data analysis, enabling people with varying levels of analytic experience to ask questions of and interact with their data. While there have been notable improvements with respect to language understanding capabilities in these systems, fundamental user experience and interaction challenges including the lack of analytic guidance (i.e., knowing what aspects of the data to consider) and discoverability of natural language input (i.e., knowing how to phrase input utterances) persist. To address these challenges, we investigate utterance recommendations that contextually provide analytic guidance by suggesting data features (e.g., attributes, values, trends) while implicitly making users aware of the types of phrasings that an NLI supports. We present SNOWY, a prototype system that generates and recommends utterances for visual analysis based on a combination of data interestingness metrics and language pragmatics. Through a preliminary user study, we found that utterance recommendations in SNOWY support conversational visual analysis by guiding the participants' analytic workflows and making them aware of the system's language interpretation capabilities. Based on the feedback and observations from the study, we discuss potential implications and considerations for incorporating recommendations in future NLIs for visual analysis., Comment: To appear at the 2021 ACM Symposium on User Interface Software and Technology (UIST); 17 pages (10 figures, 2 tables)

Humans have the innate ability to rapidly spot and react to a person’s emotional response. For computers to be able to understand expressions in a similar way, the gap between the perception of expressions between the humans and computers needs to be minimized. Inspired by human visual fixations, we propose a guided attention mechanism that facilitates the network to ‘look’ at the most important features of a face. Rather than imposing hard attention, we learn the attention maps from the intermediate representation for Action Units (AUs). We propose a joint attention learning and AU classification module with minimal increase in the network parameters. We demonstrate the efficiency of our approach on three standard datasets: BP4D, MMSE and DISFA and obtain state of the art and near state of the art results respectively.

Natural language interaction with data visualization tools often involves the use of vague subjective modifiers in utterances such as "show me the sectors that are performing" and "where is a good neighborhood to buy a house?." Interpreting these modifiers is often difficult for these tools because their meanings lack clear semantics and are in part defined by context and personal user preferences. This paper presents a system called \system that makes a first step in better understanding these vague predicates. The algorithm employs word co-occurrence and sentiment analysis to determine which data attributes and filters ranges to associate with the vague predicates. The provenance results from the algorithm are exposed to the user as interactive text that can be repaired and refined. We conduct a qualitative evaluation of the Sentifiers system that indicates the usefulness of the interface as well as opportunities for better supporting subjective utterances in visual analysis tasks through natural language.

This paper extends recent work on nonlinear Independent Component Analysis (ICA) by introducing a theoretical framework for nonlinear Independent Subspace Analysis (ISA) in the presence of auxiliary variables. Observed high dimensional acoustic features like log Mel spectrograms can be considered as surface level manifestations of nonlinear transformations over individual multivariate sources of information like speaker characteristics, phonological content etc. Under assumptions of energy based models we use the theory of nonlinear ISA to propose an algorithm that learns unsupervised speech representations whose subspaces are independent and potentially highly correlated with the original non-stationary multivariate sources. We show how nonlinear ICA with auxiliary variables can be extended to a generic identifiable model for subspaces as well while also providing sufficient conditions for the identifiability of these high dimensional subspaces. Our proposed methodology is generic and can be integrated with standard unsupervised approaches to learn speech representations with subspaces that can theoretically capture independent higher order speech signals. We evaluate the gains of our algorithm when integrated with the Autoregressive Predictive Decoding (APC) model by showing empirical results on the speaker verification and phoneme recognition tasks., To be presented at Interspeech 2020

Deep metric learning leverages carefully designed sampling strategies and loss functions that aid in optimizing the generation of a discriminable embedding space. While effective sampling of pairs is critical for shaping the metric space during training, the relative interactions between pairs, and consequently the forces exerted on these pairs that direct their displacement in the embedding space can significantly impact the formation of well separated clusters. In this work, we identify a shortcoming of existing loss formulations which fail to consider more optimal directions of pair displacements as another criterion for optimization. We propose a novel direction regularization to explicitly account for the layout of sampled pairs and attempt to introduce orthogonality in the representations. The proposed regularization is easily integrated into existing loss functions providing considerable performance improvements. We experimentally validate our hypothesis on the Cars-196, CUB-200 and InShop datasets and outperform existing methods to yield state-of-the-art results on these datasets.

Background Three-dimensional (3D) printing has revolutionized training, education, and device testing. Understanding the design and physical properties of 3D-printed models is important. Objective To systematically review the design, physical properties, accuracy, and experimental outcomes of 3D-printed vascular models used in neurointervention. Methods We conducted a systematic review of the literature between January 1, 2000 and September 30, 2018. Public/Publisher MEDLINE (PubMed), Web of Science, Compendex, Cochrane, and Inspec databases were searched using Medical Subject Heading terms for design and physical attributes of 3D-printed models for neurointervention. Information on design and physical properties like compliance, lubricity, flow system, accuracy, and outcome measures were collected. Results A total of 23 articles were included. Nine studies described 3D-printed models for stroke intervention. Tango Plus (Stratasys) was the most common material used to develop these models. Four studies described a population-representative geometry model. All other studies reported patient-specific vascular geometry. Eight studies reported complete reconstruction of the circle of Willis, anterior, and posterior circulation. Four studies reported a model with extracranial vasculature. One prototype study reported compliance and lubricity. Reported circulation systems included manual flushing, programmable pistons, peristaltic, and pulsatile pumps. Outcomes included thrombolysis in cerebral infarction, post-thrombectomy flow restoration, surgical performance, and qualitative feedback. Conclusion Variations exist in the material, design, and extent of reconstruction of vasculature of 3D-printed models. There is a need for objective characterization of 3D-printed vascular models. We propose the development of population representative 3D-printed models for skill improvement or device testing.

Coronaviruses are a group of viruses that cause various diseases in mammals and birds. In humans, they cause a range of respiratory disorders. This paper presents the analysis of the transmission of COVID19 disease and predicts the scale of the pandemic, the recovery rate as well as the fatality rate. We have used some of the well-known machine learning techniques as well as mathematical modeling techniques such as Rough Set-Support Vector Machine (RS-SVM), Bayesian Ridge and Polynomial Regression, SIR model, and RNN.

Background and aim Imparting the knowledge and skills of Anesthesiology to undergraduates can be challenging. Competency Based Undergraduate (CBUG) Curriculum for the Undergraduate medical students introduced by the Medical Council of India (MCI) aims to improve the quality of the Indian Medical Graduate (IMG). The Department of Anesthesiology and Critical Care of our college redrafted the training program and brought it in-line with the CBUG Curriculum beginning February 2019. A questionnaire based survey was conducted to assess the efficacy, satisfaction levels and the perception of the students towards the new competency based curriculum. The aim was to assess the students perception of the competency based curriculum and to evaluate two slightly different approaches to the implementation of the curriculum. Material and methods Two groups of undergraduate medical students belonging to the 6th and 8th term, underwent two different models of teaching. The 8th term students had already completed their theory classes based on the older curriculum a year ago when they were in 6th term. However, their clinics and tutorials were modelled as per the new CBUG Curriculum. The current 6th term students had their first exposure to Anesthesiology and their theory, tutorials and clinics were scheduled in the same term, simulation based training was added, the operation theatre rotation was held in the mornings at 0730hrs and the intensive care unit rounds were held in the evenings. There was no difference in the theory classes taken for the two batches, however the clinics were different. After both the batches finished their rotation, they were given the survey questionnaire to assess their perception of the model of CBUG Curriculum that they were exposed to. Results The results of the survey revealed that about 80% of the students in both groups preferred that theory classes and practical training should be conducted in parallel in 6th term. About 60% students in both groups felt that early morning clinics 0800hrs were better than mid-morning clinics at 1100hrs as they get to see and do more procedures. 66%-82% students in both groups felt that the practical training in the OT, ICU and skills lab were very helpful or extremely helpful. The most important aspect of Anesthesiology rotation was "learning basic life saving skills and simulation based learning" according to 85% students in both groups. Nearly 80% students in both groups felt that the training in Anesthesiology should be allotted more time and more weightage in undergraduate training. 72% students in 6th term and 63% students in 8th term felt more confident of handling emergencies after their Anesthesiology rotation. Conclusion The new curriculum was extremely well received by the students of both groups. The model used for 6th term students comprising of teaching theory and practical in the same term and having early morning clinics, was found to be superior as compared to the model used to teach 8th term students where there was a gap of one year between theory and practical teaching and the clinics were held midmorning.

A new imaging x-ray detector with 194μm pixel conventional FPD imaging mode for large area visualization with detector field-of-views (FOV) ranging from 12 inch to 6 inch and a 76μm pixel high resolution high-definition (Hi-Def) zoom mode for smaller region-of-interest (ROI) visualization with FOV’s from 3 to 1.5 inches in one detector panel was built to improve visualization of endovascular devices during neurovascular interventions. In this work we investigate the effect of simultaneous use of both imaging modes on entrance kerma area product (KAP) and scatter at the detector. The entrance KAP and the scatter reaching the detector during simultaneous acquisition is the same as the large FOV FPD mode acquisitions, both higher than just Hi-Def mode acquisition. When a 0.7mm thick differential ROI Cu attenuator with a circular 10mm hole was aligned to the Hi-Def detector area, the entrance air kerma is reduced by 70%. To study the effect of scatter on image quality, Hi-Def images of the devices such as guide catheter and stents placed on a head equivalent phantom were acquired with the x-ray field opened to 8inch FOV, with and without the grid. Compared to the 3 inch Hi-Def FOV, with the grid the contrast of the guide catheter tip was reduced by 11% when the FOV was opened to 8 inch; however, without the grid the reduction in contrast was 36%. With the ROI attenuator and no grid the contrast was reduced by 16% when the FOV was opened to 8inch.

E-commerce companies like Amazon, Alibaba and Flip-kart have an extensive catalogue comprising of billions of products. Matching customer search queries to plausible products is challenging due to the size and diversity of the catalogue. These challenges are compounded in apparel due to the semantic complexity and a large variation of fashion styles, product attributes and colours. Providing aids that can help the customer visualise the styles and colours matching their "search queries" will provide customers with necessary intuition about what can be done next. This helps the customer buy a product with the styles, embellishments and colours of their liking. In this work, we propose a Generative Adversarial Network (GAN) for generating images from text streams like customer search queries. Our GAN learns to incrementally generate possible images complementing the fine-grained style, colour of the apparel in the query. We incorporate a novel colour modelling approach enabling the GAN to render a wide spectrum of colours accurately. We compile a dataset from an e-commerce website to train our model. The proposed approach outperforms the baselines on qualitative and quantitative evaluations.

BACKGROUND AND PURPOSE: Quality of visualization of treatment devices during critical stages of endovascular interventions, can directly impact their safety and efficacy. Our aim was to compare the visualization of neurointerventional procedures and treatment devices using a 194-μm pixel flat panel detector mode and a 76-μm pixel complementary metal oxide semiconductor detector mode (high definition) of a new-generation x-ray detector system using a blinded-rater study. MATERIALS AND METHODS: Deployment of flow-diversion devices for the treatment of internal carotid artery aneurysms was performed under flat panel detector and high-definition-mode image guidance in a neurointerventional phantom simulating patient cranium and tissue attenuation, embedded with 3D-printed intracranial vascular models, each with an aneurysm in the ICA segment. Image-sequence pairs of device deployments for each detector mode, under similar exposure and FOV conditions, were evaluated by 2 blinded experienced neurointerventionalists who independently selected their preferred image on the basis of visualization of anatomic features, image noise, and treatment device. They rated their selection as either similar, better, much better, or substantially better than the other choice. Inter- and intrarater agreement was calculated and categorized as poor, moderate, and good. RESULTS: Both raters demonstrating good inter- and intrarater agreement selected high-definition-mode images with a frequency of at least 95% each and, on average, rated the high-definition images as much better than flat panel detector images with a frequency of 73% from a total of 60 image pairs. CONCLUSIONS: Due to their higher resolution, high-definition-mode images are sharper and visually preferred compared with the flat panel detector images. The improved imaging provided by the high-definition mode can potentially provide an advantage during neurointerventional procedures.

Aim: The primary objective of this review is to provide an updated, comprehensive overview on the efficacy of intranasal fentanyl (INF) for acute pain relief in the pediatric population. Methods: Utilizing the Preferred Reporting Instructions for Systematic Reviews and Meta-Analyses (PRISMA), we were able to screen articles based on key words to reach a final number of 10 studies. Results: All but one study showed that INF was efficacious for pain relief in this select pediatric population. Conclusion: It is evident that INF is efficacious for analgesia, but other agents should also be considered in this patient population. As a result, further research is needed to investigate the clinically efficacy of INF in an acute care setting for pediatric patients.

BackgroundDespite advances in revascularization tools for large vessel occlusion presenting as acute ischemic stroke, a significant subset of clots remain recalcitrant to current strategies. We assessed the effectiveness of a novel thrombectomy device that was specifically designed to retrieve resistant fibrin rich clots, the geometric clot extractor (GCE; Neuravi, Galway, Ireland), in an in vitro cerebrovascular occlusion stroke model.MethodsAfter introducing fibrin rich clot analogues into the middle cerebral artery of the model, we compared the rates of recanalization between GCE and Solitaire flow restoration stent retriever (SR; Medtronic, Minneapolis, Minnesota, USA; control group) cases. A maximum of three passes of each device was allowed. If the SR failed to recanalize the vessel after three passes, one pass of the GCE was allowed (rescue cases).ResultsIn a total of 26 thrombectomy cases (13 GCE, 13 SR), successful recanalization (Thrombolysis in Cerebral Infarction score of 2b or 3) was achieved 100% of the time in the GCE cases with an average of 2.13 passes per case. This rate was significantly higher compared with the Solitaire recanalization rate (7.7%, PConclusionApplication of the GCE in this experimental stroke model to retrieve typically recalcitrant fibrin rich clots resulted in higher successful recanalization rates than the SR.

Vascular pathologies such as stenosis cause changes in blood flow patterns and rates in patient vasculature. These changes are difficult to observe using conventional x-ray imaging techniques that typically operate at no more than 30 fps. However, 1000 fps High-Speed Angiography (HSA) provides sufficient temporal resolution to observe flow patterns in patient vasculature. The Actaeon detector from XCounter provides 1000 fps HSA with high spatial resolution (100 μm pixel pitch) enabling observation of detailed blood flow in 3D printed phantoms. To observe the effects of stenosis on flow, three carotid artery phantoms beginning 7.6 cm proximal to the bifurcation of the common carotid and extending 6.9 cm into the internal and external carotids with different degrees of stenosis (0%, 33%, and 66% stenosis) in the internal carotid were printed. Iodine contrast was injected into the phantom proximal to the bifurcation and images were recorded at 1000 fps. Time-density curves were generated for ROIs distal and proximal to the location of the stenosis for both branches of the carotid artery. The curves for the control phantom (no stenosis) showed little difference between the branches of the carotid while the stenotic phantoms showed an increased velocity in the flow of contrast and a lower maximum signal intensity for the stenosed internal carotid. 1000 fps HSA could be used during treatment of vascular pathologies to observe changes in flow patterns following an endovascular treatment.

Patient x-ray dose during fluoroscopically-guided neuro interventions can be reduced by using a differential Region-of-Interest (ROI) x-ray attenuator. The dose in the periphery region outside the region-of-interest (ROI) treatment area is reduced while maintaining regular dose within the ROI. In this work we present a convolutional neural network to aid in restoration of image quality in dose-reduced regions. A 0.7 mm Cu attenuator with a 10 mm circular hole in the middle was used to reduce entrance dose in the periphery. A 29 layer deep CNN was developed to derive the ROI attenuator mask image from the dose-reduced images. To train the CNN, simulated ROI dose-reduced images of various backgrounds such as anthropomorphic head and chest phantoms were generated using acquired mask images of the ROI attenuator at different positions and radiological magnifications. The image quality in the dose-reduced region of the images was restored by first dividing the CNN derived mask from the dose-reduced image and then noise in the periphery region was reduced by using a combination of Gaussian and recursive temporal filtering. A total dose-area-product reduction of 70% per frame was achieved. After image processing using the CNN derived image mask of the ROI attenuator, the SNR in the dose reduced periphery regions was improved by a factor of 3. The CNN is capable of deriving the mask without any prior knowledge of ROI attenuator position or radiological magnification. Using the CNN generated mask, the image quality in the dose reduced images was restored with minimal or no boundary artifacts.

ChemML is an open machine learning and informatics program suite that is designed to support and advance the data-driven research paradigm that is currently emerging in the chemical and materials domain. ChemML allows its users to perform various data science tasks and execute machine learning workflows that are adapted specifically for the chemical and materials context. Key features are automation, general-purpose utility, versatility, and user-friendliness in order to make the application of modern data science a viable and widely accessible proposition in the broader chemistry and materials community. ChemML is also designed to facilitate methodological innovation, and it is one of the cornerstones of the software ecosystem for data-driven in silico research outlined in our recent publication1.

Despite the advent of deep learning in computer vision, the general handwriting recognition problem is far from solved. Most existing approaches focus on handwriting datasets that have clearly written text and carefully segmented labels. In this paper, we instead focus on learning handwritten characters from maintenance logs, a constrained setting where data is very limited and noisy. We break the problem into two consecutive stages of word segmentation and word recognition respectively and utilize data augmentation techniques to train both stages. Extensive comparisons with popular baselines for scene-text detection and word recognition show that our system achieves a lower error rate and is more suited to handle noisy and difficult documents, Comment: icfhr2020

This paper introduces a new task of politeness transfer which involves converting non-polite sentences to polite sentences while preserving the meaning. We also provide a dataset of more than 1.39 instances automatically labeled for politeness to encourage benchmark evaluations on this new task. We design a tag and generate pipeline that identifies stylistic attributes and subsequently generates a sentence in the target style while preserving most of the source content. For politeness as well as five other transfer tasks, our model outperforms the state-of-the-art methods on automatic metrics for content preservation, with a comparable or better performance on style transfer accuracy. Additionally, our model surpasses existing methods on human evaluations for grammaticality, meaning preservation and transfer accuracy across all the six style transfer tasks. The data and code is located at https://github.com/tag-and-generate., Comment: To appear at ACL 2020

Current state-of-the-art methods in multi-modal fusion typically rely on generating a new shared representation space onto which multi-modal features are mapped for the goal of obtaining performance improvements by combining the individual modalities. Often, these heavily fine-tuned feature representations would have strong feature discriminability in their own spaces which may not be present in the fused subspace owing to the compression of information arising from multiple sources. To address this, we propose a new approach to fusion by enhancing the individual feature spaces through information exchange between the modalities. Essentially, domain adaptation is learnt by building a shared representation used for mutually enhancing each domain’s knowledge. In particular, the learning objective is modeled to modify the features with the overarching goal of improving the combined system performance. We apply our fusion method to the task of facial action unit (AU) recognition by learning to enhance the thermal and visible feature representations. We compare our approach to other recent fusion schemes and demonstrate its effectiveness on the MMSE dataset by outperforming previous techniques.

Introduction: The COVID-19 pandemic has provided opportunity to the Armed Forces Medical Services (AFMS) healthcare institutions to plan and execute their surge capacity facilities and identify areas for improvements in planning in the future. Material and Methods: Available medical literature on the experiences of other countries in activating surge capacities in healthcare for the pandemic were examined in detail as were existing guidelines for establishing Intensive Care Units (ICU). Personal communications with peers to understand difficulties faced in activating surge capacities were also factored in. Results: Based on the findings from these sources, a plan to establish ten-bedded ICU units specifically for COVID-19 is evolved. The best practices and latest guidelines and experiences have been collated and modified suitably to suit the AFMS in this aspect. Conclusion: Planning ICUs in ten-bedded modular units will enable the AFMS to cater to surge capacities in the future for all situations where sudden increase in number of patients is anticipated.

Background: Endovascular treatment of intracranial aneurysms (IAs) by flow diverter (FD) stents depends on flow modification. Patient-specific modeling of FD deployment and computational fluid dynamics (CFD) could enable a priori endovascular strategy optimization. We developed a fast, simplistic, expansion-free balls-weeping algorithm to model FDs in patientspecific aneurysm geometry. However, since such strong simplification could result in less accurate simulations, we also developed a fast virtual stenting workflow (VSW) that explicitly models stent expansion using pseudo-physical forces. Methods: To test which of these two fast algorithms more accurately simulates real FDs, we applied them to virtually treat three representative patient-specific IAs. We deployed Pipeline Embolization Device into 3 patient-specific silicone aneurysm phantoms and simulated the treatments using both balls-weeping and VSW algorithms in computational aneurysm models. We then compared the virtually deployed FD stents against experimental results in terms of geometry and post-treatment flow fields. For stent geometry, we evaluated gross configurations and porosity. For post-treatment aneurysmal flow, we compared CFD results against experimental measurements by particle image velocimetry. Results: We found that VSW created more realistic FD deployments than balls-weeping in terms of stent geometry, porosity and pore density. In particular, balls-weeping produced unrealistic FD bulging at the aneurysm neck, and this artifact drastically increased with neck size. Both FD deployment methods resulted in similar flow patterns, but the VSW had less error in flow velocity and inflow rate. Conclusion: In conclusion, modeling stent expansion is critical for preventing unrealistic bulging effects and thus should be considered in virtual FD deployment algorithms. Also endowed with its high computational efficiency and superior accuracy, the VSW algorithm is a better candidate for implementation into a bedside clinical tool for FD deployment simulation.

BackgroundA new dual resolution imaging x-ray detector system (Canon Medical Systems Corporation, Tochigi, Japan) has a standard resolution 194 µm pixel conventional flat-panel detector (FPD) mode and a high-resolution 76 µm high-definition (Hi-Def) mode in a single unit. The Hi-Def mode enhances the visualization of the intravascular devices.ObjectiveWe report the clinical experience and physician evaluation of this new detector system with Hi-Def mode for the treatment of intracranial aneurysms using a Pipeline embolization device (PED).MethodsDuring intervention at our institute, under large field of view (FOV) regular resolution FPD mode imaging, the catheter systems and devices were first guided to the proximity of the treatment area. Final placement and deployment of the PED was performed under Hi-Def mode guidance. A post-procedure 9-question physician survey was conducted to qualitatively assess the impact of Hi-Def mode visualization on physicians’ intraoperative decision-making. One-sample t-test was performed on the responses from the survey. Dose values reported by the x-ray unit were also recorded.ResultsTwenty-five cases were included in our study. The survey results indicated that, for each of the nine questions, the physicians in all cases indicated that the Hi-Def mode improved visualization compared with the FPD mode. For the 25 cases, the mean cumulative entrance air kerma was 2.35 Gy, the mean dose area product (DAP) was 173.71 Gy.cm2, and the mean x-ray exposure time was 39.30 min.ConclusionsThe Hi-Def mode improves visualization of flow diverters and may help in achieving more accurate placement and deployment of devices.

Natural language can be a useful modality for creating and interacting with visualizations but users often have unrealistic expectations about the intelligence of natural language systems. The gulf between user expectations and system capabilities may lead to a disappointing user experience. So — if we want to engineer a natural language system, what are the requirements around system intelligence? This work takes a retrospective look at how we answered this question in the design of Ask Data, a natural language interaction feature for Tableau. We examine two factors contributing to perceived system intelligence: the system's ability to understand the analytic intent behind an input utterance and the ability to interpret an utterance contextually (i.e. taking into account the current visualization state and recent actions). Our aim was to understand the ways in which a system would need to support these two aspects of intelligence to enable a positive user experience. We first describe a pre-design Wizard of Oz study that offered insight into this question and narrowed the space of designs under consideration. We then reflect on the impact of this study on system development, examining how design implications from the study played out in practice. Our work contributes insights for the design of natural language interaction in visual analytics as well as a reflection on the value of pre-design empirical studies in the development of visual analytic systems.

Natural language interfaces for data visualizations tools are growing in importance, but little research has been done on how a system should respond to questions that contain vague modifiers like "high" and "expensive." This paper makes a first step toward design guidelines for this problem, based on existing research from cognitive linguistics and the results of a new empirical study with 274 crowdsourcing participants. A comparison of four bar chart-based views finds that highlighting the top items according to distribution-sensitive values is preferred in most cases and is a good starting point as a design guideline.