computer aided analysis
Recently Published Documents


TOTAL DOCUMENTS

864
(FIVE YEARS 71)

H-INDEX

38
(FIVE YEARS 2)

Author(s):  
Andreas Leibetseder ◽  
Klaus Schoeffmann ◽  
Jörg Keckstein ◽  
Simon Keckstein

AbstractEndometriosis is a common gynecologic condition typically treated via laparoscopic surgery. Its visual versatility makes it hard to identify for non-specialized physicians and challenging to classify or localize via computer-aided analysis. In this work, we take a first step in the direction of localized endometriosis recognition in laparoscopic gynecology videos using region-based deep neural networks Faster R-CNN and Mask R-CNN. We in particular use and further develop publicly available data for transfer learning deep detection models according to distinctive visual lesion characteristics. Subsequently, we evaluate the performance impact of different data augmentation techniques, including selected geometrical and visual transformations, specular reflection removal as well as region tracking across video frames. Finally, particular attention is given to creating reasonable data segmentation for training, validation and testing. The best performing result surprisingly is achieved by randomly applying simple cropping combined with rotation, resulting in a mean average segmentation precision of 32.4% at 50-95% intersection over union overlap (64.2% for 50% overlap).


2021 ◽  
Vol 25 (1) ◽  
pp. 31-38
Author(s):  
E. A. Levkova ◽  
R. I. Sepiashvili ◽  
S. Z. Savin

Relevance. The article is devoted to creating prognostic models based on epidemiological and immunological data. Objective: to study the comparative dynamic epidemiological and immunological characteristics of patients with COVID-19. Materials and methods. Methodological approaches to the use of system analysis of epidemiological and immunological characteristics of patients with COVID-19 using multivariate analysis are described. The used technologies of computer-aided analysis systems, algorithms for recognizing, measuring and identifying the condition of patients, and methods of statistical data processing made it possible to create a universal information predictive model for calculating the dynamics of infectious diseases prone to generalization (pandemics), as well as to understand in which groups these new infectious diseases are most dangerous. Results and discussion. Using the methods of system analysis, the epidemiological and immunological aspects of predictive models of the coronavirus pandemic were evaluated using the most objective international data, which increased the information content of the analysis. Conclusions . Creating predictive epidemiological and immunological models of the pandemic is an urgent and promising task to combat the medical and social consequences of the spread of coronavirus infection in Russia.


2021 ◽  
Vol 10 (20) ◽  
pp. 4651
Author(s):  
Keooudone Thammavong ◽  
Suchaya Luewan ◽  
Theera Tongsong

Objective: To determine the performance of fetal cardiac volume (CV) in the detection of fetal Hb Bart’s disease among fetuses at risk at 18–22 weeks of gestation and to compare the performance with those of cardiothoracic diameter ratio (CTR) and middle cerebral artery peak systolic velocity (MCA-PSV). Methods: Fetuses at risk of Hb Bart’s disease between 18 and 22 weeks of gestation prospectively underwent echocardiography with acquisition of the volume datasets (VDS) of fetal heart, using 4D-cardiac STIC. Subsequently, off-line analysis was blindly performed to measure cardiac volume using the VOCAL technique. Results: A total of 502 fetuses at risk meeting the inclusion criteria were included in the analysis, consisting of 117 (23.3%) fetuses with Hb Bart’s disease and 385 (76.7%) unaffected fetuses. The mean (±SD) gestational age at the time of ultrasound examination was 19.70 ± 1.3 weeks. In predicting fetal Hb Bart’s disease, CV, using a cut-off Z-score of 1.7, had a sensitivity of 94.9% and specificity of 94.0%. The performance of CV was slightly better than that of CTR but very superior to that of MCA-PSV (areas under curve: 0.988, 0.974 and 0.862, respectively). Conclusions: Fetal CV has a very high performance in predicting fetal Hb Bart’s disease at mid-pregnancy, comparable with CTR and much better than MCA-PSV.


2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Tadeusz Zabolewicz ◽  
Paulina Puckowska ◽  
Paweł Brym ◽  
Kamil Oleński ◽  
Stanisław Kamiński

Abstract Bovine peptidoglycan recognition protein 1 (PGLYRP1) is an important receptor that binds to murein peptidoglycans (PGN) of Gram-positive and Gram-negative bacteria and is, therefore, involved in innate immunity. The SNP T>C rs68268284 located in the 1st exon of the PGLYRP1 gene was identified by the PCR-RFLP method in a population of 319 Holstein cows. Somatic cell count (SCC) was measured 7–10 times in each of three completed lactations to investigate whether the PGLYRP1 polymorphism is associated with SCC. Using the GLM model, it was found that cows with the TT genotype showed significantly lower somatic cell counts than those with the CC genotype during the first lactation (P = 0.023). Moreover, during lactations 1–2 and 1–3, cows with the TT genotype reveal significantly lower SCC than CT heterozygotes, at P = 0.025 and P = 0.006, respectively. Computer-aided analysis showed that rs68268284 polymorphism could modify the PGLYRP1 functions because the mutated residue is located in a domain that is important for the binding of other molecules.


Author(s):  
Siu-Cheung Kong

AbstractThis study aimed at proposing an e-Learning framework in school education. The proposed framework consisted of technology and pedagogy dimensions. The method of computer-aided analysis of learners’ reflection text was used to evaluate the pedagogical delivery of the proposed framework. A study involving 33 in-service teachers in a teacher development course was conducted to investigate the effectiveness of the pedagogical delivery. The computer-aided analysis of learners’ reflection text on “what is e-Learning” before and after the teaching of the course showed that learners deepened their understanding about the technology use and the importance of pedagogy in e-Learning, to consider more the component on theories/models/principles/strategies in pedagogical design and practices, emphasize more the reflection and discussion in learning and teaching activities, and also realize more the possibility of pedagogical decision-making using data collected from online learning. The questionnaire survey and focus group interview with learners also indicated that the pedagogical delivery of the e-Learning framework with the support from the computer-aided analysis of learners’ reflection text coupled with hierarchical visualization of analysis results was positively perceived. The research findings contribute pioneering insights into the use of a computer-aided approach for an accurate and efficient evaluation in teacher development courses on e-Learning.


Sign in / Sign up

Export Citation Format

Share Document