ECG Biometrics Using Deep Learning and Relative Score Threshold Classification

The field of biometrics is a pattern recognition problem, where the individual traits are coded, registered, and compared with other database records. Due to the difficulties in reproducing Electrocardiograms (ECG), their usage has been emerging in the biometric field for more secure applications. Inspired by the high performance shown by Deep Neural Networks (DNN) and to mitigate the intra-variability challenges displayed by the ECG of each individual, this work proposes two architectures to improve current results in both identification (finding the registered person from a sample) and authentication (prove that the person is whom it claims) processes: Temporal Convolutional Neural Network (TCNN) and Recurrent Neural Network (RNN). Each architecture produces a similarity score, based on the prediction error of the former and the logits given by the last, and fed to the same classifier, the Relative Score Threshold Classifier (RSTC).The robustness and applicability of these architectures were trained and tested on public databases used by literature in this context: Fantasia, MIT-BIH, and CYBHi databases. Results show that overall the TCNN outperforms the RNN achieving almost 100%, 96%, and 90% accuracy, respectively, for identification and 0.0%, 0.1%, and 2.2% equal error rate (EER) for authentication processes. When comparing to previous work, both architectures reached results beyond the state-of-the-art. Nevertheless, the improvement of these techniques, such as enriching training with extra varied data and transfer learning, may provide more robust systems with a reduced time required for validation.

Symptom and Disability Measurement by Common Foot and Ankle–Specific Outcome Rating Scales

Background: Well-designed foot and ankle clinical outcomes research requires region-specific subjective outcome measures. Many foot and ankle–specific instruments are now available. Determining which instruments to choose is daunting. We present a patient survey to determine the most relevant questions pertaining to them. Methods: Sixteen foot and ankle–specific outcome instruments were chosen based on popularity, emphasizing valid, reliable, responsive scores. Subjective portions were consolidated and given to 109 patients with osteochondral lesion of the talus (OLT) (39), ankle instability (35), and ankle arthritis (35). Outcome instruments were measured according to 4 criteria: number of questions endorsed by 51%, number with high mean importance, number with low mean importance, and number with the highest-ranking frequency importance product. Instruments were then ranked based on relative score, with the highest scores indicating the instrument was the most useful for that sample. Results: The Foot and Ankle Outcome Score (FAOS) achieved the highest score in all 4 categories for OLT, followed by Foot and Ankle Ability Measure (FAAM) and American Academy of Orthopaedic Surgeons (AAOS) Foot and Ankle Score. The FAOS achieved the highest score in all 4 categories for ankle instability, followed by FAAM and AAOS. For osteoarthritis, the FAOS achieved the highest relative score followed by FAAM and AAOS. The AOF, Ankle Osteoarthritis Score, and AAS are instruments commonly used that had lower relative scores. Conclusion: The FAOS, FAAM, and AAOS Foot and Ankle Score contain several items important to patients with osteochondral lesions of the talus, ankle instability, and ankle osteoarthritis. Level of Evidence: Level II, prospective comparative study.

Fine-Grained Named Entity Typing over Distantly Supervised Data Based on Refined Representations

Fine-Grained Named Entity Typing (FG-NET) is a key component in Natural Language Processing (NLP). It aims at classifying an entity mention into a wide range of entity types. Due to a large number of entity types, distant supervision is used to collect training data for this task, which noisily assigns type labels to entity mentions irrespective of the context. In order to alleviate the noisy labels, existing approaches on FG-NET analyze the entity mentions entirely independent of each other and assign type labels solely based on mention's sentence-specific context. This is inadequate for highly overlapping and/or noisy type labels as it hinders information passing across sentence boundaries. For this, we propose an edge-weighted attentive graph convolution network that refines the noisy mention representations by attending over corpus-level contextual clues prior to the end classification. Experimental evaluation shows that the proposed model outperforms the existing research by a relative score of upto 10.2% and 8.3% for macro-f1 and micro-f1 respectively.

Adolescent friendly health services: where are we actually standing?

Background: Services for adolescents are highly fragmented, poorly coordinated and uneven in quality. Adolescent-friendly services (AFHS) should be able to attract young people, meet their needs comfortably and with sensitivity, and retain young clients for continuing care. With this background, the present study was carried out to study the implementation status of Adolescent friendly health services and to determine compliance with quality standards.Methods: Present cross sectional study was carried out at 10 Urban Health Centers of Ahmedabad Municipal Corporation. Study population comprised of adolescents (10 to 19 years). All adolescents aged 10 to 19 years (n=99) and all health care providers present at the time of visit at the facility were interviewed. Data was collected using WHO quality measurement tools. Data was analyzed using WHO scoring sheet for data analysis.Results: Relative score for confidentiality, privacy, equitability, accessibility and knowledge gap in adolescent is 58.3%. Relative score for health care providers’ competencies to work with adolescents and to provide them with required services is 42.6%. Relative score for Observation tool used for facility inventory is 45.6%. Overall relative score is 47.3% of maximum possible score.Conclusions:As the score is near to the lower limit of class interval of 40% to 80%, it implies need of improvement.

No apparent influence of psychometrically-defined schizotypy on orientation-dependent contextual modulation of visual contrast detection

We investigated the relationship between psychometrically-defined schizotypy and the ability to detect a visual target pattern. Target detection is typically impaired by a surrounding pattern (context) with an orientation that is parallel to the target, relative to a surrounding pattern with an orientation that is orthogonal to the target (orientation-dependent contextual modulation). Based on reports that this effect is reduced in those with schizophrenia, we hypothesised that there would be a negative relationship between the relative score on psychometrically-defined schizotypy and the relative effect of orientation-dependent contextual modulation. We measured visual contrast detection thresholds and scores on the Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE) from a non-clinical sample (N= 100). Contrary to our hypothesis, we find an absence of a monotonic relationship between the relative magnitude of orientation-dependent contextual modulation of visual contrast detection and the relative score on any of the subscales of the O-LIFE. The apparent difference of this result with previous reports on those with schizophrenia suggests that orientation-dependent contextual modulation may be an informative condition in which schizophrenia and psychometrically-defined schizotypy are dissociated. However, further research is also required to clarify the strength of orientation-dependent contextual modulation in those with schizophrenia.

Patterns of Change in Psychological Variables Leading up to Competition in Superior Versus Inferior Performers

The study explored patterns of change in a number of potentially performance-related variables (i.e., fatigue, social support, self-efficacy, autonomous motivation, mental skills) during the lead-up to a competitive triathlon, and whether these patterns of change differed for relatively superior versus inferior performers. Forty-two triathletes completed an inventory measuring the study variables every other day during a 2-week period leading up to competition. Performance was assessed using participants’ race time, and using a self-referenced relative score compared with personal best times. Multilevel growth curve analyses revealed significant differences in growth trajectories over the 2-week period in mental skills use, social support, and fatigue. The results provide novel insight into how athletes’ fluctuating psychological state in the 2 weeks before competition may be crucial in determining performance.

Analysis haemostatic system gene polymorphism in pregnant women without complications from Russia and Ukraine

Polymorphism of F5 1691G>A, F2 20210G>A, FGB –455G>A, ITGB3 1565Т>С, PAI1 –675 5G>4G, MTHFR 677C>Т genes in pregnant women from Russia and Ukraine was studied by biochip methods. No differences in distribution of F5, F2 and ITGβ3 gene polymorphism were detected. Higher rates of –455G/A FGB and –675 5G/4G PAI1 genotypes in ukrainians compared to pregnant women from Russia were found. Also variable distribution of MTHFR gene polymorphism in women from different countries was registered. The complex approach based on the calculation of relative “score” as a sum of relevant genetic polymorphisms has detected somewhat elevated risk of trombophilia for pregnant women from Ukraine compared to this one from Russia.

Correlation of Early Metastatic Response by123I-Metaiodobenzylguanidine Scintigraphy With Overall Response and Event-Free Survival in Stage IV Neuroblastoma

Purpose: Metaiodobenzylguanidine (MIBG), specifically taken up in cells of sympathetic origin, provides a highly sensitive and specific indicator for the detection of metastases in neuroblastoma. The aim of this study was to correlate early response to therapy by MIBG scan, using a semiquantitative scoring method, with the end induction response and event-free survival (EFS) rate in stage IV neuroblastoma. Patients and Methods: Seventy-five children older than 1 year and with stage IV neuroblastoma had 123I-MIBG scans at diagnosis, after two and four cycles of induction therapy, and before autologous stem-cell transplantation. The scans were read by two independent observers (concordance > 95%) using a semiquantitative method. Absolute and relative (score divided by initial score) MIBG scores were then correlated with overall pretransplantation response, bone marrow response, and EFS. Results: The pretransplantation response rate was 81%, and the 3-year EFS rate was 32%, similar to a concomitant group of 375 stage IV patients. The median relative MIBG scores after two, four, and six cycles were 0.5, 0.24, and 0.12, respectively. The probability of having a complete response or very good partial response before transplantation was significantly higher if the relative score after two cycles was ≤ 0.5, or, if after four cycles, the relative score was ≤ 0.24. Patients with a relative score of ≤ 0.5 after two cycles or a score of ≤ 0.24 after four cycles had an improved EFS rate (P = .053 and .045, respectively). Conclusion: Semiquantitative MIBG score early in therapy provides valuable prognostic information for overall response and EFS, which may be useful in tailoring treatment.