Search

In order to behave adaptively and to successfully interact with our environment, we must learn from the consequences of our actions. Anticipation and processing of rewards is omni-present in modern behavioral experiments since it regulates interaction and is altered in a diverse set of psychiatric disorders. Our understanding of neural processes underlying the feedback on reward or punishment has largely advanced due to research in electroencephalography (EEG). Within this context, the feedback-Related Negativity (FRN), an event-related potential (ERP) and EEG variable tracts responsiveness to reward and reward learning. Furthermore, it is a valuable assessment tool for a modern descriptive-dimensional diagnostic system by the National Institute for Mental Health (NIMH) called Research-Domain-Criteria (RDoC). It needs to be discussed whether FRN can functionally act as a biomarker passing certain criteria from a theoretical and mechanistic point of view. Showing consistency in related psychopathologies would imply great capacity of FRN in future EEG-research and diagnostic investigation. In this regard, we will use the Internalizing-Externalizing-Spectrum (IES), a well-established nosology system, testing the practicality of FRN along a continuous psychopathology system in a metanalytic approach.

Scientific reports 11(1), 5078 (2021). doi:10.1038/s41598-021-84496-z, Published by Macmillan Publishers Limited, part of Springer Nature, [London]

Frontiers in psychology 13, 762079 (2022). doi:10.3389/fpsyg.2022.762079, Published by Frontiers Research Foundation, Lausanne

The absolute neutrino mass is still a missing parameter in the modern landscape of particle physics.The HOLMES experiment aims to perform a direct measurement of the neutrino mass with a sensitivity of the order of 2 eV. The neutrino mass will be studied through the calorimetric measurement of the decay products of the weak process decay of 163 Ho. To achieve the target sensitivity, HOLMES will combine the use of 1000 low temperature TES microcalorimeters with the use of a sophisticated software, designed for the signal processing and data analysis. The detector and readout technology will be pushed to their limits in order to preserve the detectors performance and the high multiplexing factor at the same time. This contribution will provide an overview on the status of the major tasks that will bring HOLMES to life: from the isotope production and embedding to the detector production and readout.

This short paper presents the design decisions taken and challenges encountered in completing SemEval Task 6, which poses the problem of identifying and categorizing offensive language in tweets. Our proposed solutions explore Deep Learning techniques, Linear Support Vector classification and Random Forests to identify offensive tweets, to classify offenses as targeted or untargeted and eventually to identify the target subject type.

Invasive coronary angiography (ICA) is the gold standard in Coronary Artery Disease (CAD) imaging. Detection of the end-diastolic frame (EDF) and, in general, cardiac phase detection on each temporal frame of a coronary angiography acquisition is of significant importance for the anatomical and non-invasive functional assessment of CAD. This task is generally performed via manual frame selection or semi-automated selection based on simultaneously acquired ECG signals - thus introducing the requirement of simultaneous ECG recordings. We evaluate the performance of a purely image based workflow based on deep neural networks for fully automated cardiac phase and EDF detection on coronary angiographies. A first deep neural network (DNN), trained to detect coronary arteries, is employed to preselect a subset of frames in which coronary arteries are well visible. A second DNN predicts cardiac phase labels for each frame. Only in the training and evaluation phases for the second DNN, ECG signals are used to provide ground truth labels for each angiographic frame. The networks were trained on 17800 coronary angiographies from 3900 patients and evaluated on 27900 coronary angiographies from 6250 patients. No exclusion criteria related to patient state, previous interventions, or pathology were formulated. Cardiac phase detection had an accuracy of 97.6%, a sensitivity of 97.6% and a specificity of 97.5% on the evaluation set. EDF prediction had a precision of 97.4% and a recall of 96.9%. Several sub-group analyses were performed, indicating that the cardiac phase detection performance is largely independent from acquisition angles and the heart rate of the patient. The average execution time of cardiac phase detection for one angiographic series was on average less than five seconds on a standard workstation., 16 pages, 9 figures

Motivated by state-of-the-art performances acrossa wide variety of areas, over the last few years Machine Learning has drawn a significant amount of attention from thehealthcare domain. Despite their potential in enabling personalized medicine applications, the adoption of Deep Learningbased solutions in clinical workflows has been hindered in many cases by the strict regulations concerning the privacy ofpatient health data. We propose a solution that relies on Fully Homomorphic Encryption, particularly on the MORE scheme,as a mechanism for enabling computations on sensitive health data, without revealing the underlying data. The chosen variantof the encryption scheme allows for the computations in the Neural Network model to be directly performed on floatingpoint numbers, while incurring a reasonably small computational overhead. For feasibility evaluation, we demonstrateon the MNIST digit recognition task that Deep Learning canbe performed on encrypted data without compromising theaccuracy. We then address a more complex task by training amodel on encrypted data to estimate the outputs of a wholebodycirculation (WBC) model. These results underline thepotential of the proposed approach to outperform currentsolutions by delivering comparable results to the unencryptedDeep Learning based solutions, in a reasonable amount oftime. Lastly, the security aspects of the encryption scheme areanalyzed, and we show that, even though the chosen encryptionscheme favors performance and utility at the cost of weakersecurity, it can still be used in certain practical applications.

Motivated by state-of-the-art performances across a wide variety of areas, over the last few years Machine Learning has drawn a significant amount of attention from the healthcare domain. Despite their potential in enabling personalized medicine applications, the adoption of Deep Learning based solutions in clinical workflows has been hindered in many cases by the strict regulations concerning the privacy of patient health data. We propose a solution that relies on Fully Homomorphic Encryption, particularly on the MORE scheme, as a mechanism for enabling computations on sensitive health data, without revealing the underlying data. The chosen variant of the encryption scheme allows for the computations in the Neural Network model to be directly performed on floating point numbers, while incurring a reasonably small computational overhead. For feasibility evaluation, we demonstrate on the MNIST digit recognition task that Deep Learning can be performed on encrypted data without compromising the accuracy. We then address a more complex task by training a model on encrypted data to estimate the outputs of a wholebody circulation (WBC) model. These results underline the potential of the proposed approach to outperform current solutions by delivering comparable results to the unencrypted Deep Learning based solutions, in a reasonable amount of time. Lastly, the security aspects of the encryption scheme are analyzed, and we show that, even though the chosen encryption scheme favors performance and utility at the cost of weaker security, it can still be used in certain practical applications.