Motor Unit Spike Trains Recontruction from sEMG Signals for Gesture Classification on a Low-Power Processing Platform

Hand gesture recognition based on surface electromyography (sEMG) signals is a promising approach for the development of intuitive human-machine interfaces (HMIs) in domains such as robotics and prosthetics. The sEMG signal arises from the muscles' electrical activity, and can thus be used to recognize hand gestures. The decoding from sEMG signals to actual control signals is non-trivial; typically, control systems map sEMG patterns into a set of gestures using machine learning, failing to incorporate any physiological insight. This master thesis aims at developing a bio-inspired hand gesture recognition system based on neuromuscular spike extraction rather than on simple pattern recognition. The system relies on a decomposition algorithm based on independent component analysis (ICA) that decomposes the sEMG signal into its constituent motor unit spike trains, which are then forwarded to a machine learning classifier. Since ICA does not guarantee a consistent motor unit ordering across different sessions, 3 approaches are proposed: 2 ordering criteria based on firing rate and negative entropy, and a re-calibration approach that allows the decomposition model to retain information about previous sessions. Using a multilayer perceptron (MLP), the latter approach results in an accuracy up to 99.4% in a 1-subject, 1-degree of freedom scenario. Afterwards, the decomposition and classification pipeline for inference is parallelized and profiled on the PULP platform, achieving a latency < 50 ms and an energy consumption < 1 mJ. Both the classification models tested (a support vector machine and a lightweight MLP) yielded an accuracy > 92% in a 1-subject, 5-classes (4 gestures and rest) scenario. These results prove that the proposed system is suitable for real-time execution on embedded platforms and also capable of matching the accuracy of state-of-the-art approaches, while also giving some physiological insight on the neuromuscular spikes underlying the sEMG.

Biogas and bio-hydrogen: production and uses. A review

The first part of this essay aims at investigating the already available and promising technologies for the biogas and bio-hydrogen production from anaerobic digestion of different organic substrates. One strives to show all the peculiarities of this complicate process, such as continuity, number of stages, moisture, biomass preservation and rate of feeding. The main outcome of this part is the awareness of the huge amount of reactor configurations, each of which suitable for a few types of substrate and circumstance. Among the most remarkable results, one may consider first of all the wet continuous stirred tank reactors (CSTR), right to face the high waste production rate in urbanised and industrialised areas. Then, there is the up-flow anaerobic sludge blanket reactor (UASB), aimed at the biomass preservation in case of highly heterogeneous feedstock, which can also be treated in a wise co-digestion scheme. On the other hand, smaller and scattered rural realities can be served by either wet low-rate digesters for homogeneous agricultural by-products (e.g. fixed-dome) or the cheap dry batch reactors for lignocellulose waste and energy crops (e.g. hybrid batch-UASB). The biological and technical aspects raised during the first chapters are later supported with bibliographic research on the important and multifarious large-scale applications the products of the anaerobic digestion may have. After the upgrading techniques, particular care was devoted to their importance as biofuels, highlighting a further and more flexible solution consisting in the reforming to syngas. Then, one shows the electricity generation and the associated heat conversion, stressing on the high potential of fuel cells (FC) as electricity converters. Last but not least, both the use as vehicle fuel and the injection into the gas pipes are considered as promising applications. The consideration of the still important issues of the bio-hydrogen management (e.g. storage and delivery) may lead to the conclusion that it would be far more challenging to implement than bio-methane, which can potentially “inherit” the assets of the similar fossil natural gas. Thanks to the gathered knowledge, one devotes a chapter to the energetic and financial study of a hybrid power system supplied by biogas and made of different pieces of equipment (natural gas thermocatalitic unit, molten carbonate fuel cell and combined-cycle gas turbine structure). A parallel analysis on a bio-methane-fed CCGT system is carried out in order to compare the two solutions. Both studies show that the apparent inconvenience of the hybrid system actually emphasises the importance of extending the computations to a broader reality, i.e. the upstream processes for the biofuel production and the environmental/social drawbacks due to fossil-derived emissions. Thanks to this “boundary widening”, one can realise the hidden benefits of the hybrid over the CCGT system.

Bio-QA-GNN: Reasoning with Language Models and Knowledge Graphs for Interpretable Biomedical Question Answering

Nowadays the idea of injecting world or domain-specific structured knowledge into pre-trained language models (PLMs) is becoming an increasingly popular approach for solving problems such as biases, hallucinations, huge architectural sizes, and explainability lack—critical for real-world natural language processing applications in sensitive fields like bioinformatics. One recent work that has garnered much attention in Neuro-symbolic AI is QA-GNN, an end-to-end model for multiple-choice open-domain question answering (MCOQA) tasks via interpretable text-graph reasoning. Unlike previous publications, QA-GNN mutually informs PLMs and graph neural networks (GNNs) on top of relevant facts retrieved from knowledge graphs (KGs). However, taking a more holistic view, existing PLM+KG contributions mainly consider commonsense benchmarks and ignore or shallowly analyze performances on biomedical datasets. This thesis start from a propose of a deep investigation of QA-GNN for biomedicine, comparing existing or brand-new PLMs, KGs, edge-aware GNNs, preprocessing techniques, and initialization strategies. By combining the insights emerged in DISI's research, we introduce Bio-QA-GNN that include a KG. Working with this part has led to an improvement in state-of-the-art of MCOQA model on biomedical/clinical text, largely outperforming the original one (+3.63\% accuracy on MedQA). Our findings also contribute to a better understanding of the explanation degree allowed by joint text-graph reasoning architectures and their effectiveness on different medical subjects and reasoning types. Codes, models, datasets, and demos to reproduce the results are freely available at: \url{https://github.com/disi-unibo-nlp/bio-qagnn}.