4,354 research outputs found
ENAS-B: Combining ENAS with Bayesian Optimisation for Automatic Design of Optimal CNN Architectures for Breast Lesion Classification from Ultrasound Images
No full textEfficient Neural Architecture Search (ENAS) is a recent development in searching for optimal cell structures for Convolutional Neural Network (CNN) design. It has been successfully used in various applications including ultrasound image classification for breast lesions. However, the existing ENAS approach only optimises cell structures rather than the whole CNN architecture nor its trainable hyperparameters. This paper presents a novel framework for automatic design of CNN architectures by combining strengths of ENAS and Bayesian Optimisation in two folds. Firstly, we use ENAS to search for optimal normal and reduction cells. Secondly, with the optimal cells and a suitable hyperparameter search space, we adopt Bayesian Optimisation to find the optimal depth of the network and optimal configuration of the trainable hyperparameters. To test the validity of the proposed framework, a dataset of 1,522 breast lesion ultrasound images is used for the searching and modelling. We then evaluate the robustness of the proposed approach by testing the optimized CNN model on three external datasets consisting of 727 benign and 506 malignant lesion images. We further compare the CNN model with the default ENAS-based CNN model, and then with CNN models based on the state-of-the-art architectures. The results (error rate of no more than 20.6% on internal tests and 17.3% on average of external tests) showed that the proposed framework generates robust and light CNN models
ENAS: A new software for spike train analysis and simulation
No full textInternational audienceAs one gains more intuitions and results on the importance of concerted activity in spike trains, models are developed to extract potential canonical principles underlying spike coding. These methods shed a new light on spike train dynamics. However, they require time and expertise to be implemented efficiently, making them hard to use in a daily basis by neuroscientists or modelers. To bridge this gap, we developed the license free multiplatform software ENAS integrating tools for spike trains analysis and simulation. These tools are accessible through a friendly Graphical User Interface that avoids any scripting or writing code from user. Most of them have been implemented to run in parallel to reduce the time and memory consumption. One of the main strength of ENAS when compared to competing software is to provide statistical analysis with Maximum Entropy-Gibbs distributions taking into account both spatial and temporal correlations as constraints, allowing to introduce causality and memory in statistics. Conversely, given this analysis or other known statistics, ENAS one can also generate new spike trains. These methods result from a series of work and they have already been applied to the analysis of retina data. This comes in addition to basic visualizations and classical analysis for statistics of spike trains analysis. All these tools are generic and can be applied to any spike train. Interestingly, ENAS also includes specific tools dedicated for vision, and the retina in particular. For example, one can jointly visualize stimulus and spiking activity or estimate receptive fields. ENAS also includes a virtual retina simulator extending former Virtual Retina simulator to include lateral connections in the IPL. We hope that ENAS will become a useful tool for neuroscientists to analyse spike trains and we hope to improve it thanks to user feedback. Our goal is to progressively enrich ENAS with the latest research results, in order to facilitate transfer of new methods to the community. It is downloadable from https://enas.inria.fr where documentation, tutorials and samples of spike trains are available
Labcode 9B - HIV neutralization Protocol automated Plaque Reduction Assay with GHOST cells
No full textThis document contains a published paper with a detailed protocol on a high-throughput, high-content automated plaque reduction (APR) assay based on
automated microscopy and image analysis that allows evaluation of neutralization and inhibition of cell-cell fusion within the same assay. Neutralization of virus particles is measured as a reduction in the number of fluorescent plaques, and inhibition of cell-cell fusion as a reduction in plaque area.
PMID: 2517603
Sequence of occurrence of individual ENAS (%).
No full text<p>LO—lobed; BL—blebbed; NO—notched; CO—condensed; BU—bud; VA—vacuolated and MN—micronuclei. Values below each ENA indicate its frequency. Analysis included 14,000 cells <i>per</i> treatment.</p><p>Sequence of occurrence of individual ENAS (%).</p
Thomas Rotch accounts payable, Kendal 1818-1823
No full textJohn Benn, Enas Moore, Jacob Feters, Bartly Williams submit their bill to Thomas Rotch for 516 pounds of bar iron, $41.28. 7.75" x 7.1
Replication Data for: Mental Health Burden and Predictors among Egyptian Healthcare Workers during the COVID-19 Pandemic
No full textthis is the questionnaire used for collection of dat
Labcode 9B - HIV neutralization Protocol automated Plaque Reduction Assay with GHOST cells
No full textThis document contains a published paper with a detailed protocol on a high-throughput, high-content automated plaque reduction (APR) assay based on
automated microscopy and image analysis that allows evaluation of neutralization and inhibition of cell-cell fusion within the same assay. Neutralization of virus particles is measured as a reduction in the number of fluorescent plaques, and inhibition of cell-cell fusion as a reduction in plaque area.
PMID: 2517603
Replication Data for: Mental Health Burden and Predictors among Egyptian Healthcare Workers during the COVID-19 Pandemic
No full textspss file for data collected for original articl
The role of experienced practitioners in promoting Saudi female pre-service teachers’ cognitive and social engagement in learning to teach with technology through participation in an online professional learning community
No full textTeacher education programmes are increasingly focused on engaging pre-service teachers in the process of learning to teach with technology. Learning to teach with technology is considered a complex process that requires pre-service teachers to engage both cognitively and socially. One of the possible approaches that can enhance pre-service teachers’ engagement in the learning to teach is by engaging pre-service teachers in an online professional learning community (PLC) with experienced practitioners who are expert in-service teachers in the field of technology use in education. Therefore, this study seeks to explore the role of experienced practitioners in promoting Saudi female pre-service teachers’ cognitive and social engagement in learning to teach with technology through participation in an online PLC. This study adopted a mixed method and experimental approach. Two practitioners and 76 female pre-service teachers participated in the study. The pre-service teachers were assigned to one of two groups, with 38 pre-service teachers in each group. The practitioners only participated in the experimental group. The study was divided into three phases. The first phase investigated the role of practitioners in developing pre-service teachers’ technological pedagogical content knowledge (TPACK). The results revealed that the experimental group had significantly higher scores in technological pedagogical knowledge (TPK) and TPACK than the control group. The interview findings show that practitioners increased pre-service teachers’ awareness of up-to-date technology, technologies designed for specific subjects, the pedagogical affordance of technology, and factors that need to be considered during the use of technology. The second phase examined the role of practitioners in promoting pre-service teachers’ knowledge construction and social interaction by employing content analysis (CA), social network analysis (SNA), and a TPACK questionnaire. The practitioners increased pre-service teachers’ levels of knowledge construction, high-cognitive discourse as well as network size and the extent of collaboration between those with varying levels of TPACK knowledge in both knowledge sharing and building networks. In both groups, collaboration in the knowledge building network predominantly occurred with peers in the same discipline. The third phase explored both pre-service teachers and practitioners’ experience regarding their participation in an online PLC via interviews. The interview findings revealed that practitioners enhanced the quality of the discussion as well as enhancing the depth of the discussion. Another significant finding was that the practitioners enhanced the social interaction among pre-service teachers by strengthening their self-confidence, deepening the ethics of collaborative work within an Islamic cultural perspective, strengthening the sense of community, making interdisciplinary activities, linking pre-service teachers with each other, and encouraging them to use their external support as a resource. The practitioners found that this experience enhanced their professional growth. This research provides suggestions for enhancing pre-service teachers’ engagement with practitioners in an online PLC
- …
