Similar to other programming models, compilers for SYCL, the open programming model for heterogeneous computing based on C++, would benefit from access to higher-level intermediate representations. The loss of high-level structure and semantics caused by premature lowering to low-level intermediate representations and the inability to reason about host and device code simultaneously present major challenges for SYCL compilers. The MLIR compiler framework, through its dialect mechanism, allows to model domain-specific, high-level intermediate representations and provides the necessary facilities to address these challenges. This work therefore describes practical experience with the design and implementation of an MLIR-based SYCL compiler. By modeling key elements of the SYCL programming model in host and device code in the MLIR dialect framework, the presented approach enables the implementation of powerful device code optimizations as well as analyses across host and device code. Compared to two LLVM-based SYCL implementations, this yields speedups of up to 4.3x on a collection of SYCL benchmark applications. Finally, this work also discusses challenges encountered in the design and implementation and how these could be addressed in the future.
The ROS 2 Navigation Stack (Nav2) has emerged as a widely used software component providing the underlying basis to develop a variety of high-level functionalities. However, when used in outdoor environments such as orchards and vineyards, its functionality is notably limited by the presence of obstacles and/or situations not commonly found in indoor settings. One such example is given by tall grass and weeds that can be safely traversed by a robot, but that can be perceived as obstacles by LiDAR sensors, and then force the robot to take longer paths to avoid them, or abort navigation altogether. To overcome these limitations, domain specific extensions must be developed and integrated into the software pipeline. This paper presents a new, lightweight approach to address this challenge and improve outdoor robot navigation. Leveraging the multi-scale nature of the costmaps supporting Nav2, we developed a system that using a depth camera performs pixel level classification on the images, and in real time injects corrections into the local cost map, thus enabling the robot to traverse areas that would otherwise be avoided by the Nav2. Our approach has been implemented and validated on a Clearpath Husky and we demonstrate that with this extension the robot is able to perform navigation tasks that would be otherwise not practical with the standard components.
Magnetic Resonance imaging based Electrical Properties Tomography (MR-EPT) is a non-invasive technique that measures the electrical properties (EPs) of biological tissues. In this work, we present and numerically investigate the performance of an unrolled, physics-assisted method for 2D MR-EPT reconstructions, where a cascade of Convolutional Neural Networks is used to compute the contrast update. Each network takes in input the EPs and the gradient descent direction (encoding the physics underlying the adopted scattering model) and returns as output the updated contrast function. The network is trained and tested in silico using 2D slices of realistic brain models at 128 MHz. Results show the capability of the proposed procedure to reconstruct EPs maps with quality comparable to that of the popular Contrast Source Inversion-EPT, while significantly reducing the computational time.
In this paper we propose a novel approximate floating-point divider based on bidimensional linear approximation. In our approach, the mantissa quotient is seen as a function of the two input mantissas of the divider. The domain of this two-variable function is partitioned into $nx ny$ subregions, named tiles, where $nx, ny$ are chosen as powers of two. In each tile the quotient is approximated with a linear combination of the input mantissas. To achieve fine accuracy, an optimization problem is formulated within each tile to determine the optimal coefficients for the linear combination, which minimize the Mean Relative Error Distance (MRED) of the divider. Furthermore, to make hardware implementation more effective, the minimization problem is appropriately modified to search for optimal quantized coefficients. The hardware structure of the divider only requires a small look-up table to store the linear approximation coefficients, and a carry save adder tree. The proposed architecture is highly tunable at design-time over a wide range of accuracy, depending on the number of tiles chosen for the approximation. The obtained results demonstrate error performance and hardware features superior to the state-of-the-art. The proposed dividers define the Pareto front, considering the trade-off between power-delay-product vs. MRED and area-delay-product vs. MRED, for MRED in the range of $4 10^{-3}-2 10^{-2}$ . Application results for JPEG compression and tone mapping further highlight the strength of our proposal, which exhibits Structural Similarity Index (SSIM) very close to 1 in all cases and Peak Signal-to-Noise Ratio (PSNR) up to 45 dB.
Abstract Understanding what drives the evolution of nectar‐robbing strategies is key for gaining insight into the functioning of pollination networks. However, nectar robbing is often an anecdotal behaviour, difficult to quantify and record through field observations, especially in hummingbirds, limiting our understanding of how ecological networks change across communities. Here, we report new records of nectar robbing by Peru's endemic Black Metaltail ( Metallura phoebe ) in a high‐elevation forest at ca. 4000 m a.s.l. and how this species uses either legitimate pollination feeding or nectar robbing in relation to corolla lengths. Furthermore, by analysing 452 citizen science records of photographic observations, we found that 36% of the photographs depicting a foraging event in this species were actually nectar‐robbing events. After identifying the plant species in all photographs involving foraging events, we describe how nectar robbing conducted by this hummingbird species is strongly associated with flowers that have longer corollas. We propose that the hummingbird‐flower interactions in harsh high‐altitude environments, where resources and competition vary markedly across seasons, can offer insight into the ecological drivers of nectar‐robbing behaviour in hummingbirds.
Background The minimally invasive infrapubic approach (MIIA) for inflatable penile prosthesis (IPP) placement has shown favorable peri‐operative safety and efficacy profile, but scarce data exist on long‐term follow‐up. Objectives We investigated the safety and efficacy of IPP implantation via the MIIA after a minimum 5‐year follow‐up. Materials and methods We identified data of implanted patients prospectively included in our institutional database. Complications and functional outcomes were assessed by using validated tools. Specifically, quality of life and patient satisfaction were evaluated by the Quality of Life and Sexuality with Penile Prosthesis (QoLSPP) questionnaire. Kaplan–Meier method was used to analyze IPP survival (defined as a working IPP). Results Overall, 67 patients implanted by MIIA with a median (IQR) age of 64 years (61−70) were included. The median (IQR) follow‐up duration was 71 months (63−80). Fifteen (22%) patients experienced complications: minor (Clavien ≤2) events included changes in penile sensitivity ( n = 1; 1.5%), orgasmic dysfunction ( n = 1; 1.5%), pain ( n = 5; 7%), urinary tract infection ( n = 2; 3%), and chronic discomfort ( n = 1; 1.5%); major (Clavien 3) complications were represented by mechanical failure ( n = 3; 4.5%), IPP infection ( n = 1; 1.5%), and cylinder protrusion ( n = 1; 1.5%). The estimated IPP survival was 94% (95% CI, 91.4−96.6), 92.5% (95% CI, 89.7−95.3), and 92.5% (95% CI, 89.7−95.3) at 3, 5, and 7 years after implantation, respectively. In patients using the device at follow‐up ( n = 61; 91%), median (IQR) scores for QoLSPP domains demonstrated favorable functional outcomes and patient satisfaction: functional 21 (19−23), personal 16 (15−18), relational 14 (12−15), and social 12 (11−14). Discussion and conclusion This study represents the longest follow‐up using validated tools to assess the outcomes of IPP implantation via MIIA so far. IPP placement via MIIA confirms to be safe and to offer high satisfaction to both patients and partners at mid‐term evaluation.
Across research on incarceration and insularity, recent trajectories have challenged conventional understandings of islands and prisons as isolated, autonomous units. Instead, they have directed attention toward their capacity to establish relations, circuits and routes. Beyond the focus on mobility and exchanges, this literature criticised the association between insularity, incarceration and confinement as the outcome of specific colonial epistemologies. This article builds on these literatures to investigate the case of Sardinia, a large Mediterranean island that plays a key role in the Italian carceral regime by providing a destination for thousands of Italian convicts. Despite being an Italian and European region, Sardinia's past and present exhibit distinct colonial qualities, which are visible in the structure of its carceral estate. The case study exemplifies how islands can hardly be interpreted as isolated units, given that their histories and geographies have implications that extend far beyond their coasts. In the case of Sardinia, the island combines modern penitentiaries with dynamics reminiscent of previous periods in the history of punishment, specifically penal colonies and convict transportation. This case study illustrates the benefits of an interdisciplinary approach that is willing to question the conceptual categories adopted, particularly those of island, prison and colony.
Annual influenza vaccination is one of the main public health measures able to drastically reduce the burden of this infectious disease. Some evidence suggests ‘trained immunity’ triggered by influenza vaccine might reduce the risk of SARS‐CoV‐2 infection. Adjuvanted influenza vaccines are known to induce a broader cross‐reactive immunity. No studies investigated the effect of adjuvanted vs . non‐adjuvanted influenza vaccines on the risk of symptomatic SARS‐CoV‐2 infection. A case–control analysis nested in a cohort of subjects aged ≥65 years and immunized with adjuvanted or non‐adjuvanted influenza vaccines was conducted. Although no statistically significant (OR = 0.87; P = .082) difference between the two vaccine types was observed for the 9‐month follow‐up period, a 17% (OR = 0.83; P = .042) reduction in the odds of COVID‐19 was observed for adjuvanted vaccines with a 6‐month follow‐up. Further evidence is needed, but these results might have implications given the complexity of the upcoming winter seasons, in which the co‐occurrence of influenza, SARS‐CoV‐2 and other respiratory infections (e.g., syncytial virus) might be unpredictable.
Testing serum C‐reactive protein (CRP) levels can help determine whether there is a need for antibiotics and can limit prescribing of antibiotics for illnesses that are likely viral or non‐infectious in origin. Using Health Search, an Italian primary care database, we identified all patients, aged 15 years or older, being registered in the period between 1 January 2000 and 31 December 2019 and newly diagnosed with upper respiratory tract infections (URTIs) or COPD‐ or asthma‐related exacerbations. From the date of these diagnoses, patients were followed up until occurrence of antibiotic prescription (for these indications) up to 31 December 2019. The association between the CRP testing and the outcome was investigated using a nested case–control analysis. In a cohort of 469 684 patients being diagnosed for URTI (83%), COPD‐ (11%) and asthma (7%)‐related exacerbations, 28 688 (6.11%) were prescribed with antibiotics because of the aforementioned indications. Of note, 98% of cases, nominally those prescribed with antibiotics, were not tested with CRP. However, those receiving antibiotics were more likely to have been previously tested for CRP than controls who did not receive antibiotics (833/28 601 [3%] and 4128/277 968 [1.5%]; OR 2.0 [95% CI: 1.8–2.1]). Our findings indicate that most of the antibiotic prescriptions for the investigated conditions were given without any prior CRP testing. A small minority of GPs did properly use CRP to determine whether antibiotics were needed. Further guidance is needed in Italy on the use of CRP in guiding antibiotic prescribing in primary care.