Publications

EEG-based brain-machine interfaces (BMIs) offer an intuitive approach for individuals with motor impairments to control prosthetic or rehabilitation devices. Decoding movement intentions plays a vital role in accurately translating the motor execution plans of subjects, such as identifying the desired grasp type or target position. In this study, EEG signals were recorded from seven healthy subjects during self-paced reaching and grasping tasks. Power Spectral Density (PSD) and Entropy were extracted as features to assess their efficacy in discrimination between different brain states related to movement planning. Classification between anticipation of movement and resting-state periods were evaluated using machine-learning methods (Linear Discriminant Analysis, Quadratic Discriminant Analysis, and Support Vector Machines). The achieved results provide strong evidence for the feasibility of decoding movement intention, laying the foundation for future applications and advancements in the field. Attaining high accuracy in decoding movement intention holds significant potential for the translational applications of BMIs in the fields of biomedical engineering and rehabilitation.

Pompe disease (PD) is a monogenic autosomal recessive disorder caused by biallelic pathogenic variants of the GAA gene encoding lysosomal alpha-glucosidase; its loss causes glycogen storage in lysosomes, mainly in the muscular tissue. The genotype–phenotype correlation has been extensively discussed, and caution is recommended when interpreting the clinical significance of any mutation in a single patient. As there is no evidence that environmental factors can modulate the phenotype, the observed clinical variability in PD suggests that genetic variants other than pathogenic GAA mutations influence the mechanisms of muscle damage/repair and the overall clinical picture. Genes encoding proteins involved in glycogen synthesis and catabolism may represent excellent candidates as phenotypic modifiers of PD. The genes analyzed for glycogen synthesis included UGP2, glycogenin (GYG1-muscle, GYG2, and other tissues), glycogen synthase (GYS1-muscle and GYS2-liver), GBE1, EPM2A, NHLRC1, GSK3A, and GSK3B. The only enzyme involved in glycogen catabolism in lysosomes is α-glucosidase, which is encoded by GAA, while two cytoplasmic enzymes, phosphorylase (PYGB-brain, PGL-liver, and PYGM-muscle) and glycogen debranching (AGL) are needed to obtain glucose 1-phosphate or free glucose. Here, we report the potentially relevant variants in genes related to glycogen synthesis and catabolism, identified by whole exome sequencing in a group of 30 patients with late-onset Pompe disease (LOPD). In our exploratory analysis, we observed a reduced number of variants in the genes expressed in muscles versus the genes expressed in other tissues, but we did not find a single variant that strongly affected the phenotype. From our work, it also appears that the current clinical scores used in LOPD do not describe muscle impairment with enough qualitative/quantitative details to correlate it with genes that, even with a slightly reduced function due to genetic variants, impact the phenotype.

The nutraceutical supplementation of Spirulina (Arthrospira platensis) in dogs and cats has not yet been investigated. The aim of this study was to evaluate if the dietary supplementation of increasing amounts of Spirulina for 6 weeks is palatable to pets and to assess the owner’s perception of such supplementation. The owners of the 60 dogs and 30 cats that participated in this study were instructed to daily provide Spirulina tablets starting with a daily amount of 0.4 g, 0.8 g, and 1.2 g for cats as well as small dogs, medium dogs, and large dogs, respectively, and allowing a dose escalation of 2× and 3× every 2 weeks. The daily amount (g/kg BW) of Spirulina ranged from 0.08 to 0.25 for cats, from 0.06 to 0.19 for small-sized dogs, from 0.05 to 0.15 for medium-sized dogs, and from 0.04 to 0.12 for large-sized dogs. Each owner completed a questionnaire at the time of recruitment and the end of each 2-week period. No significant effect on the fecal score, defecation frequency, vomiting, scratching, lacrimation, general health status, and behavioral attitudes was detected by the owners’ reported evaluations. Most animals accepted Spirulina tablets either administrated alone or mixed with food in the bowl. Daily supplementation of Spirulina for 6 weeks in the amounts provided in this study is therefore palatable and well tolerated by dogs and cats.

Brain-machine interfaces (BMIs) are alternative communication channels that have allowed healthy and disabled people to control external devices from brain signals. In the last decades, the growing attention towards neurorobotics has led to the proliferation of several BMI-based systems for controlling different devices including telepresence robots, powered wheelchairs, robotic arms, and upper/lower-limb exoskeletons. Despite the potentialities of these systems, it has emerged the necessity to create new forms of interaction between the human and the robot in order to increase the granularity of the user’s commands which are, in turn, translated into specific robot’s actions. In this preliminary work, we present how artificial intelligence can be exploited to design and tune a model able to convert the user’s intention into continuous robot’s movements.

The collaborative transport of objects between humans and robots is one of the main areas of focus in physical Human-Robot Interaction (pHRI). Ensuring the operator’s safety and maintaining collision-free motion of the robot during transportation are crucial challenges in this context. Consider a collaborative co-manipulation scenario where the operator modifies the trajectory being executed by the robot. In such cases, the robot may deviate from its previously calculated path, potentially resulting in collisions. In this work, we propose a method to estimate the maximum collision-free volume around the path of the robot. This volume represents the permissible deviation introduced by the human worker while ensuring that no collisions occur. To evaluate the effectiveness of the proposed algorithm, we test it in a real industrial scenario.

Hand-eye calibration is the problem of estimating the spatial transformation between a reference frame, usually the base of a robot arm or its gripper, and the reference frame of one or multiple cameras. Generally, this calibration is solved as a non-linear optimization problem, what instead is rarely done is to exploit the underlying graph structure of the problem itself. Actually, the problem of hand-eye calibration can be seen as an instance of the Simultaneous Localization and Mapping (SLAM) problem. Inspired by this fact, in this work we present a pose-graph approach to the hand-eye calibration problem that extends a recent state-of-the-art solution in two different ways: i) by formulating the solution to eye-on-base setups with one camera; ii) by covering multi-camera robotic setups. The proposed approach has been validated in simulation against standard hand-eye calibration methods. Moreover, a real application is shown. In both scenarios, the proposed approach overcomes all alternative methods. We release with this paper an open-source implementation of our graph-based optimization framework for multi-camera setups.

Robot grasping has been widely studied in the last decade. Recently, Deep Learning made possible to achieve remarkable results in grasp pose estimation, using depth and RGB images. However, only few works consider the choice of the object to grasp. Moreover, they require a huge amount of data for generalizing to unseen object categories. For this reason, we introduce the Few-shot Semantic Grasping task where the objective is inferring a correct grasp given only five labelled images of a target unseen object. We propose a new deep learning architecture able to solve the aforementioned problem, leveraging on a Few-shot Semantic Segmentation module. We have evaluated the proposed model both in the Graspnet dataset and in a real scenario. In Graspnet, we achieve 40,95% accuracy in the Few-shot Semantic Grasping task, outperforming baseline approaches. In the real experiments, the results confirmed the generalization ability of the network.