Preventing worker’s musculoskeletal disorders with wearable inertial sensors: ergonomics and modelling of physical human-robot interactions

The most widespread work-related diseases are musculoskeletal disorders (MSD) caused by awkward postures and excessive effort to upper limb muscles during work operations. The use of wearable IMU sensors could monitor the workers constantly to prevent hazardous actions, thus diminishing work injuries. In this thesis, procedures are developed and tested for ergonomic analyses in a working environment, based on a commercial motion capture system (MoCap) made of 17 Inertial Measurement Units (IMUs). An IMU is usually made of a tri-axial gyroscope, a tri-axial accelerometer, and a tri-axial magnetometer that, through sensor fusion algorithms, estimates its attitude. Effective strategies for preventing MSD rely on various aspects: firstly, the accuracy of the IMU, depending on the chosen sensor and its calibration; secondly, the correct identification of the pose of each sensor on the worker’s body; thirdly, the chosen multibody model, which must consider both the accuracy and the computational burden, to provide results in real-time; finally, the model scaling law, which defines the possibility of a fast and accurate personalization of the multibody model geometry. Moreover, the MSD can be diminished using collaborative robots (cobots) as assisted devices for complex or heavy operations to relieve the worker's effort during repetitive tasks. All these aspects are considered to test and show the efficiency and usability of inertial MoCap systems for assessing ergonomics evaluation in real-time and implementing safety control strategies in collaborative robotics. Validation is performed with several experimental tests, both to test the proposed procedures and to compare the results of real-time multibody models developed in this thesis with the results from commercial software. As an additional result, the positive effects of using cobots as assisted devices for reducing human effort in repetitive industrial tasks are also shown, to demonstrate the potential of wearable electronics in on-field ergonomics analyses for industrial applications.

Data-driven multivariate flood hazard modeling

The present Dissertation shows how recent statistical analysis tools and open datasets can be exploited to improve modelling accuracy in two distinct yet interconnected domains of flood hazard (FH) assessment. In the first Part, unsupervised artificial neural networks are employed as regional models for sub-daily rainfall extremes. The models aim to learn a robust relation to estimate locally the parameters of Gumbel distributions of extreme rainfall depths for any sub-daily duration (1-24h). The predictions depend on twenty morphoclimatic descriptors. A large study area in north-central Italy is adopted, where 2238 annual maximum series are available. Validation is performed over an independent set of 100 gauges. Our results show that multivariate ANNs may remarkably improve the estimation of percentiles relative to the benchmark approach from the literature, where Gumbel parameters depend on mean annual precipitation. Finally, we show that the very nature of the proposed ANN models makes them suitable for interpolating predicted sub-daily rainfall quantiles across space and time-aggregation intervals. In the second Part, decision trees are used to combine a selected blend of input geomorphic descriptors for predicting FH. Relative to existing DEM-based approaches, this method is innovative, as it relies on the combination of three characteristics: (1) simple multivariate models, (2) a set of exclusively DEM-based descriptors as input, and (3) an existing FH map as reference information. First, the methods are applied to northern Italy, represented with the MERIT DEM (∼90m resolution), and second, to the whole of Italy, represented with the EU-DEM (25m resolution). The results show that multivariate approaches may (a) significantly enhance flood-prone areas delineation relative to a selected univariate one, (b) provide accurate predictions of expected inundation depths, (c) produce encouraging results in extrapolation, (d) complete the information of imperfect reference maps, and (e) conveniently convert binary maps into continuous representation of FH.

Nonlinear modelling of non-structural masonry walls under a low-intensity earthquake motion: calibration through a parametric study

This thesis aims to understand the behavior of a low-rise unreinforced masonry building (URM), the typical residential house in the Netherlands, when subjected to low-intensity earthquakes. In fact, in the last decades, the Groningen region was hit by several shallow earthquakes caused by the extraction of natural gas. In particular, the focus is addressed to the internal non-structural walls and to their interaction with the structural parts of the building. A simple and cost-efficient 2D FEM model is developed, focused on the interfaces representing mortar layers that are present between the non-structural walls and the rest of the structure. As a reference for geometries and materials, it has been taken into consideration a prototype that was built in full-scale at the EUCENTRE laboratory of Pavia (Italy). Firstly, a quasi-static analysis is performed by gradually applying a prescribed displacement on the roof floor of the structure. Sensitivity analyses are conducted on some key parameters characterizing mortar. This analysis allows for the calibration of their values and the evaluation of the reliability of the model. Successively, a transient analysis is performed to effectively subject the model to a seismic action and hence also evaluate the mechanical response of the building over time. Moreover, it was possible to compare the results of this analysis with the displacements recorded in the experimental tests by creating a model representing the entire considered structure. As a result, some conditions for the model calibration are defined. The reliability of the model is then confirmed by both the reasonable results obtained from the sensitivity analysis and the compatibility of the values obtained for the top displacement of the roof floor of the experimental test, and the same value acquired from the structural model.

Energy consumption of parallel algorithms for solving linear systems on HPC architecture

Modern High-Performance Computing HPC systems are gradually increasing in size and complexity due to the correspondent demand of larger simulations requiring more complicated tasks and higher accuracy. However, as side effects of the Dennard’s scaling approaching its ultimate power limit, the efficiency of software plays also an important role in increasing the overall performance of a computation. Tools to measure application performance in these increasingly complex environments provide insights into the intricate ways in which software and hardware interact. The monitoring of the power consumption in order to save energy is possible through processors interfaces like Intel Running Average Power Limit RAPL. Given the low level of these interfaces, they are often paired with an application-level tool like Performance Application Programming Interface PAPI. Since several problems in many heterogeneous fields can be represented as a complex linear system, an optimized and scalable linear system solver algorithm can decrease significantly the time spent to compute its resolution. One of the most widely used algorithms deployed for the resolution of large simulation is the Gaussian Elimination, which has its most popular implementation for HPC systems in the Scalable Linear Algebra PACKage ScaLAPACK library. However, another relevant algorithm, which is increasing in popularity in the academic field, is the Inhibition Method. This thesis compares the energy consumption of the Inhibition Method and Gaussian Elimination from ScaLAPACK to profile their execution during the resolution of linear systems above the HPC architecture offered by CINECA. Moreover, it also collates the energy and power values for different ranks, nodes, and sockets configurations. The monitoring tools employed to track the energy consumption of these algorithms are PAPI and RAPL, that will be integrated with the parallel execution of the algorithms managed with the Message Passing Interface MPI.

Continual learning on nano drones

Miniaturized flying robotic platforms, called nano-drones, have the potential to revolutionize the autonomous robots industry sector thanks to their very small form factor. The nano-drones’ limited payload only allows for a sub-100mW microcontroller unit for the on-board computations. Therefore, traditional computer vision and control algorithms are too computationally expensive to be executed on board these palm-sized robots, and we are forced to rely on artificial intelligence to trade off accuracy in favor of lightweight pipelines for autonomous tasks. However, relying on deep learning exposes us to the problem of generalization since the deployment scenario of a convolutional neural network (CNN) is often composed by different visual cues and different features from those learned during training, leading to poor inference performances. Our objective is to develop and deploy and adaptation algorithm, based on the concept of latent replays, that would allow us to fine-tune a CNN to work in new and diverse deployment scenarios. To do so we start from an existing model for visual human pose estimation, called PULPFrontnet, which is used to identify the pose of a human subject in space through its 4 output variables, and we present the design of our novel adaptation algorithm, which features automatic data gathering and labeling and on-device deployment. We therefore showcase the ability of our algorithm to adapt PULP-Frontnet to new deployment scenarios, improving the R2 scores of the four network outputs, with respect to an unknown environment, from approximately [−0.2, 0.4, 0.0,−0.7] to [0.25, 0.45, 0.2, 0.1]. Finally we demonstrate how it is possible to fine-tune our neural network in real time (i.e., under 76 seconds), using the target parallel ultra-low power GAP 8 System-on-Chip on board the nano-drone, and we show how all adaptation operations can take place using less than 2mWh of energy, a small fraction of the available battery power.

A Highly Improved Method For Sensitive Determination Of Amitriptyline In Pharmaceutical Formulations Using An Unmodified Carbon Nanotube Electrode In The Presence Of Sulfuric Acid.

The present paper describes a novel, simple and reliable differential pulse voltammetric method for determining amitriptyline (AMT) in pharmaceutical formulations. It has been described for many authors that this antidepressant is electrochemically inactive at carbon electrodes. However, the procedure proposed herein consisted in electrochemically oxidizing AMT at an unmodified carbon nanotube paste electrode in the presence of 0.1 mol L(-1) sulfuric acid used as electrolyte. At such concentration, the acid facilitated the AMT electroxidation through one-electron transfer at 1.33 V vs. Ag/AgCl, as observed by the augmentation of peak current. Concerning optimized conditions (modulation time 5 ms, scan rate 90 mV s(-1), and pulse amplitude 120 mV) a linear calibration curve was constructed in the range of 0.0-30.0 μmol L(-1), with a correlation coefficient of 0.9991 and a limit of detection of 1.61 μmol L(-1). The procedure was successfully validated for intra- and inter-day precision and accuracy. Moreover, its feasibility was assessed through analysis of commercial pharmaceutical formulations and it has been compared to the UV-vis spectrophotometric method used as standard analytical technique recommended by the Brazilian Pharmacopoeia.

Extreme Harmonic Generation In Electrically Driven Spin Resonance.

We report the observation of multiple harmonic generation in electric dipole spin resonance in an InAs nanowire double quantum dot. The harmonics display a remarkable detuning dependence: near the interdot charge transition as many as eight harmonics are observed, while at large detunings we only observe the fundamental spin resonance condition. The detuning dependence indicates that the observed harmonics may be due to Landau-Zener transition dynamics at anticrossings in the energy level spectrum.

Magnesium-deficient High-fat Diet: Effects On Adiposity, Lipid Profile And Insulin Sensitivity In Growing Rats.

To determine if magnesium deficiency aggravates the effects of a high-fat diet in growing rats in terms of obesity, lipid profile and insulin resistance. The study population comprised 48 newly weaned male Wistar Hannover rats distributed into four groups according to diet, namely, control group (CT; n = 8), control diet provided ad libitum; pair-feeding control group (PF; n = 16), control diet but in the same controlled amount as animals that received high-fat diets; high-fat diet group (HF; n = 12), and magnesium-deficient high-fat diet group (HFMg(-); n = 12). The parameters investigated were adiposity index, lipid profile, magnesium status, insulin sensitivity and the phosphorylation of proteins involved in the insulin-signaling pathway, i.e. insulin receptor β-subunit, insulin receptor substrate 1 and protein kinase B. The HF and HFMg(-) groups were similar regarding gain in body mass, adiposity index and lipid profile, but were significantly different from the PF group. The HFMg(-) group exhibited alterations in magnesium homeostasis as revealed by the reduction in urinary and bone concentrations of the mineral. No inter-group differences were observed regarding glucose homeostasis. Protein phosphorylation in the insulin-signaling pathway was significantly reduced in the high-fat groups compared with the control groups, demonstrating that the intake of fat-rich diets increased insulin resistance, a syndrome that was aggravated by magnesium deficiency. Under the experimental conditions tested, the intake of a magnesium-deficient high-fat diet led to alterations in the insulin-signaling pathway and, consequently, increased insulin resistance.

Interplay Between Cell Growth And Cell Cycle In Plants.

The growth of organs and whole plants depends on both cell growth and cell-cycle progression, but the interaction between both processes is poorly understood. In plants, the balance between growth and cell-cycle progression requires coordinated regulation of four different processes: macromolecular synthesis (cytoplasmic growth), turgor-driven cell-wall extension, mitotic cycle, and endocycle. Potential feedbacks between these processes include a cell-size checkpoint operating before DNA synthesis and a link between DNA contents and maximum cell size. In addition, key intercellular signals and growth regulatory genes appear to target at the same time cell-cycle and cell-growth functions. For example, auxin, gibberellin, and brassinosteroid all have parallel links to cell-cycle progression (through S-phase Cyclin D-CDK and the anaphase-promoting complex) and cell-wall functions (through cell-wall extensibility or microtubule dynamics). Another intercellular signal mediated by microtubule dynamics is the mechanical stress caused by growth of interconnected cells. Superimposed on developmental controls, sugar signalling through the TOR pathway has recently emerged as a central control point linking cytoplasmic growth, cell-cycle and cell-wall functions. Recent progress in quantitative imaging and computational modelling will facilitate analysis of the multiple interconnections between plant cell growth and cell cycle and ultimately will be required for the predictive manipulation of plant growth.

Classification Of Frankfurters By Ft-raman Spectroscopy And Chemometric Methods.

Frankfurters are widely consumed all over the world, and the production requires a wide range of meat and non-meat ingredients. Due to these characteristics, frankfurters are products that can be easily adulterated with lower value meats, and the presence of undeclared species. Adulterations are often still difficult to detect, due the fact that the adulterant components are usually very similar to the authentic product. In this work, FT-Raman spectroscopy was employed as a rapid technique for assessing the quality of frankfurters. Based on information provided by the Raman spectra, a multivariate classification model was developed to identify the frankfurter type. The aim was to study three types of frankfurters (chicken, turkey and mixed meat) according to their Raman spectra, based on the fatty vibrational bands. Classification model was built using partial least square discriminant analysis (PLS-DA) and the performance model was evaluated in terms of sensitivity, specificity, accuracy, efficiency and Matthews's correlation coefficient. The PLS-DA models give sensitivity and specificity values on the test set in the ranges of 88%-100%, showing good performance of the classification models. The work shows the Raman spectroscopy with chemometric tools can be used as an analytical tool in quality control of frankfurters.

Evaluation Of Soft-tissue Lesions With (18)f-fdg Pet/ct: Initial Results Of A Prospective Trial.

Although MRI is utilized for planning the resection of soft-tissue tumors, it is not always capable of differentiating benign from malignant lesions. The risk of local recurrence of soft-tissue sarcomas is increased when biopsies are performed before resection and by inadequate resections. PET associated with computed tomography using fluorodeoxyglucose labeled with fluorine-18 ((18)F-FDG PET/CT) may help differentiate between benign and malignant tumors, thus avoiding inadequate resections and making prior biopsies unnecessary. The purpose of this study was to evaluate the usefulness of (18)F-FDG PET/CT in differentiating benign from malignant solid soft-tissue lesions. Patients with solid lesions of the limbs or abdominal wall detected by MRI were submitted to (18)F-FDG PET/CT. The maximum standardized uptake value (SUVmax) cutoff was determined to differentiate malignant from benign tumors. Regardless of the (18)F-FDG PET/CT results all patients underwent biopsy and surgery. MRI was performed in 54 patients, and 10 patients were excluded because of purely lipomatose or cystic lesions. (18)F-FDG PET/CT was performed in the remaining 44 patients. Histopathology revealed 26 (59%) benign and 18 (41%) malignant soft-tissue lesions. A significant difference in SUVmax was observed between benign and malignant soft-tissue lesions. The SUVmax cutoff of 3.0 differentiated malignant from benign lesions with 100% sensitivity, 83.3% specificity, 89.6% accuracy, 78.3% positive predictive value, and 100% negative predictive value. (18)F-FDG PET/CT seems to be able to differentiate benign from malignant soft-tissue lesions with good accuracy and very high negative predictive value. Incorporating (18)F-FDG PET/CT into the diagnostic algorithm of these patients may prevent inadequate resections and unnecessary biopsies.

Comparison Of Different Transfer Impression Techniques Accuracy For Osseointegrated Implants.

Abstract The aim of this study was to evaluate three transfer techniques used to obtain working casts of implant-supported prostheses through the marginal misfit and strain induced to metallic framework. Thirty working casts were obtained from a metallic master cast, each one containing two implant analogues simulating a clinical situation of three-unit implant-supported fixed prostheses, according to the following transfer impression techniques: Group A, squared transfers splinted with dental floss and acrylic resin, sectioned and re-splinted; Group B, squared transfers splinted with dental floss and bis-acrylic resin; and Group N, squared transfers not splinted. A metallic framework was made for marginal misfit and strain measurements from the metallic master cast. The misfit between metallic framework and the working casts was evaluated with an optical microscope following the single-screw test protocol. In the same conditions, the strain was evaluated using strain gauges placed on the metallic framework. The data was submitted to one-way ANOVA followed by the Tukey's test (α=5%). For both marginal misfit and strain, there were statistically significant differences between Groups A and N (p<0.01) and Groups B and N (p<0.01), with greater values for the Group N. According to the Pearson's test, there was a positive correlation between the variables misfit and strain (r=0.5642). The results of this study showed that the impression techniques with splinted transfers promoted better accuracy than non-splinted one, regardless of the splinting material utilized.

Evaluation Of Red Cell And Reticulocyte Parameters As Indicative Of Iron Deficiency In Patients With Anemia Of Chronic Disease.

The purpose of this study was to evaluate the effectiveness of mature red cell and reticulocyte parameters under three conditions: iron deficiency anemia, anemia of chronic disease, and anemia of chronic disease associated with absolute iron deficiency. Peripheral blood cells from 117 adult patients with anemia were classified according to iron status, and inflammatory activity, and the results of a hemoglobinopathy investigation as: iron deficiency anemia (n=42), anemia of chronic disease (n=28), anemia of chronic disease associated with iron deficiency anemia (n=22), and heterozygous β thalassemia (n=25). The percentage of microcytic red cells, hypochromic red cells, and levels of hemoglobin content in both reticulocytes and mature red cells were determined. Receiver operating characteristic analysis was used to evaluate the accuracy of the parameters in differentiating between the different types of anemia. There was no significant difference between the iron deficient group and anemia of chronic disease associated with absolute iron deficiency in respect to any parameter. The percentage of hypochromic red cells was the best parameter to discriminate anemia of chronic disease with and without absolute iron deficiency (area under curve=0.785; 95% confidence interval: 0.661-0.909, with sensitivity of 72.7%, and specificity of 70.4%; cut-off value 1.8%). The formula microcytic red cells minus hypochromic red cells was very accurate in differentiating iron deficiency anemia and heterozygous β thalassemia (area under curve=0.977; 95% confidence interval: 0.950-1.005; with sensitivity of 96.2%, and specificity of 92.7%; cut-off value 13.8). The indices related to red cells and reticulocytes have a moderate performance in identifying absolute iron deficiency in patients with anemia of chronic disease.

Strain Gauges's Analysis On Implant-retained Prosthesis' Cast Accuracy.

A proper cast is essential for a successful rehabilitation with implant prostheses, in order to produce better structures and induce less strain on the implants. The aim of this study was to evaluate the precision of four different mold filling techniques and verify an accurate methodology to evaluate these techniques. A total of 40 casts were obtained from a metallic matrix simulating three unit implant-retained prostheses. The molds were filled using four different techniques in four groups (n = 10): Group 1 - Single-portion filling technique; Group 2 - Two-step filling technique; Group 3 - Latex cylinder technique; Group 4 - Joining the implant analogs previously to the mold filling. A titanium framework was obtained and used as a reference to evaluate the marginal misfit and tension forces in each cast. Vertical misfit was measured with an optical microscope with an increase of 120 times following the single-screw test protocol. Strain was quantified using strain gauges. Data were analyzed using one-way ANOVA (Tukey's test) (α =0.05). The correlation between strain and vertical misfit was evaluated by Pearson test. The misfit values did not present statistical difference (P = 0.979), while the strain results showed statistical difference between Groups 3 and 4 (P = 0.027). The splinting technique was considered to be as efficient as the conventional technique. The strain gauge methodology was accurate for strain measurements and cast distortion evaluation. There was no correlation between strain and marginal misfit.

Urothelial Carcinogen Resistance Driven By Stronger Toll-like Receptor 2 (tlr2) And Uroplakin Iii (up Iii) Defense Mechanisms: A New Model.

The objective of the study was to illustrate the applicability and significance of the novel Lewis urothelial cancer model compared to the classic Fisher 344. Fischer 344 and Lewis females rats, 7 weeks old, were intravesical instilled N-methyl-N-nitrosourea 1.5 mg/kg every other week for a total of four doses. After 15 weeks, animals were sacrificed and bladders analyzed: histopathology (tumor grade and stage), immunohistochemistry (apoptotic and proliferative indices) and blotting (Toll-like receptor 2-TLR2, Uroplakin III-UP III and C-Myc). Control groups received placebo. There were macroscopic neoplastic lesions in 20 % of Lewis strain and 70 % of Fischer 344 strain. Lewis showed hyperplasia in 50 % of animals, normal bladders in 50 %. All Fischer 344 had lesions, 20 % papillary hyperplasia, 30 % dysplasia, 40 % neoplasia and 10 % squamous metaplasia. Proliferative and apoptotic indices were significantly lower in the Lewis strain (p < 0.01). The TLR2 and UP III protein levels were significantly higher in Lewis compared to Fischer 344 strain (70.8 and 46.5 % vs. 49.5 and 16.9 %, respectively). In contrast, C-Myc protein levels were significantly higher in Fischer 344 (22.5 %) compared to Lewis strain (13.7 %). The innovative Lewis carcinogen resistance urothelial model represents a new strategy for translational research. Preservation of TLR2 and UP III defense mechanisms might drive diverse urothelial phenotypes during carcinogenesis in differently susceptible individuals.