Search

IntroductionArtificial Intelligence (AI) methods are being increasingly investigated as a means to generate predictive models applicable in the clinical practice. In this study, we developed a model to predict the efficacy of immunotherapy (IO) in patients with advanced non-small cell lung cancer (NSCLC) using eXplainable AI (XAI) Machine Learning (ML) methods.MethodsWe prospectively collected real-world data from patients with an advanced NSCLC condition receiving immune-checkpoint inhibitors (ICIs) either as a single agent or in combination with chemotherapy. With regards to six different outcomes - Disease Control Rate (DCR), Objective Response Rate (ORR), 6 and 24-month Overall Survival (OS6 and OS24), 3-months Progression-Free Survival (PFS3) and Time to Treatment Failure (TTF3) - we evaluated five different classification ML models: CatBoost (CB), Logistic Regression (LR), Neural Network (NN), Random Forest (RF) and Support Vector Machine (SVM). We used the Shapley Additive Explanation (SHAP) values to explain model predictions.ResultsOf 480 patients included in the study 407 received immunotherapy and 73 chemo- and immunotherapy. From all the ML models, CB performed the best for OS6 and TTF3, (accuracy 0.83 and 0.81, respectively). CB and LR reached accuracy of 0.75 and 0.73 for the outcome DCR. SHAP for CB demonstrated that the feature that strongly influences models’ prediction for all three outcomes was Neutrophil to Lymphocyte Ratio (NLR). Performance Status (ECOG-PS) was an important feature for the outcomes OS6 and TTF3, while PD-L1, Line of IO and chemo-immunotherapy appeared to be more important in predicting DCR.ConclusionsIn this study we developed a ML algorithm based on real-world data, explained by SHAP techniques, and able to accurately predict the efficacy of immunotherapy in sets of NSCLC patients.

The physical boundaries of our body do not define what we perceive as self. This malleable representation arises from the neural integration of sensory information coming from the environment. Manipulating the visual and haptic cues produces changes in body perception, inducing the Full Body Illusion (FBI), a vastly used approach to exploring humans’ perception. After pioneering FBI demonstrations, issues arose regarding its setup, using experimenter-based touch and pre-recorded videos. Moreover, its outcome measures are based mainly on subjective reports, leading to biased results, or on heterogeneous objective ones giving poor consensus on their validity. To address these limitations, we developed and tested a multisensory platform allowing highly controlled experimental conditions, thanks to the leveraged use of innovative technologies: Virtual Reality (VR) and Transcutaneous Electrical Nerve Stimulation (TENS). This enabled a high spatial and temporal precision of the visual and haptic cues, efficiently eliciting FBI. While it matched the classic approach in subjective measures, our setup resulted also in significant results for all objective measurements. Importantly, FBI was elicited when all 4 limbs were multimodally stimulated but also in a single limb condition. Our results behoove the adoption of a comprehensive set of measures, introducing a new neuroscientific platform to investigate body representations.

Over the past several years, electromyography (EMG) signals have been used as a natural interface to interact with computers and machines. Recently, deep learning algorithms such as Convolutional Neural Networks (CNNs) have gained interest for decoding the hand movement intention from EMG signals. However, deep networks require a large dataset to train appropriately. Creating such a database for a single subject could be very time-consuming. In this study, we addressed this issue from two perspectives: (i) we proposed a subject-transfer framework to use the knowledge learned from other subjects to compensate for a target subject’s limited data; (ii) we proposed a task-transfer framework in which the knowledge learned from a set of basic hand movements is used to classify more complex movements, which include a combination of mentioned basic movements. We introduced two CNN-based architectures for hand movement intention detection and a subject-transfer learning approach. Classifiers are tested on the Nearlab dataset, a sEMG hand/wrist movement dataset including 8 movements and 11 subjects, along with their combination, and on open-source hand sEMG dataset “NinaPro DataBase 2 (DB2).” For the Nearlab database, the subject-transfer learning approach improved the average classification accuracy of the proposed deep classifier from 92.60 to 93.30% when classifier was utilizing 10 other subjects’ data via our proposed framework. For Ninapro DB2 exercise B (17 hand movement classes), this improvement was from 81.43 to 82.87%. Moreover, three stages of analysis in task-transfer approach proved that it is possible to classify combination hand movements using the knowledge learned from a set of basic hand movements with zero, few samples and few seconds of data from the target movement classes. First stage takes advantage of shared muscle synergies to classify combined movements, while second and third stages take advantage of novel algorithms using few-shot learning and fine-tuning to use samples from target domain to further train the classifier trained on the source database. The use of information learned from basic hand movements improved classification accuracy of combined hand movements by 10%.

C-band technology is emerging as an exciting innovative approach to the creation of compact new accelerators. Besides the possibility to sustain higher gradients and higher repetition rate operation at normal conducting temperature, it also allows one to increase the machine performance in terms of beam brightness. We propose the design study of a normal conducting, high gradient C-band injector aiming at the production of high brightness electron beams—up to 2.6×10^{3} TA/m^{2}—at high repetition rate—up to 1 KHz—as desired to enhance the capabilities of modern radiation sources. This paper reports on beam dynamics studies that guided the injector design looking for a good compromise between the machine compactness and performances. For this purpose, a new standing wave C-band gun consisting of 2.6 cells has been designed to enhance the final beam brightness together with its own mode launcher to ensure the needed peak field at the cathode—up to 180 MV/m. Several working points in terms of peak current, transverse emittance and brightness are explored, showing the flexibility allowed by the proposed photoinjector. The integration of the gun mode launcher, that relies on a four-port configuration, in the injector design and its effect on the beam quality are also reported.

Background: Functional Electrical Stimulation (FES) represents a promising technique for promoting functional recovery in individuals with neuromuscular diseases. Traditionally, current pulses are delivered through self‐adhesive hydrogel Ag/AgCl electrodes, which allow a good contact with the skin, are easy‐to‐use and have a moderate cost. However, skin adherence decreases after a few uses and skin irritations can originate. Recently, textile electrodes have become an attractive alternative as they assure increased durability, easy integration into clothes and can be conveniently cleaned, improving the wearability of FES. However, as various manufacture processes were attempted, their clear validation is lacking. This proof‐of‐concept study proposes a novel set of ink‐based printed textile electrodes and compares them to adhesive hydrogel electrodes in terms of impedance, stimulation performance and perceived comfort. Methods: The skin–electrode impedance was evaluated for both types of electrodes under different conditions. These electrodes were then used to deliver FES to the Rectus Femoris of 14 healthy subjects to induce its contraction in both isometric and dynamic conditions. This allowed to compare the two types of electrodes in terms of sensory, motor, maximum and pain thresholds, FES‐induced range of motion during dynamic tests, FES‐induced torque during isometric tests and perceived stimulation comfort. Results: No statistically significant differences were found both in terms of stimulation performance (Wilcoxon test) and comfort (Generalized Linear Mixed Model). Conclusion: The results showed that the proposed ink‐based printed textile electrodes can be effectively used as alternative to hydrogel ones. Further experiments are needed to evaluate their durability and their response to sterilizability and stretching tests. [ABSTRACT FROM AUTHOR]

Parkinsonism is presented as a motor syndrome characterized by rigidity, tremors, and bradykinesia, with Parkinson's disease (PD) being the predominant cause. The discovery that those motor symptoms result from the death of dopaminergic cells in the substantia nigra led to focus most of parkinsonism research on the basal ganglia (BG). However, recent findings point to an active involvement of the cerebellum in this motor syndrome. Here, we have developed a multiscale computational model of the rodent brain's BG–cerebellar network. Simulations showed that a direct effect of dopamine depletion on the cerebellum must be taken into account to reproduce the alterations of neural activity in parkinsonism, particularly the increased beta oscillations widely reported in PD patients. Moreover, dopamine depletion indirectly impacted spike-time-dependent plasticity at the parallel fiber-Purkinje cell synapses, degrading associative motor learning as observed in parkinsonism. Overall, these results suggest a relevant involvement of cerebellum in parkinsonism associative motor symptoms. [ABSTRACT FROM AUTHOR]

We reproduced a decision-making network model using the neural simulator software neural simulation tool (NEST), and we embedded the spiking neural network in a virtual robotic agent performing a simulated behavioral task. The present work builds upon the concept of replicability in neuroscience, preserving most of the computational properties in the initial model although employing a different software tool. The proposed implementation successfully obtains equivalent results from the original study, reproducing the salient features of the neural processes underlying a binary decision. Furthermore, the resulting network is able to control a robot performing an in silico visual discrimination task, the implementation of which is openly available on the EBRAINS infrastructure through the neuro robotics platform (NRP).

It is common for animals to use self-generated movements to actively sense the surrounding environment. For instance, rodents rhythmically move their whiskers to explore the space close to their body. The mouse whisker system has become a standard model for studying active sensing and sensorimotor integration through feedback loops. In this work, we developed a bioinspired spiking neural network model of the sensorimotor peripheral whisker system, modeling trigeminal ganglion, trigeminal nuclei, facial nuclei, and central pattern generator neuronal populations. This network was embedded in a virtual mouse robot, exploiting the Human Brain Project's Neurorobotics Platform, a simulation platform offering a virtual environment to develop and test robots driven by brain-inspired controllers. Eventually, the peripheral whisker system was adequately connected to an adaptive cerebellar network controller. The whole system was able to drive active whisking with learning capability, matching neural correlates of behavior experimentally recorded in mice.

Abstract Background Upper limb assistive devices can compensate for muscular weakness and empower the user in the execution of daily activities. Multiple devices have been recently proposed but there is still a lack in the scientific comparison of their efficacy. Methods We conducted a cross-over multi-centric randomized controlled trial to assess the functional improvement at the upper limb level of two arms supports on 36 patients with muscular dystrophy. Participants tested a passive device (i.e., Wrex by Jaeco) and a semi-active solution for gravity compensation (i.e., Armon Ayura). We evaluated devices’ effectiveness with an externally-assessed scale (i.e., Performance of the Upper Limb-PUL-module), a self-perceived scale (i.e., Abilhand questionnaire), and a usability scale (i.e., System Usability Scale). Friedman’s test was used to assess significant functional gain for PUL module and Abilhand questionnaire. Moreover, PUL changes were compared by means of the Friedman’s test. Results Most of the patients improved upper limb function with the use of arm supports (median PUL scores increase of 1–3 points). However, the effectiveness of each device was related to the level of residual ability of the end-user. Slightly impaired patients maintained the same independence without and with assistive devices, even if they reported reduced muscular fatigue for both devices. Moderately impaired patients enhanced their arm functionality with both devices, and they obtained higher improvements with the semi-active one (median PUL scores increase of 9 points). Finally, severely impaired subjects benefited only from the semi-active device (median PUL scores increase of 12 points). Inadequate strength was recognized as a barrier to passive devices. The usability, measured by the System Usability Scale, was evaluated by end-users “good” (70/100 points) for the passive, and “excellent” (80/100 points) for the semi-active device. Conclusions This study demonstrated that assistive devices can improve the quality of life of people suffering from muscular dystrophy. The use of passive devices, despite being low cost and easy to use, shows limitations in the efficacy of the assistance to daily tasks, limiting the assistance to a predefined horizontal plane. The addition of one active degree of freedom improves efficacy and usability especially for medium to severe patients. Further investigations are needed to increase the evidence on the effect of arm supports on quality of life and diseases’ progression in subjects with degenerative disorders. Trial registration clinicaltrials.gov, NCT03127241, Registered 25th April 2017. The clinical trial was also registered as a post-market study at the Italian Ministry of Health.