Search

Accurate vehicle localization at the level of individual lanes is crucial to ensure the safe and efficient operation of autonomous vehicles, serving as a cornerstone for the development of future Advanced Driving Assistance Systems (ADAS). Contemporary localization methods relying on Global Navigation Satellite Systems (GNSS) often fall short of achieving the necessary precision, necessitating the involvement of additional systems. These supplementary systems frequently depend on the output of road line detectors, whose performance can be hindered by various factors, including adverse weather conditions and heavy traffic, resulting in noisy or sporadically missing data. This study introduces a probabilistic algorithm designed to precisely estimate the actual lane positioning of a vehicle in the specific context of multi-lane roads, such as highways, without relying on GNSS data. The proposed algorithm is built upon a Hidden Markov Model that exploits the output of a generic line detector, a common component of contemporary driving assistance systems. This model ensures consistent lane localization estimates even when faced with noisy or intermittently missing data. Experiments demonstrate the algorithm's effectiveness, providing a reliable estimate of the vehicle in-lane position in challenging datasets containing highway scenarios with hundreds of lane changes. This contributes to the enhancement of existing literature, achieving an accuracy of 86.71% over a segment exceeding 50 km. These results, improving by almost 10% over our previous efforts, suggest that our approach has the potential to enable new ADAS functionalities and offer a robust localization scheme for use in the context of autonomous driving scene understanding.

Deferasirox (DFX) is used for the management of iron overload (IOL) in many haematological malignancies including myelofibrosis (MF). The ‘RUX-IOL’ study retrospectively collected 69 MF patients treated with ruxolitinib (RUX) and DFX for IOL to assess: safety, efficacy in term of iron chelation response (ICR) and erythroid response (ER), and impact on overall survival of the combination therapy. The RUX–DFX therapy was administered for a median time of 12.4 months (interquartile range 3.1–71.2). During treatment, 36 (52.2%) and 34 (49.3%) patients required RUX and DFX dose reductions, while eight (11.6%) and nine (13.1%) patients discontinued due to RUX- or DFX-related adverse events; no unexpected toxicity was reported. ICR and ER were achieved by 33 (47.8%) and 32 patients (46.4%) respectively. Thirteen (18.9%) patients became transfusion-independent. Median time to ICR and ER was 6.2 and 2 months respectively. Patients achieving an ER were more likely to obtain an ICR also (p = 0.04). In multivariable analysis, the absence of leukocytosis at baseline (p = 0.02) and achievement of an ICR at any time (p = 0.02) predicted improved survival. In many MF patients, the RUX–DFX combination provided ICR and ER responses that correlated with improved outcome in the absence of unexpected toxicities. This strategy deserves further clinical investigation.

Objective: Alterations in voice intensity and quality may constitute a social life limitation in people with multiple sclerosis (MS), but only 2% of cases receive speech therapy. Especially the Lee Silverman Voice Treatment (LSVT)-Loud is a highly effective intensive method for voice intensity, requiring subjects’ repeated attendance at the clinic. Telerehabilitation may represent a feasible solution to bypass potential barriers related to speech therapy attendance, scaling up the beneficial effects of the treatment to a broader population. The proposed protocol aims to test the feasibility and the pilot efficacy of the LSVT-Loud delivered in telerehabilitation (Tele-LSVT-Loud), compared to the same treatment delivered in the clinic (LSVT-Loud). Methods: A single-blinded, parallel, two-arm, pilot randomized (1:1 ratio) controlled trial will be performed involving 20 people with MS. Patients will be allocated to 4 weeks of Tele-LSVT-Loud by accessing a telerehabilitation platform at home or LSVT-Loud conventionally delivered in the clinic. Feasibility and pilot effectiveness will be evaluated three times: before (T0), after the treatment (T1), and 3-month follow-up (T2). Feasibility measures will include adherence, adverse events, user experience, motivation, engagement, and acceptability. Vocal intensity during a 1-minute monologue will be the primary outcome measure. Secondary outcome measures will be the vocal quality during a 1-minute monologue, sustained /a/ voice intensity, quality and stability, voice use in daily life, voice subjective perception in daily life, and quality of life. Results: Expected results will be (1) high feasibility of Tele-LSVT-Loud and (2) a non-inferiority effect of Tele-LSVT-Loud compared with face-to-face treatment delivery on voice intensity and quality outcomes. Conclusions: Tele-LSVT-Loud may be a feasible intervention for MS alteration in voice intensity and quality with a non-inferior effect compared to LSVT-Loud.

BACKGROUND: Vocal disorders are frequent in people with multiple sclerosis (MS). Cognitive impairment, fatigue, depression, and other clinical characteristics can be associated with treatment effectiveness in rehabilitation. Finding baseline characteristics that identify those who are responding to treatment can help the clinical decision-making process, which can then help improve the effectiveness of voice treatment. We developed a model to identify factors associated with treatment-related improvement on voice intensity in people with MS. METHODS: Data are from a randomized controlled trial of the effects of voice therapy. Forty-four people with MS were enrolled and randomized to receive Lee Silverman Voice Treatment LOUD, specifically addressing voice intensity, or conventional speech-therapy group. Voice intensity (dB) was measured during monologue before and after treatment and was used to differentiate those who responded (posttreatment voice intensity?> 60 dB) from those who did not. Possible associated factors were cognitive impairment, fatigue, depression, disability, and disease duration. Associations were assessed by univariate logistic regression and univariate and multivariate linear regressions. RESULTS: Mean ± SD monologue voice intensity is improved in the whole sample (before rehabilitation: 51.8 ± 4.2 dB; and after rehabilitation 57.0 ± 6.5 dB; P <.001), and 11 people with MS (27.5%) responded to treatment. Specificity of treatment was associated with the return to normal voice intensity (OR, 14.28; 95% CI, 12.17-309.56) and we found a linear association between voice improvement and the specificity of treatment (6.65 [SE = 1.54] dB; P <.05). Moreover, the analysis revealed a nonlinear association between improvement and fatigue, suggesting increased benefits for people with MS with moderate fatigue. Other factors were not significantly associated with treatment effectiveness. CONCLUSIONS: Moderate fatigue and the specificity of the inter

Background: Coronavirus disease 2019 (COVID-19) broke out in China in December 2019 and now is a pandemic all around the world. In Italy, Northern regions were hit the hardest during the first wave. We aim to explore the prevalence and the exposure characteristics of physiotherapists (PTs) working in different Italian regions during the first wave of COVID-19. Methods: Between April and May 2020 a structured anonymous online survey was distributed to all PTs registered in the National Professional Registry to collect prevalence data of a confirmed diagnosis of COVID-19 (i.e., nasopharyngeal swab and/or serological test). A bottom-up agglomerative nesting hierarchical clustering method was applied to identify groups of regions based on response rate. Multivariable logistic regression was used to explore personal and work-related factors associated with a confirmed diagnosis of COVID-19. Results: A total of 15,566 PTs completed the survey (response rate 43.3%). The majority of respondents (57.7%) were from Northern regions. Considering all respondents, the number of confirmed COVID-19 cases in Northern and Central Italy, was higher compared to those in Southern Italy (6.9% vs. 1.8%, P < 0.001); focusing the analysis on respondents who underwent nasopharyngeal swab and/or serological test led to similar findings (14.1% vs. 6.4%, P < 0.001). Working in Northern and Central regions was associated with a higher risk of confirmed diagnosis of COVID-19 compared to Southern regions (OR 3.4, 95%CI 2.6 to 4.3). PTs working in Northern and Central regions were more likely to be reallocated to a different unit and changing job tasks, compared to their colleagues working in the Southern regions (10.5% vs 3.7%, P < 0.001). Conclusions: Work-related risk factors were differently distributed between Italian regions at the time of first pandemic wave, and PTs working in the Northern and Central regions were more at risk of a confirmed diagnosis of COVID-19, especially when w

Point clouds registration is a fundamental step of many point clouds processing pipelines; however, most algorithms are tested on data that are collected ad-hoc and not shared with the research community. These data often cover only a very limited set of use cases; therefore, the results cannot be generalized. Public datasets proposed until now, taken individually, cover only a few kinds of environment and mostly a single sensor. For these reasons, we developed a benchmark, for localization and mapping applications, using multiple publicly available datasets. In this way, we are able to cover many kinds of environment and many kinds of sensor that can produce point clouds. Furthermore, the ground truth has been thoroughly inspected and evaluated to ensure its quality. For some of the datasets, the accuracy of the ground truth measuring system was not reported by the original authors, therefore we estimated it with our own novel method, based on an iterative registration algorithm. Along with the data, we provide a broad set of registration problems, chosen to cover different types of initial misalignment, various degrees of overlap, and different kinds of registration problems. Lastly, we propose a metric to measure the performances of registration algorithms: it combines the commonly used rotation and translation errors together, to allow an objective comparison of the alignments. This work aims at encouraging authors to use a public and shared benchmark, instead of data collected ad-hoc, to ensure objectivity and repeatability, two fundamental characteristics in any scientific field.

Successful discontinuation of tyrosine kinase inhibitors has been achieved in patients with chronic-phase chronic myeloid leukemia (CML). Careful molecular monitoring after discontinuation warrants safe and prompt resumption of therapy. We retrospectively evaluated how molecular monitoring has been conducted in Italy in a cohort of patients who discontinued tyrosine kinase inhibitor (TKI) treatment per clinical practice. The outcome of these patients has recently been reported—281 chronic-phase CML patients were included in this subanalysis. Median follow-up since discontinuation was 2 years. Overall, 2203 analyses were performed, 17.9% in the first three months and 38.4% in the first six months. Eighty-six patients lost major molecular response (MMR) in a mean time of 5.7 months—65 pts (75.6%) during the first six months. We evaluated the number of patients who would experience a delay in diagnosis of MMR loss if a three-month monitoring schedule was adopted. In the first 6 months, 19 pts (29.2%) would have a one-month delay, 26 (40%) a 2-month delay. Very few patients would experience a delay in the following months. A less intense frequency of monitoring, particularly after the first 6 months off treatment, would not have affected the success of treatment-free remission (TFR) nor put patients at risk of progression.

Global localization is an important and widely studied problem for many robotic applications. Place recognition approaches can be exploited to solve this task, e.g., in the autonomous driving field. While most vision-based approaches match an image w.r.t. an image database, global visual localization within LiDAR-maps remains fairly unexplored, even though the path toward high definition 3D maps, produced mainly from LiDARs, is clear. In this work we leverage Deep Neural Network (DNN) approaches to create a shared embedding space between images and LiDAR-maps, allowing for image to 3D-LiDAR place recognition. We trained a 2D and a 3D DNN that create embeddings, respectively from images and from point clouds, that are close to each other whether they refer to the same place. An extensive experimental activity is presented to assess the effectiveness of the approach w.r.t. different learning paradigms, network architectures, and loss functions. All the evaluations have been performed using the Oxford Robotcar Dataset, which encompasses a wide range of weather and light conditions.

Introduction: Fatigue, and balance and gait disorders can impact on physical activity (PA) levels in persons with Multiple Sclerosis (pwMS). To date, several studies have examined PA in pwMS during daily life, but little is known about PA levels in pwMS during a rehabilitation period. The present study investigated PA levels (daily steps, light physical activity (LPA), and moderate-vigorous physical activity (MVPA)) in inpatient and outpatients with MS during their rehabilitation period and described the relationship between objectively measured PA and levels of disability, quality of life, fatigue, and self-efficacy. Methods: In this exploratory cross-sectional study, we examined 40 pwMS during their inpatient or outpatient rehabilitation regime. Participants included in the study (N = 34) wore a Fitbit Versa tracker for one week recording daily steps, minutes of LPA, and minutes of MVPA (primary outcomes). They underwent a clinical assessment of physical activity levels (Godin Leisure-Time Exercise Questionnaire(GLTEQ)), fatigue (Fatigue Severity Scale(FSS)), walking ability (10 Meter Walk Test(10MWT), 2-Min Walk Test(2MWT), 12-item Multiple Sclerosis Walking scale(MSWS-12)), quality of life (12-Item Short Form Survey (SF-12)), and self-efficacy (Self-Efficacy in Multiple Sclerosis scale (SEMS)) (secondary outcomes). Multiple linear regressions (MLR) models were used to test whether the demographic difference between the two groups influenced the estimation of objective variables measured by Fitbit. Finally, correlations between objectively measured physical activity and subjective clinical scales were estimated with Spearman correlations. Results: Our sample consisted of 21 females and 13 males with a mean (interquartile range) age of 52 (20) years and an Expanded Disability Status scale(EDSS) score of 6.0 (1.50) points; baseline characteristics of inpatients (N = 18) and outpatients (N = 16) differed statistically only in EDSS levels (p-value = 0.008) and use of

Aim: The rehabilitation of voice disorders is an unmet need in multiple sclerosis (MS). The Lee Silverman Voice Treatment (LSVT LOUD) is a well-documented and effective speech treatment, developed to treat voice disorders in Parkinson Disease. The purpose of the present study was to examine the viability of applying the LSVT LOUD to individuals with MS and verify short- and long-term improvements in acoustic and perceptual voice parameters. Methods: A single subject design was performed in a consecutive sample of 8 subjects with MS. The subjects’ voice was recorded with PRAAT software for 5 days at baseline during the 16 treatment sessions, and at follow-up (FU) 6/12 months later. PRAAT provided data on sustained /a/ (SPL/a/) voice intensity and maximum phonation time (MPT/a/) of sustained /a/, and on functional sentences voice intensity. In addition, self-assessment questionnaire Voice Handicap Index, the perceptual GIRBAS scale and intensity of monologue were collected at first day of baseline, post-treatment and at FU. In the treatment phase each subject received treatment according to LSVT LOUD protocol. Visual analysis calculated for daily acoustic variables was used to determine baseline stability and analyse changes following treatment. The Wilcoxon test was used to assess statistically significant differences between baseline and post treatment. Results: All participants completed the LSVT LOUD programme; one participant dropped out at FU. Improvements in acoustic analysis were found: SPL/a/ improved on average (± standard deviation) 11.64 ± 4.19 dB with 7 subjects showing statistically significant improvement (P < 0.05); MPT/a/ improved on average 1.2 ± 1.53seconds, while intensity of functional sentences improved on average 8.11 ± 3.46 dB with 4 and 5 subjects showed statistically significant improvement, respectively. Intensity of monologue improved 14.90 ± 3.33 dB. Acoustic values are maintained or increased at FU respect to baseline. All subjects imp

Background: Hypophonia is a prevailing problem in people with multiple sclerosis (PwMS). However, evidence supporting the effectiveness of voice rehabilitation is lacking. Objective: The aim of this study was to identify the most effective method to reduce hypophonia. Methods: In this randomized controlled trial, 44 PwMS were randomized to intensive and high-effort voice treatment groups, the LSVT-LOUD®, and conventional treatment group. Subjects received 16 treatments (4 sessions/week) lasting 45 minutes. The primary outcome was voice intensity (dB) in monologue, vocalization, and sentences while voice handicap index (VHI) measured voice self-perception. Outcomes were assessed by a blinded observer at baseline, post-treatment, and 15-month follow-up (FU). Results: Linear models revealed a significant post-intervention between-group mean difference in favor of LSVT-LOUD for monologue: +6.3 dB (95% CI: 2.5 to 10.1); vocalization: +7.4 dB (95% CI: 2.3 to 12.5); and sentences: +9.5 dB (95% CI: 4.7 to 14.3). However, 43.7% PwMS in the LSVT-LOUD and 10% in the conventional treatment group obtained a full recovery of voice intensity (>60 dB) post-treatment, Fisher’s test = 13.3, p < 0.01. However, these improvements were not maintained at FU. Between-group differences at VHI were −10.8 (95% CI: −21.2 to −0.4) and −11.3 (95% CI: −24.3 to −1.7) in favor of LSVT-LOUD at post and FU. Conclusion: LSVT-LOUD can be a valid treatment to increase voice intensity in PwMS. However, results suggest the need for FU interventions targeting maintenance.

Loop closure detection is an essential component of simultaneous localization and mapping (SLAM) systems, which reduces the drift accumulated over time. Over the years, several deep learning approaches have been proposed to address this task; however, their performance has been subpar compared to handcrafted techniques, especially while dealing with reverse loops. In this article, we introduce the novel loop closure detection network (LCDNet) that effectively detects loop closures in light detection and ranging (LiDAR) point clouds by simultaneously identifying previously visited places and estimating the six degrees of freedom relative transformation between the current scan and the map. LCDNet is composed of a shared encoder, a place recognition head that extracts global descriptors, and a relative pose head that estimates the transformation between two point clouds. We introduce a novel relative pose head based on the unbalanced optimal transport theory that we implement in a differentiable manner to allow for end-to-end training. Extensive evaluations of LCDNet on multiple real-world autonomous driving datasets show that our approach outperforms state-of-the-art loop closure detection and point cloud registration techniques by a large margin, especially while dealing with reverse loops. Moreover, we integrate our proposed loop closure detection approach into a LiDAR SLAM library to provide a complete mapping system and demonstrate the generalization ability using different sensor setup in an unseen city.

It is judged safe to discontinue treatment with tyrosine kinase inhibitors (TKI) for chronic myeloid leukemia (CML) in experimental trials on treatment-free remission (TFR). We collected a total of 293 Italian patients with chronic phase CML who discontinued TKI in deep molecular response. Seventy-two percent of patients were on treatment with imatinib, and 28% with second generation TKI at the time of discontinuation. Median duration of treatment with the last TKI was 77 months [Interquartile Range (IQR) 54;111], median duration of deep molecular response was 46 months (IQR 31;74). Duration of treatment with TKI and duration of deep molecular response were shorter with second generation TKI than with imatinib (P<0.001). Eighty-eight percent of patients discontinued as per clinical practice, and reasons for stopping treatment were: toxicity (20%), pregnancy (6%), and shared decision between treating physician and patient (62%). After a median follow up of 34 months (range, 12-161) overall estimated TFR was 62% (95%CI: 56;68). At 12 months, TFR was 68% (95%CI: 62;74) for imatinib, 73% (95%CI: 64;83) for second generation TKI. Overall median time to restart treatment was six months (IQR 4;11). No progressions occurred. Although our study has the limitation of a retrospective study, our experience within the Italian population confirms that discontinuation of imatinib and second generation TKI is feasible and safe in clinical practice

This paper deals with the task of ego-vehicle localization at intersections, a significant task in autonomous road driving. We propose an online vision-based method that can hence be applied if the intersection is visible. It relies on stereo images and on a coarse street-level pose estimate, used to retrieve intersection data from a digital map service. Pixel-level semantic segmentation, and 3D reconstruction from state-of-the art Deep Neural Networks are coupled with an intersection model; this allows good positioning accuracy compared to the state-of-the-art in this task. To demonstrate the effectiveness of the method and make it possible to compare it with other methods, an extensive activity has been conducted in order to set up a dataset of approaches to an intersection, which has then been used to benchmark the proposed method. The dataset is made available to the community, and it currently includes more than forty intersection approaches, from KITTI. Another important contribution of the paper is the definition of criteria for the comparison of different methods, on recorded datasets. The proposed method achieves nearly sub-meter accuracy in difficult real conditions.

Detecting buildings in the surroundings of an urban vehicle and matching them to building models available on map services is an emerging trend in robotics localization for urban vehicles. In this paper, we present a novel technique, which improves a previous work by detecting building façade, their positions, and finding the correspondences with their 3D models, available in Open-StreetMap. The proposed technique uses segmented point clouds produced using stereo images, processed by a convolutional neural network. The point clouds of the façades are then matched against a reference point cloud, produced extruding the buildings’ outlines, which are available on OpenStreetMap (OSM). In order to produce a lane-level localization of the vehicle, the resulting information is then fed into our probabilistic framework, called Road Layout Estimation (RLE). We prove the effectiveness of this proposal, testing it on sequences from the well-known KITTI dataset and comparing the results concerning a basic RLE version without the proposed pipeline.

Background: Therapeutic drug monitoring (TDM) is a tool to personalize and optimize dosing by measuring the drug concentration and subsequently adjusting the dose to reach a target concentration or exposure. The evidence to support TDM is however often ranked as expert opinion. Limitations in study design and sample size have hampered definitive conclusions of the potential added value of TDM. Objectives: We aim to give expert opinion and discuss the main points and limitations of available data from antibiotic TDM trials and emphasize key elements for consideration in design of future clinical studies to quantify the benefits of TDM. Sources: The sources were peer-reviewed publications, guidelines and expert opinions from the field of TDM. Content: This review focuses on key aspects of antimicrobial TDM study design: describing the rationale for a TDM study, assessing the exposure of a drug, assessing susceptibility of pathogens and selecting appropriate clinical endpoints. Moreover we provide guidance on appropriate study design.

Purpose: The aim of this study was to develop an exercise protocol to improve maximal mouth opening (MMO), tongue protrusion (Tprot), tongue strength (Tstren), and lip strength (Lstren), and to assess its effects on subjects with scleroderma. Methods: We performed four replicated single-system studies in a consecutive sample of subjects with scleroderma. An instrumented assessment measured MMO, Tprot, Tstren, and Lstren. Each day, subjects were assessed and performed orofacial exercises conducted by speech therapists. Treatments were first aimed at improving mouth physical characteristics by impairment-oriented exercises and then to improve skills with function-oriented exercises. Results: The mean phase differences between assessment and treatment phases across subjects were from 0.88 to 9.56 mm in MMO, from 2.03 to 12.3 mm in Tprot, from-0.12 to 5.35 N in Tstren, and from-0.84 to 5.19 N in Lstren. After treatment, 3 subjects crossed the 5th percentile discriminating normal from abnormal performances for both Tstren and Tprot, while this occurred in 2 subjects for MMO and Lstren. Conclusions: The results of this study suggest that rehabilitation appears to be useful in reducing tongue and lip impairments and in improving oral functions in subjects with scleroderma.