194 research outputs found
Wav2Vec2 model trained on TIMIT train set, following Baevski et al. (2020)
No full textThis is a wav2vec2 automatic speech recognition model, fine-tuned on train set of TIMIT dataset, used as a part of the pipeline for extracting average vowel entropy (AVE, Isaev et al. (2023), "Uncertainty of Vowel Predictions as a Digital Biomarker for Ataxic Dysarthria", The Cerebellum, in press
LASERCOOLED RaF AS A LABORATORY FOR TESTING FUNDAMENTAL SYMMETRIES
No full textAuthor Institution: Clemens-Schopf Institute, Technical University of Darmstadt, 64287 Darmstadt, GermanyModern spectroscopical methods allow to reach an accuracy in measurements of molecular spectra that is sufficient for testing symmetry violation in fundamental weak interactions. As was demonstrated in recent experimental search for a permanent electric dipole moment of the electron in YbF [1] and PbO [2], diatomic molecules with an open electronic shell present extremely powerful instrument to study violations of space parity (-odd) and space parity and time reversal (-odd) symmetries. In this talk we discuss the enhancement of -odd and -odd effects in diatomic molecules and consider some prospective candidates for carrying out successful measurements. Radium monofluoride is shown to be a very prospective molecular candidate for observing possible symmetry violations due to its predicted suitability to direct laser cooling [3], large enhancement factors (both electronic and nuclear) [4] and efficient (for the given class of molecules) production route. \\~ REFERENCES: \begin{enumerate} \item J. J. Hudson, D. M. Kara, I. J. Smallman, B. E. Sauer, M. R. Tarbutt and E. A. Hinds, Nature, 473 (2011) \item S. Eckel, P. Hamilton, E. Kirilov, H.W. Smith and D. DeMille, http://arxiv.org/abs/1303.3075v1 \item T. A. Isaev, S. Hoekstra and R. Berger, Phys.Rev.A, 82, 5, 2010 \item T. A. Isaev and R. Berger, http://arxiv.org/abs/1302.5682 \end{enumerate
Use of Machine Learning and Computer Vision Methods for Building Behavioral and Electrophysiological Biomarkers for Brain Disorders
No full textResearch on biomarkers of brain disorders is an actively developing area. Biomarkers may allow for the early detection of diseases, which is essential for early intervention and improved outcomes. Biomarkers for monitoring the changes in the patient’s state can potentially increase the efficiency of clinical trials. Digital biomarkers, which emerged in recent years, rely on applications of machine learning methods to the data gathered by low-cost sensors, often embedded in consumer devices. Digital biomarkers have the potential to provide low-cost and more objective, granular, and sensitive to change metrics than traditional clinical ratings used in assessments of neurological and neurodevelopmental disorders. On the other hand, in traditional electrophysiological methods measuring brain activity, such as electroencephalography (EEG), biomarkers historically were based on visual analysis by clinicians, classical signal processing measures, or event-related potential (ERP) technique. Search for machine learning-based EEG biomarkers is an active area of research. This dissertation aims to build novel digital behavioral and EEG-based biomarkers and outcome measures by applying machine learning to behavioral, EEG, and concurrently recorded behavioral and EEG data. Machine learning models for the detection of gaze, human face and body landmarks, and automatic speech recognition achieve good performance on publicly available datasets. However, applying these models to a new clinical dataset immediately incurs a dataset shift problem, since the conditions under which real clinical video and audio data are recorded differe from the training dataset (e.g. different video camera angles, or audio noise). Furthermore, clinical datasets are in general much smaller than those used for training such models, and there are not enough human resources in the clinical setting to perform data labeling, making re-training not feasible. Yet, the question remains – whether the predictions from pre-trained models can provide valuable insight into human behavior and neurophysiology in the clinical setting, and whether they can be a source of clinically relevant findings.
In this dissertation, we first explore this question in two use cases: (1) building digital measures of caregiver-child interaction in neurodevelopmental disorders using pre-trained pose detection deep learning models; (2) creating a digital biomarker of ataxic dysarthria using pre-trained automatic speech recognition deep learning models. We show that in the first case, our method enables to distinguish different clusters of caregiver responsiveness which are associated with a child’s caregiver- and clinician-reported socialization, communication, and language abilities, thus demonstrating the feasibility of using digital measures of caregiver-child interaction in clinical trials. In the second case, we demonstrate the convergent validity of our novel biomarker with clinician-reported scores and the greater sensitivity to change than clinician-reported scores on a longitudinal dataset.
Second, we propose a novel deep learning model for detecting seizures in neonates from EEG data. We demonstrate the model’s high generalizability by evaluating it on an independent dataset from another hospital and show that model by design can be applied in different facilities with different EEG hardware. This approach has the potential to be clinically validated and will allow to scale up studies of neonatal seizures by increasing the sample sizes (including data from multiple clinical centers).
Finally, we turn to the problem of combining EEG and behavioral biomarkers, which can improve biomarker sensitivity, but also provide new insights into brain-behavior relationships. In the study of autism, we propose a new metric of attentional preference to social/non-social stimuli and show that not only it distinguishes between autistic and neurotypical children, but also is differently associated with brain activity as measured by EEG. Then we turn to the question of scaling up EEG and behavior studies and provide the tool that allows measuring participants’ attention to the screen during EEG recording. This tool will allow to reduce human effort and make measurements of participants’ visual attention more objective, thus scaling up data preprocessing and allowing for multi-center studies of concurrent EEG and behavior.
</p
Information support of corporate governance and strategic management using analytical software
No full textImplementation of electric drive of withdrawal-roll set in horizontal continuous casting machine for the production of small billets
No full text
AUTHOR STRATEGIES IN RYURIK IVNEV’S “SUN IN THE COFFIN”. BOOK OF POEMS
No full textThe author’s problem in Ryurik Ivnev’s ‘The Sun in the Coffin’ (1921) is studied from the author’s perspectives. It is concluded that at least three Ryurik Ivnev’s strategies as an author can be distinguished: the first, the least obvious, coming from his youthful excitement about Nadson’s neo-romantic poetry, Baudelaire’s and Blok’s symbolism and appearing in a very weakened form, can be called a decadent one. The second strategy is determined by the poet’s recent belonging to egofuturism. The third one was formed under the influence of imaginism aesthetics, to which the poet was attracted when writing the book. Ryurik Ivnev’s strategies, spontaneously but sometimes consciously, were aimed at forming the literary image of the author, identified with the persona. The themes and motives of the book are concentrated around the hypertrophic and screened in the text «I» which becomes the center of Ryurik Ivnev’s poetic worldview. All the strategies are intricately intertwined and give rise to the quality of the style that distinguishes sharply ‘The Sun in the Coffin’ against the background of the Russian poetry of the 1920s and which can be referred to as surrealism
Standard and Modified SST Models with the Consideration of the Streamline Curvature for Separated Flow Calculation in a Narrow Channel with a Conical Dimple on the Heated Wall
No full textThe testing of the standard and modified SST models of the transfer of shear stresses was carried out on an example of calculating the heat transfer with an intense detached flow around a conical dimple with a slope angle of 45° on the heated wall of a narrow channel. It was shown that the standard turbulence model by Menter SST (MSST) of 2003, widely used in the packages Fluent, CFX, StarCCM+, etc., significantly underestimated the intensity of the return flow. A correction of this model was presented that took into account the influence of the curvature of streamlines within the framework of the Rodi-Leshziner-Isaev (RLI) approach for spatial separated flows. It was found that the predictions for the RLI MSST 2003 were close to the predictions for the original standard MSST 1993, in which the eddy viscosity was calculated using the vorticity modulus. At the same time, the predictions based on the modified one, following Smirnov-Menter (SM) MSST 2003, included in the ANSYS model catalog did not differ too much from the standard MSST 2003. The preference of the MSST modified within the RLI 2003 for calculating the heat transfer in intense separated flows was substantiated
Numerical Simulation of a Periodic Quasi-Switching Mode of Flow around a Conical Dimple with a Slope Angle of 10 Degrees on the Wall of a Narrow Channel Using URANS
No full textThe applicability of the solution of the unsteady Reynolds-averaged Navier–Stokes equations (URANS) for the numerical simulation of the periodic quasi-switching regime of vortex generation and heat transfer in a deep conical dimple with a slope angle of 10∘ on the wall of a narrow channel is substantiated. To calculate the turbulent regime, the model of shear stress transfer by Menter 2003, modified taking into account the influence of the curvature of streamlines within the framework of the Rodi-Leshziner-Isaev approach, is used. At Reynolds number Re=104, the oscillation period of the transverse Rz and longitudinal forces Rx, as well as the total heat transfer Numm to the control section of the heated channel wall with a dimple, is set equal to 60, which corresponds to the Strouhal number St=0.0167. Computer visualization of swirling jet-vortex flows demonstrates focus-type sources on the side faces of the dimple. In the self-oscillating mode, a two-cell vortex system is formed with different intensities at the oscillation period Rz. Periodic changes in friction, Nusselt numbers and temperature are recorded in the longitudinal and transverse median sections of the dimple and reflect the oscillations of the vortex structure from left to right and from right to left. The formation of a fan jet is shown, which oscillates relative to the plane of longitudinal symmetry, causing a redistribution of power and thermal loads
- …
