Single Anomaly Research Articles

HeartWare HVAD Market Withdrawal and Impact on the Pediatric Field.

HeartWare HVAD Market Withdrawal and Impact on the Pediatric Field.

SAND

With the increasing demand for real-time analytics and decision making, anomaly detection methods need to operate over streams of values and handle drifts in data distribution. Unfortunately, existing approaches have severe limitations: they either require prior domain knowledge or become cumbersome and expensive to use in situations with recurrent anomalies of the same type. In addition, subsequence anomaly detection methods usually require access to the entire dataset and are not able to learn and detect anomalies in streaming settings. To address these problems, we propose SAND, a novel online method suitable for domain-agnostic anomaly detection. SAND aims to detect anomalies based on their distance to a model that represents normal behavior. SAND relies on a novel steaming methodology to incrementally update such model, which adapts to distribution drifts and omits obsolete data. The experimental results on several real-world datasets demonstrate that SAND correctly identifies single and recurrent anomalies without prior knowledge of the characteristics of these anomalies. SAND outperforms by a large margin the current state-of-the-art algorithms in terms of accuracy while achieving orders of magnitude speedups.

HEREDITARY LOAD IN MEAT AND EGG CHICKENS OF VARIOUS ORIGIN

The aim of the study was to determine the spectrum and frequency of hereditary genetic defects of embryo development in meat and egg hens of different genetic origin, obtained during an experiment to study the effectiveness of crossing roosters of imported meat crosses with meat and egg females of domestic selection. All embryos that died during egg incubation were carefully examined to detect morphopathology. The main research method is pathomorphological analysis of genetic defects of dead embryos. The spectrum and frequency of morphological and anatomical hereditary defects of embryos were established during pathological and anatomical examination of incubation waste. During the autopsy of dead embryos, morphological disorders in the structure of the skeleton were determined, as well as various disproportions of its individual parts.
 In the meat-egg hens of the original maternal form during three years of monitoring with the highest frequency (2.11–6.06%) revealed such a defect in embryo development as "donald duck". This indicates that the bird is "contaminated" with a "harmful gene" that causes this abnormality. It is worth noting that in groups of chickens that were artificially inseminated with the semen of roosters of meat crosses, this hereditary defect was also recorded. In the descendants of the first and second generations, this anomaly was absent, possibly due to its transition to a heterozygous state. Thus, we can conclude that the allele that expresses this lethal has in its genotype a bird of the local subpopulation "K". Only in meat and egg hens of domestic selection a wide variety of double anomalies was found – four species with a frequency of 1.05–1.12%. Most likely, double abnormalities are the result of homozygosity of dead embryos simultaneously for two lethal genes. Whereas, in chickens of other studied groups, only single anomalies of dead embryos were found, among them "exencephaly", "crossed beak", "shortened beak". Genetic defect of embryos "exencephaly" was found in chickens of most groups. It is obvious that the gene that determines the manifestation of this anomaly is significantly distributed in meat and egg hens of domestic selection and roosters used crosses. And in chickens of the created synthetic population "K-5" only this hereditary defect is revealed. In meat and egg hens of subpopulation "K" (F10) of the original maternal form and in the group of females, which were inseminated with polysperm of meat cocks of the cross "Cobb-500", with a low frequency (0.78–1.12%) a single "crossed beak" anomaly was found among the frozen embryos. This anomaly was not detected in the offspring of the first generation and in groups of chickens inseminated with semen of overripe brood stock. And only in the "cobb" bird of the "K-51" group among the hereditary defects of the dead embryos was this anomaly identified. In meat-egg hens of F10 subpopulation "K" with a low frequency (1.12%) was found a hereditary defect of embryos "shortened beak". Then this anomaly was found only in the "cobb" bird of both the first and second generation. The data show that the hereditary factor that determines this defect, the descendants inherited from the cocks of the cross "Cobb-500". The level of genetic burden in meat and egg hens of the local subpopulation "K", depending on the generation, ranged from 5.26 to 8.99% and was wavy in nature. In the descendants of the first generation it is low – 1.64–4.17%, in the "ross" crossbreeds is much smaller compared to the "cobb". The descendants of the second generation of different origins also had a low hereditary burden – in the range of 4.35–5.88%, which does not exceed the maximum allowable species value of 8.0%.
 The results of the research showed that crossing meat-egg hens of domestic selection with roosters of imported meat crosses "Cobb-500" and "Ross-308" did not lead to significant "contamination" of genotypes of F1-F2 offspring with lethal genes. This indicates a low share in the heredity of the used offspring of hidden carriers of "defective" genes. This may be due to the high selection of birds of foreign crosses, which does not contribute to the accumulation in its gene pool of lethal genes that cause embryonic developmental abnormalities. In chickens of the created synthetic population "K-5" the level of genetic burden is low (3.03%), which indicates the relative "purity" of this bird from lethal genes.

Connection Between Indications of Binocular Single Vision Anomalies and Punctum Proximum Among University Students of India

Purpose: To evaluate the connection between indications of binocular single vision anomalies and punctum proximum among university students of India. Method: This was a cross-sectional prospective study of 90 randomly selected subjects age ranging from18-26 years. The subjects were paramedical students studying in A.J Institute of medical science, Mangalore. After going through a detailed ophthalmic examination including standard binocular vision examination if the subject met the inclusion criteria, according to CISS questionnaire students were divided into two groups (asymptomatic and symptomatic).NPC was measured with RAF rule asking the subject to focus on a single vertical line as a target on the rule. After data collection, data were analyzed in an excel spreadsheet using, paired t-test. Result: NPC break and recovery value are more in symptomatic (9.479±0.3653 and 10.845±0.4865) subjects than asymptomatic subjects (6.033±1.3966 and 7.598±1.3921), with a highly significant difference between the two groups. Conclusion: Strong relationship was found between anomalies binocular single vision symptoms and punctum proximum (PP). The assurance of PP is useful to separate the indicative of symptomatic subjects from the asymptomatic. PP was more in symptomatic groups when compared to asymptomatic groups.

First report of polymorphisms in MTRR, GATA4, VEGF, and ISL1 genes in Pakistani children with isolated ventricular septal defects (VSD)

BackgroundVentricular septal defects (VSDs) are malformations in the septum separating the heart’s ventricles. VSDs may present as a single anomaly (isolated/nonsyndromic VSD) or as part of a group of phenotypes (syndromic VSD). The exact location of the defect is crucial in linking the defect to the underlying genetic cause. The number of children visiting cardiac surgery units is constantly increasing. However, there are no representative data available on the genetics of VSDs in Pakistani children.MethodsTwo hundred forty-two subjects (121 VSD children and 121 healthy controls) were recruited from pediatric cardiac units of Lahore. The clinical and demographic data of the subjects were collected. A total of four SNPs, one each from MTRR, GATA4, VEGF, and ISL1 genes were genotyped by PCR-RFLP.ResultsThe results showed that the minor allele (T) frequency (MAFs) for the MTRR gene variant rs1532268 (c.524C > T) was 0.20 and 0.41 in the controls and the cases, respectively, with the genotype frequencies 3, 35, 62% in the controls and 12, 59 and 29% in the cases for TT, CT, CC genotypes, respectively (allelic OR: 5.73, CI: 3.82–8.61, p-value: 5.11 × 10− 7). For the GATA4 variant rs104894073 (c.886G > A), the MAF for the controls and the cases was 0.16 and 0.37, respectively, the frequencies of AA, GA and GG genotypes were 2, 28, and 70% in the controls and 5, 64 and 31% of the cases (allelic OR: 3.08, CI: 2.00–4.74, p-value: 8.36 × 10− 8). The rs699947 (c.-2578C > A) of VEGF gene showed MAF 0.36 and 0.53 for the controls and cases, respectively, with the genotype frequencies 13, 42, and 45% in the controls and 22, 15, and 63% in the cases for the AA, CA, CC (allelic OR: 2.03, CI: 1.41–2.92, p-value: 0.0001). The ISL1 gene variant rs6867206 (g.51356860 T > C), the MAFs were 0.26 and 0.31 in the controls and cases, respectively. The genotype frequencies were 48, 52, 0% in the controls and 39, 61, 0% in the cases for TT, TC, CC genotypes (allelic OR: 0.27, CI: 0.85–1.89, p-value: 0.227). The MTRR, GATA4 and VEGF variants showed association while ISL1 variant did not appear to be associated with the VSD in the recruited cohort.ConclusionThis first report in Pakistani children demonstrates that single nucleotide polymorphisms in genes encoding transcription factors, signaling molecules and structural heart genes involved in fetal heart development are associated with developmental heart defects., however further work is needed to validate the results of the current investigation.

Unsupervised and scalable subsequence anomaly detection in large data series

Subsequence anomaly (or outlier) detection in long sequences is an important problem with applications in a wide range of domains. However, the approaches that have been proposed so far in the literature have severe limitations: they either require prior domain knowledge or become cumbersome and expensive to use in situations with recurrent anomalies of the same type. In this work, we address these problems and propose NormA, a novel approach, suitable for domain-agnostic anomaly detection. NormA is based on a new data series primitive, which permits to detect anomalies based on their (dis)similarity to a model that represents normal behavior. The experimental results on several real datasets demonstrate that the proposed approach correctly identifies all single and recurrent anomalies of various types, with no prior knowledge of the characteristics of these anomalies (except for their length). Moreover, it outperforms by a large margin the current state-of-the art algorithms in terms of accuracy, while being orders of magnitude faster.

Clinicopathological and Genomic Characterization of a Simmental Calf with Generalized Bovine Juvenile Angiomatosis

Simple SummaryVascular anomalies represent a heterogeneous group of rare disorders encompassing both vascular malformations and tumors, which can be congenital or arise shortly after birth. They often pose a diagnostic challenge in human and veterinary medicine, and the referring nomenclature is equivocal. Bovine juvenile angiomatosis (BJA), a clinical condition belonging to this group of disorders, encompasses vascular malformations and tumors arising in calves. Usually, such vascular anomalies are not further investigated on a molecular genetic level, mainly because of a lack of resources and diagnostic tools, as well as the low value and short lifespan of the affected animals. Here we report the clinical, pathological, immunohistochemical, and genetic features of a Simmental calf that displayed multiple cutaneous, subcutaneous, and visceral vascular hamartomas compatible with a generalized form of BJA. Whole-genome sequencing identified six coding variants, including four heterozygous variants in the PREX1, UBE3B, PCDHGA2, and ZSWIM6 genes, which occurred only in the BJA-affected calf and were absent in the global control cohort of more than 4500 cattle. Assuming a germline mutation as etiology, one of these variants might be responsible for the vascular malformations identified in this calf.Bovine juvenile angiomatosis (BJA) comprises a group of single or multiple proliferative vascular anomalies in the skin and viscera of affected calves. The purpose of this study was to characterize the clinicopathological phenotype of a 1.5-month-old Simmental calf with multiple cutaneous, subcutaneous, and visceral vascular hamartomas, which were compatible with a generalized form of BJA, and to identify genetic cause for this phenotype by whole-genome sequencing (WGS). The calf was referred to the clinics as a result of its failure to thrive and the presence of multiple cutaneous and subcutaneous nodules, some of which bled abundantly following spontaneous rupture. Gross pathology revealed similar lesions at the inner thoracic wall, diaphragm, mediastinum, pericardium, inner abdominal wall, and mesentery. Histologically, variably sized cavities lined by a single layer of plump cells and supported by a loose stroma with occasional acute hemorrhage were observed. Determined by immunochemistry, the plump cells lining the cavities displayed a strong cytoplasmic signal for PECAM-1, von Willebrand factor, and vimentin. WGS revealed six private protein-changing variants affecting different genes present in the calf and absent in more than 4500 control genomes. Assuming a spontaneous de novo mutation event, one of the identified variants found in the PREX1, UBE3B, PCDHGA2, and ZSWIM6 genes may represent a possible candidate pathogenic variant for this rare form of vascular malformation.

1027 Diagnostic yield of exome sequencing in prenatal diagnosis

Exome sequencing (ES) is increasingly utilized for prenatal evaluation of sonographic structural anomalies. Given high costs, identifying cases most likely to yield a diagnosis is warranted. We hypothesize that the diagnostic yield of ES varies by number and type of fetal anomalies. We analyzed two cohorts of fetuses that underwent ES due to one or more structural anomalies or nonimmune hydrops. The diagnostic yield (pathogenic or likely pathogenic (P/LP) variants) was compared in fetuses with one structural anomaly, multiple structural anomalies, and effusions or nonimmune hydrops (NIHF) with or without a concurrent structural anomaly. Cases with a single structural anomaly + polyhydramnios or growth restriction (FGR) were categorized as multiple anomalies, and cases with an effusion or NIHF and a structural anomaly were categorized as hydrops. We also compared the proportion of variants of uncertain significance that were thought to potentially be associated with the phenotype across these subgroups (VUS). Chi square or Fisher’s exact test compared proportions and logistic regression generated odds ratios. The cohort included 315 pregnancies with median age 33 years (IQR 29-36), 45.1% (142/315) nulliparous, and median gestational age at diagnosis of anomaly of 22.3 weeks (IQR 20.0-25.7). In all, 19.7% (62/315) had a single structural anomaly, 25.4% (80/315) had multiple anomalies, and 54.9% (173/315) had effusions/NIHF. The overall diagnostic yield of ES was 22.9% (72/315). The yield for isolated anomalies was 16.1% (10/62), for multiple anomalies was 17.5% (14/80), and for effusions/NIHF was 27.8% (48/173) (p=0.07) (Table). The OR for diagnostic yield of ES (P/LP) for multiple anomalies vs a single anomaly was 1.31 (95% CI 0.54-3.16) and for effusions/NIHF vs a single anomaly was 1.47 (95% CI 1.03-2.11). The diagnostic yields of ES for single or multiple structural anomalies are similar, while the yield in cases of NIHF is higher. Further study is needed to identify optimal candidates for prenatal ES.View Large Image Figure ViewerDownload Hi-res image Download (PPT)

Identifying multi-metal prospect using regional soil and stream sediment geochemical data in Bidou, Nyong Series, North West of Congo Craton

The Congo Craton is a well-known mineralized Province in Africa being host to a wide variety of mineral deposits. Exploration work in the Nyong Series over the years has yielded only gold and iron prospects. Analysis of 460 regional soils and 60 stream sediment samples revealed new mineral prospects in the area. Au content ranges from below detection limit (bdl) to 216 ppb and bdl to 18.1 ppm in soil and stream sediments, respectively. Ag contents are within the range of 5 ppb to 237 ppb in soils and from 0.01 to 1.02 ppm in stream sediments. An uncommon positive correlation is noticed between Au and U in the soils while Au–Ag correlation is observed in stream sediments. La attains a maximum concentration of 123,500 ppb in soil and >1.5% in stream sediments implying a possible mineralization of La in this study area. Principal component analysis (PCA) presents loadings of high field strength elements and chalcophile elements in factors 1 and 2 of both soil and stream sediments illustrating input from the catchment geology. Factor maps turn out a single anomaly for Au–U in soil. Streams draining the Bidou prospect yield wider Au anomalies within the prospect. Surficial geochemical maps display Bidou hill as a single urge prospect with possible disseminated mineralization throughout the entire hill. Other potential minerals of interest within Bidou prospect include Ag, Cu, Th, U, and REEs.

Anomalies Detection Using Isolation in Concept-Drifting Data Streams

Detecting anomalies in streaming data is an important issue for many application domains, such as cybersecurity, natural disasters, or bank frauds. Different approaches have been designed in order to detect anomalies: statistics-based, isolation-based, clustering-based, etc. In this paper, we present a structured survey of the existing anomaly detection methods for data streams with a deep view on Isolation Forest (iForest). We first provide an implementation of Isolation Forest Anomalies detection in Stream Data (IForestASD), a variant of iForest for data streams. This implementation is built on top of scikit-multiflow (River), which is an open source machine learning framework for data streams containing a single anomaly detection algorithm in data streams, called Streaming half-space trees. We performed experiments on different real and well known data sets in order to compare the performance of our implementation of IForestASD and half-space trees. Moreover, we extended the IForestASD algorithm to handle drifting data by proposing three algorithms that involve two main well known drift detection methods: ADWIN and KSWIN. ADWIN is an adaptive sliding window algorithm for detecting change in a data stream. KSWIN is a more recent method and it refers to the Kolmogorov–Smirnov Windowing method for concept drift detection. More precisely, we extended KSWIN to be able to deal with n-dimensional data streams. We validated and compared all of the proposed methods on both real and synthetic data sets. In particular, we evaluated the F1-score, the execution time, and the memory consumption. The experiments show that our extensions have lower resource consumption than the original version of IForestASD with a similar or better detection efficiency.

Пренатальная диагностика персистирующей правой пупочной вены

Objective. To determine the incidence of persistent right umbilical vein (PRUV), as well as to study the prognostic value of as a marker of fetal developmental anomalies and neonatal outcomes. Patients and methods. 12,945 pregnant women whose gestation period was 18–41 weeks were examined. Ultrasound scanning was performed using a 3–5 MHz probe. A detailed fetal anatomy assessment was carried out, PRUV was diagnosed on the transverse scan of the fetal abdominal cavity. Results. 23 fetuses with PRUV were revealed: in 22 fetuses – an intrahepatic PRUV as a single anomaly was found, in 1 fetus – an extrahepatic PRUV (frequency 1:562). In all cases of intrahepatic PRUV, associated anomalies of fetal development were not revealed, pregnancy proceeded without complications with a favorable postnatal outcome. In one case of extrahepatic PRUV, cardiomegaly and hyper-echogenic bowel were diagnosed. Conclusion. The presence of PRUV in fetus is a marker of developmental anomalies and it warns a doctor about possible complications. Pregnancy, complicated by PRUV, requires a detailed dynamic observation, as in 30% of cases, there is a fetal growth retardation. The results of our study indicate that PRUV, when it is the only anomaly, has a favorable outcome, but based on the data of other authors, the fetus should be more carefully examined for other congenital malformations. Key words: persistent right umbilical vein, prenatal diagnosis, ultrasound diagnosis

An Efficient IDS Framework for DDoS Attacks in SDN Environment

The rapid usage of the Internet for the last few decades has lead to the deployment of high-speed networks in commercial and educational institutions. As network traffic is increasing, security challenges are also increasing in the high-speed network. Although the Intrusion Detection System (IDS) has a significant role in spotting potential attacks, the heavy traffic flow causes severe technical challenges relating to monitoring and detecting the network activities. Moreover, the devastating nature of the Distributed Denial-of-Service (DDoS) attack draws out as a significant cyber-attack regardless of the emergence of Software Defined Network (SDN) architecture. This paper proposes a novel framework to address the performance issues of IDS and the design issues of SDN about DDoS attacks by incorporating intelligence in the data layer using Data Plane Development Kit (DPDK) in the SDN architecture. This novel framework is named as DPDK based DDoS Detection (D3) framework, since DPDK provides fast packet processing and monitoring in the data plane. Moreover, the statistical anomaly detection algorithm implemented in the data plane as Virtual Network Function (VNF) using DPDK offers fast detection of DDoS attacks. The experimental results of the D3 framework guarantee both efficiency and effect of the novel IDS framework. The publicly available CIC DoS datasets also ensure the detection effect of a single statistical anomaly detection algorithm against the DDoS attack.

Successful Implantation of Two Stents in the Right Coronary Artery in a Patient with a Single Coronary Artery Anomaly

Single coronary artery is a congenital cardiac anomaly in which the right coronary artery or left main coronary artery originates from a single aortic ostium. Although this coronary anomaly is rare, it has a wide range of clinical manifestations, from being asymptomatic to sudden cardiac death, and the treatment strategies in this anomaly are not clear yet. In this report; we presented single coronary artery originating from the right sinus of valsalva, and successful stenting of the middle and distal right coronary artery in a patient with acute coronary syndrome.

Combined Surgical Treatment of Atherosclerotic Coronary Artery Disease and Moderate Aortic Valve Stenosis in Patient with Concomitant Lipton’s R-III Type of Single Coronary Artery Anomaly

A single coronary artery is a very rare condition, commonly associated with other congenital anomalies. It could be generally considered as neither benign nor malignant form of congenital coronary artery anomalies since its pathophysiological and clinical implications grossly depend on different anatomical patterns defined by the site of origin and distribution of the branches. By presenting the patient who underwent successful coronary artery bypass grafting and aortic valve replacement surgery in a presence of isolated single coronary artery, we intend to emphasize natural and procedural risks and distinguish casual from causal in this extremely rare clinical and surgical scenario.

PROGRESSION OF DIABETIC NON-PROLIFERATIVE RETINOPATHY IN TYPE 2 DIABETES MELLITUS: THE CONNECTION WITH THE BLOOD ENDOTHELIAL MONOCYTE-ACTIVATING POLYPEPTIDE-II LEVEL

Relevance. The numerous pro-inflammatory and antiangiogenic properties of endothelial monocyte-activating polypeptide-II (EMAP-II) suggest its possible role in the onset and progression of diabetic non-proliferative retinopathy (DNPR) in type 2 diabetes mellitus (T2DM).
 Objective – is to determine the blood EMAP-II in the DM2 patients and to establish its connection with the progression of DNPR.
 Material and methods. We examined 91 patients with DM2 (182 eyes), who were divided into groups: 1st – there was no DNPR in both eyes and 2nd – there was no retinopathy in one eye, and isolated vascular anomalies were noted in the other (ETDRS level 14, 15). The control group included 25 patients of the corresponding age and gender. The patients were re-examined after 1 year. The level of EMAP-II was determined by the enzyme immunoassay in blood plasma once at the beginning of the study. Statistical packages MedStat and MedCalc v.15.1 (MedCalc Software bvba) were used for statistical research.
 Results. The analysis of clinical and laboratory parameters showed that the initial manifestations of diabetic retinal lesions were manifested in 27.5% of patients after 7.16±1.11 years and were accompanied by greater glycemia. The level of EMAP-II in DM2 was many times higher than in the control, which depended on the presence of diabetic vascular changes in the retina: in patients without changes in the retina (group 1) – by 3.7 times, and in patients with initial vascular anomalies (group 2) – 5.2 times (p<0.001). The level of EMAP-II at the beginning of the study was associated with the progression of diabetic changes in the retina after 1 year – with their presence, it was 1.5 times higher than without them (p<0.001). Stratification by stage of DNPR after 1 year also showed the dependence of the severity of diabetic changes in the retina on the initial level of EMAP-II: in the presence of single vascular anomalies and initial DNPR, it was increased by 3-4 times, while with moderate DNPR – 5.9 times (p<0.001 for all comparisons).
 Conclusion. Thus, a significant increase in the level of EMAP-II in T2DM was established, and the dependence of the initial diabetic changes in the retina and the degree of their progression in 1 year after the increasing of the blood EMAP-II level.

A multidisciplinary approach to management of abdominal lymphatic malformations

Background/purposeAbdominal lymphatic malformations (LM) are a rare subset of vascular anomaly caused by abnormal development of the lymphatic system. They are classified as macrocystic, microcystic or combination macrocystic and microcystic. Surgical resection, percutaneous sclerotherapy, and medical therapy are all employed to treat these complex and often symptomatic lesions. No standardized treatment algorithm exists currently. The purpose of this study was to establish a multidisciplinary treatment approach to abdominal LMs. MethodsA retrospective observational study was conducted from 2013 to 2019 on patients with abdominal LMs at a single tertiary children's hospital vascular anomalies center. Demographics, imaging, and treatment modality were recorded. Clinical and/or radiographic response to the primary treatment modality as well as complications was the outcomes measured. ResultsNineteen patients (12 macrocystic, 5 microcystic and 2 combined) were identified, with a median age at diagnosis of 2.2 years (range 0.1–20.8 years). Sclerotherapy was the most common primary treatment, followed by surgical resection and sirolimus. No difference in clinical response (p = 0.58) or complications (p = 0.31) was observed based on primary treatment or subtypes. ConclusionsBased on our institutional experience, we propose an LM subtype-based treatment algorithm for abdominal LMs. It employs a multidisciplinary approach, and results in satisfactory patient outcomes with minimal complications. Level of evidenceLevel III, retrospective comparative study.

Combining SCADA and vibration data into a single anomaly detection model to predict wind turbine component failure

AbstractReducing downtime through predictive or condition‐based maintenance is a promising strategy to help reduce costs associated with wind farm operation and maintenance. To help effectively monitor wind turbine condition, operators now rely on multiply sources of data to make informed operational decisions which can minimise downtime, increasing availability and profitability of any given site. Two of such approaches are SCADA temperature and vibration monitoring, which are typically performed in isolation and compared over time for both fault diagnostics and reliability analysis. Presenting two separate case studies, this paper describes a methodology to bring multiple data sources together to diagnose faults by using a single‐class support vector machine classifier to assess normal behaviour model error, with results showing that anomalies can be detected more consistently when compared to more standard approaches of analysing each data source in isolation.

Anomaly Detection in Automated Vehicles Using Multistage Attention-Based Convolutional Neural Network

Connected and Automated Vehicles (CAVs), owing to their characteristics such as seamless and real-time transfer of data, are imperative infrastructural advancements to realize the emerging smart world. The sensor-generated data are, however, vulnerable to anomalies caused due to faults, errors, and/or cyberattacks, which may cause accidents resulting in fatal casualties. To help in avoiding such situations by timely detecting anomalies, this study proposes an anomaly detection method that incorporates a combination of a multi-stage attention mechanism with a Long Short-Term Memory (LSTM)-based Convolutional Neural Network (CNN), namely, MSALSTM-CNN. The data streams, in the proposed method, are converted into vectors and then processed for anomaly detection. We also designed a method, namely, weight-adjusted fine-tuned ensemble: WAVED, which works on the principle of average predicted probability of multiple classifiers to detect anomalies in CAVs and benchmark the performance of the MSALSTM-CNN method. The MSALSTM-CNN method effectively enhances the anomaly detection rate in both low and high magnitude cases of anomalous instances in the dataset with the gain of up to 2.54% in F-score for detecting different single anomaly types. The method achieves the gain of up to 3.24% in F-score in the case of detecting mixed anomaly types. The experiment results show that the MSALSTM-CNN method achieves promising performance gain for both single and mixed multi-source anomaly types as compared to the state-of-the-art and benchmark methods.

Abstract P175: A Twitter Analysis Of Hypertension And Allergies: Discordance Between Social Media Discussion And Disease Morbidity

Background: Social media data may be used to shed light on how engaged the general public is in conversations surrounding elevated blood pressure (BP) as an important health problem. Methods: We gathered Twitter data from 1/1/2016 to 11/23/2019 using the ‘TweetScraper’ program (https://github.com/jonbakerfish/TweetScraper). We scraped tweets with content related to BP and, to generate comparator data, we similarly collected tweets related to allergies during the same time period. We then plotted the number of tweets over time (in days) for each health topic and compared the between-group frequency of tweets over time. Results: We observed that the frequency of tweets related to BP was consistently lower than that related to allergy, notwithstanding lower rates of allergy related tweets during the non-peak allergy seasons ( Figure ). The number of tweets related to BP never exceeded those related to allergy, per month, but there were single day anomalies: BP tweets exceeded allergy tweets by 78% on 11/13/2017 and by 96% on 11/14/2027, following the release of updated hypertension guidelines. Conclusions: We demonstrate both the feasibility and utility of using social media data to assess the general public’s interest level in certain health topics over time. We found persistently lower levels of engagement around elevated BP (asymptomatic but conferring high mortality risk) compared to allergies (symptomatic but conferring low mortality risk) despite their similarly wide prevalence. As popular social media methods of communication evolve, so will the challenges and opportunities related to harnessing these data as part of efforts to improve public health.

Preliminary studies on deep-penetrating geochemical methods in exploration for concealed volcanic-type uranium deposit

Volcanic-type uranium deposit is one of the most important type of uranium resources in China. It has been proved by the scientific deep drilling work accomplished in 2015 that uranium mineralization can be detected at the depth of more than one thousand meters, where there are no significant radiometric or other exploration geochemical anomalies that can be acquired on ground surface. With the progress of exploration for uranium resources in China, deposits which are exposed on surface or shallow-buried are all almost discovered already. It has become a challenging task to explore the deep-buried deposits for uranium geologists. Application experiments are carried out in Xiangshan uranium deposit field using two kinds of deep-penetrating geochemical methods, including the Selective Leaching of Mobile Metals(MOMEO) and Electro-geochemistry techniques(CHIM).The U ore bodies are manifested as a clear CHIM anomaly of U. The adsorbed and exchangeable forms of U obtained with the MOMEO technique is demonstrated as lower single peak anomalies compared with the CHIM technique. The ‘total’ U element content shows no clear anomaly over the uranium mineralization, while there is only a single point high value with an intensity up to 12.5 ppm of U. The polonium-210 shows no significant anomalies over the uranium mineralization.According to the results obtained in this paper, two kinds of deep-penetrating geochemical signatures demonstrate more effective role in detecting the concealed uranium mineralization than traditional methods. The deep-penetrating geochemical methods are bound to play an important role in exploring for deep-buried and concealed uranium deposits in China.