75 research outputs found
Potential dual synergy between electrochemotherapy and sequence of immunotherapies in metastatic melanoma: A case report
No full textImmune checkpoint inhibitors have changed the natural history of advanced melanoma. Despite this, a notable proportion of patients immediately relapse or develop resistance during immunotherapy, especially with the appearance of superficial metastases and consequently with a dramatic impact on clinical outcomes. Local treatment by electrochemotherapy (ECT), parallel to regional control with palliative aim, seems to release neoantigens potentially determining a significant systemic anticancer immune reactivation. The present study reported a case of a patient with metastatic melanoma receiving Pembrolizumab, electrochemotherapy and then Ipilimumab for in-transit and finally locoregional lymph nodes and distant bone metastases with experience of clinic-radiological remission. Specifically, the present patient progressed during adjuvant treatment with in-transit metastases on the scalp; he underwent two cycle of ECT obtaining partial and then unexpected and very fast nearly complete response with the Ipilimumab treatment. Concomitantly, he developed grade 4 endocrine adverse events (hypophysitis and diabetes mellitus type I) as immune-related toxicities. At 12 months from ECT the patient is in ECOG Performance Status 0 and he has resumed a regular social life. In our experience, ECT in two administrations increased and accelerated the response of Ipilimumab. The present confirmed its promising contribution in inducing a powerful immune response in order to overcome primary or acquired resistance to immune checkpoint inhibitors such as anti-programmed death antigen-1 drugs
Numerical modelling of existing tunnels subjected to time dependent threats
No full textDealing with time-dependent phenomena during tunnel excavation and management is a complex topic that engineers and geologists need to face worldwide. Threats to tunnel stability are posed by the swelling/creep behaviour of geomaterials. These phenomena occur during the construction of the tunnel but can evolve for a long period after tunnel operation and can jeopardize the tunnel integrity with time. Although the topic of time-dependent behaviour has been repeatedly addressed in the literature, less attention has been paid to methodologies to deal with it during the serviceability of tunnels. Therefore, appropriate and reliable methods are needed to analyse and predict tunnel behaviour during their service life and/or refurbishment processes. This paper will focus on existing tunnels subjected to the time-dependent behaviour of the surrounding ground to identify the methodological approach for designing maintenance works and/or managing the associated risks. Examples from the Authors' experience are used to support the discussion
Influence of agricultural management practices on soil organic carbon stock and distribution in topsoil and subsoil as revealed by a mid-term trial
No full textCover crops and no–tillage are agricultural practices used to improve soil organic carbon (OC) sequestration, although most studies are often limited to examining only the topsoil. In this study the influence of different cropping systems (CONV – integrated management without cover crop and conventional tillage, ORG – organic management with cover and temporary intercropping crops and conventional tillage, and NOTILL – integrated management with cover crops and no–tillage) was evaluated on OC quantity and distribution in topsoil (0–20 cm), midsoil (20–40 cm), and subsoil (40–60 cm) of a 10–year wheat–maize rotation trial. A physical–chemical fractionation was performed to isolate OC among labile (water soluble and particulate OC, WEOC and POC, respectively), stable (OC in sand–size and silt– and clay–size aggregates, SSA and SCA, respectively), and resistant (NaClO oxidation) pools. Further, soil samples were characterised for13C and15N natural abundance, phenols and glomalin (GRSP) contents, and microbial activity. The soil OC stock in the 0–60 cm depth was similar for CONV, ORG and NOTILL. This was attributed to enhanced mineralisation processes promoted by the addition of N–rich fresh legume cover crop residues in ORG and NOTILL soils that exceeded the rate of OC stabilisation. However, specific contributions of the functional OC pools to the total stock and along the soil depth intervals occurred. For ORG and NOTILL, the implementation with cover crops favoured the development of a stable macrostructure and the accumulation of OC in SSAs, whereas CONV system mainly accumulated OC in SCAs. When the system was implemented with both cover crops and no–tillage, as for NOTILL, almost half of the total soil OC stock was stored in the topsoil, mostly as POC and associated with SSAs. Regardless of agricultural management, 53–68 % of the total OC stock was found in the layers below the topsoil
A Nonlinear Autoencoder for Kinematic Synergy Extraction from Movement Data Acquired with HTC Vive Trackers
No full textHow the human central nervous system (CNS) copes with the several degrees of freedom (DoF) of the muscle-skeletal system for the generation of complex movements has not been fully understood yet. Many studies in literature have stated that likely the CNS does not independently control DoF but combines few building blocks that consider the synergistic actuation of each DoF. Such building blocks are called synergies. Synergies have been defined both at muscle level, i.e. muscle synergies, and kinematic level, i.e. kinematic synergies. Kinematic synergies consider the synergistic movement of several human articulations during the performance of a complex task, e.g. a reaching-grasping task. The principal component analyses (PCA) is the most used approach in literature for the kinematic synergy extraction. However, the PCA only considers linear correlations among DoFs which can be considered as the most-simple model of inter-joint coupling. In this work, we have extracted synergies from kinematics data (five upper limb angles) acquired during 12 different reaching movements with a tracking system based on the HTC Vive Trackers. After the extraction of the upper-limb joint angles with the OpenSim software, the kinematic synergies have been extracted using nonlinear under-complete autoencoders. Different models of nonlinear autoencoders were investigated and evaluated with R2 index and normalized reconstruction error. The results showed that 4 synergies were enough for describing the 0.973 ± 0.005 (R2 index of log sigmoid model) and 0.979 ± 0.004 (R2 index of tan sigmoid model) of the movement variance for the entire experiment with respectively a Normalized Reconstruction Error (ERMS) of 0.03 ± 0.005 and 0.034 ± 0.004. Comparing the non-linear autoencoders (AE) with the standard linear PCA it emerged that the AE performance are comparable with the PCA results. However, more experiments are needed to perform a deep comparison on a dataset including more joint angles
A framework for automatic Knowledge Base generation from observation data sets
No full textIn the Semantic Web of Everything, observation data collected from sensors and devices disseminated in smart environments must be annotated in order to produce a Knowledge Base (KB) or Knowledge Graph (KG) which can be used subsequently for inference. Available approaches allow defining complex data models for mapping tabular data to KBs/KGs: while granting high flexibility, they can be difficult to use. This paper introduces a framework for automatic KB generation in Web Ontology Language (OWL) 2 from observation data sets. It aims at simplicity both in usage and in expressiveness of generated KBs, in order to enable reasoning with SWoE inference engines in pervasive and embedded devices. An illustrative example from a precision farming case study clarifies the approach and early performance results support its computational sustainability
A Deep Learning Instance Segmentation Approach for Global Glomerulosclerosis Assessment in Donor Kidney Biopsies
No full textThe histological assessment of glomeruli is fundamental for determining if a kidney is suitable for transplantation. The Karpinski score is essential to evaluate the need for a single or dual kidney transplant and includes the ratio between the number of sclerotic glomeruli and the overall number of glomeruli in a kidney section. The manual evaluation of kidney biopsies performed by pathologists is time-consuming and error-prone, so an automatic framework to delineate all the glomeruli present in a kidney section can be very useful. Our experiments have been conducted on a dataset provided by the Department of Emergency and Organ Transplantations (DETO) of Bari University Hospital. This dataset is composed of 26 kidney biopsies coming from 19 donors. The rise of Convolutional Neural Networks (CNNs) has led to a realm of methods which are widely applied in Medical Imaging. Deep learning techniques are also very promising for the segmentation of glomeruli, with a variety of existing approaches. Many methods only focus on semantic segmentation—which consists in segmentation of individual pixels—or ignore the problem of discriminating between non-sclerotic and sclerotic glomeruli, so these approaches are not optimal or inadequate for transplantation assessment. In this work, we employed an end-to-end fully automatic approach based on Mask R-CNN for instance segmentation and classification of glomeruli. We also compared the results obtained with a baseline based on Faster R-CNN, which only allows detection at bounding boxes level. With respect to the existing literature, we improved the Mask R-CNN approach in sliding window contexts, by employing a variant of the Non-Maximum Suppression (NMS) algorithm, which we called Non-Maximum-Area Suppression (NMAS). The obtained results are very promising, leading to improvements over existing literature. The baseline Faster R-CNN-based approach obtained an F-Measure of 0.904 and 0.667 for non-sclerotic and sclerotic glomeruli, respectively. The Mask R-CNN approach has a significant improvement over the baseline, obtaining an F-Measure of 0.925 and 0.777 for non-sclerotic and sclerotic glomeruli, respectively. The proposed method is very promising for the instance segmentation and classification of glomeruli, and allows to make a robust evaluation of global glomerulosclerosis. We also compared Karpinski score obtained with our algorithm to that obtained with pathologists’ annotations to show the soundness of the proposed workflow from a clinical point of view
liBERTa: Local Intelligence via Browser Extensions for Real-Time Applications
No full textAs an application and service platform, the World Wide Web spans from simple informational websites to rich social media and Software-as-a-Service (SaaS) clients. While innovative capabilities are increasingly provided by Deep Learning (DL) Artificial Intelligence (AI) architectures such as pre-trained trans- formers, so far Web applications and services have integrated them only via cloud-based implementations. Deep-Learning-as- a-Service (DLaaS) is establishing itself for professional and personal use, with prevalent business models including pay-per- use and monthly subscriptions. With growing concerns over data privacy, response latency, and service costs, executing DL inference directly within the user’s browser appears as a com- pelling alternative to cloud-based solutions. This paper introduces local intelligence via Browser Extension for Real-Time applications (liBERTa), a modular browser extension-based architecture for real-time client-side DL inference. By operating entirely within the browser, liBERTa reduces reliance on external servers. Its modular design consists of independent layers for data extraction, model inference, and results presentation, granting flexibility and adaptability across different kinds of applications and services. Experimental results from a case study on website privacy policy classification demonstrate the feasibility of the approach, showing that lightweight transformer models can achieve competitive accuracy while maintaining inference times suitable for real- world use on commodity hardware
- …
