The use of games to help children eyes testing

A doctor cannot perform a good diagnosis without the patient collaboration. One of the major problem for ophthalmologists with children is to capture and maintain their attention while performing their tests. Sometimes children give wrong answers, or not accurate, since they are no longer interested in the task. In this paper we use the serious game paradigm to help children eyes testing. We ask the children to perform a vision acuity and a daltonism test using our game PlayWithEyes. Children have to recognize symbols projected on a wall and point them in a touch interface which displays all the possible answers. Tests performed in a kindergarten with 65 children have shown that the use of our game helps to obtain children cooperation because they have fun, so their attention may last longer, thus improving the possibility to perform a correct diagnosis especially for very young children. This is particular important for some sight defects like amblyopia (lazy eye). Moreover, the system allows us to identify visual acuity reduction in two children. Tests also highlight some limitations of the tool which have been promptly fixed.     

Optimal sampling for the estimation of dispersion parameters in soil columns using an Iterative Genetic Algorithm

In groundwater modelling, an appropriate soil characterization is strongly recommended to evaluate both the fate and transport of solutes and the performance of groundwater remediation criterion, though parameter estimation techniques are often blocked by several inherent difficulties (i.e. ill-posedness and insufficient quantity and quality of observation data). In this paper, an iterative decision model is built and tested in order to locate the position of a fixed number of sample points in a soil column experiment to obtain optimal parameter estimation (OPE), minimizing the parametric uncertainty and the overall cost of the experimental campaign. Starting from an initial guess of chosen points (given by a fraction of the total disposable ones), an Iterative Genetic Algorithm (IGA) is capable of finding the best points able to minimize a first-order approximation of the parameter covariance matrix. The parameter estimates are updated under a Bayesian scheme, using exclusively the observations collected after the earlier run of minimization, and the iterative process stops when the imposed convergence criterion based on the parameter values is reached. An important contribution of this work is the development of an effective direct search algorithm (IGA) for solving the sampling network optimization problem at a laboratory scale.     

Analysis and verification of SMIL documents

In this paper, we consider the problem of automatic verification of SMIL documents and present a tool which can assist the user in the complex task of authoring a multimedia presentation. The tool is based on a formal semantics defining the temporal aspects of SMIL elements by means of a set of inference rules. The rules, in the spirit of Hoare’s semantics, describe how the execution of a piece of code changes the state of the computation of a player. If any temporal conflict is found, the system returns a message to the user pointing out the element which contains the conflict and its motivation. This helps the user to develop robust and clear code.     

Modelling Synchronized Hypermedia Presentations

This paper presents a synchronization model for hypermedia presentations. Several media, continuous, like video and audio files, and non-continuous, like text pages and images, are delivered separately in a distributed environment like the World Wide Web, and presented to the user in a coordinated way.The model is based on a set of synchronization relationships which define media behavior during presentation playback, channels in which to play them, and the effects of user interaction. The model is suited for a wide range of applications, among which self and distance education, professional training, Web advertising, cultural heritage promotion and news-on-demand are good representatives. The model is formally described in terms of changes in the presentation state due to media-related events. A virtual exhibition is analyzed as a test bed to validate the model.     

Quantification of the Immune Content in Neuroblastoma: Deep Learning and Topological Data Analysis in Digital Pathology

We introduce here a novel machine learning (ML) framework to address the issue of the quantitative assessment of the immune content in neuroblastoma (NB) specimens. First, the EUNet, a U-Net with an EfficientNet encoder, is trained to detect lymphocytes on tissue digital slides stained with the CD3 T-cell marker. The training set consists of 3782 images extracted from an original collection of 54 whole slide images (WSIs), manually annotated for a total of 73,751 lymphocytes. Resampling strategies, data augmentation, and transfer learning approaches are adopted to warrant reproducibility and to reduce the risk of overfitting and selection bias. Topological data analysis (TDA) is then used to define activation maps from different layers of the neural network at different stages of the training process, described by persistence diagrams (PD) and Betti curves. TDA is further integrated with the uniform manifold approximation and projection (UMAP) dimensionality reduction and the hierarchical density-based spatial clustering of applications with noise (HDBSCAN) algorithm for clustering, by the deep features, the relevant subgroups and structures, across different levels of the neural network. Finally, the recent TwoNN approach is leveraged to study the variation of the intrinsic dimensionality of the U-Net model. As the main task, the proposed pipeline is employed to evaluate the density of lymphocytes over the whole tissue area of the WSIs. The model achieves good results with mean absolute error 3.1 on test set, showing significant agreement between densities estimated by our EUNet model and by trained pathologists, thus indicating the potentialities of a promising new strategy in the quantification of the immune content in NB specimens. Moreover, the UMAP algorithm unveiled interesting patterns compatible with pathological characteristics, also highlighting novel insights into the dynamics of the intrinsic dataset dimensionality at different stages of the training process. All the experiments were run on the Microsoft Azure cloud platform.     

Experiments on basic thermalhydraulic phenomena relevant for LWR containments: Gas mixing and transport induced by buoyant jets in a multi-compartment geometry

In the frame of the OECD SETH project twenty-four experiments have been carried out on basic gas transport and mixing phenomena in the large-scale, multi-compartment PANDA facility. The experiments consist of several series based on the flow configuration driving the gas transport including plume and jet injection of saturated or superheated steam. A variety of test configurations with well-defined initial and boundary conditions have been investigated, including the parametric effects of the geometry of the injection, the composition and velocity of the injected mixture, the initial ambient composition, the vent location and the initial temperatures. The tests have been designed to be used for the validation of computer codes that are capable of analyzing the thermalhydraulic safety of the containment of light water reactors (LWRs). The results obtained form an extensive database especially valuable for the assessment of the capabilities of three-dimensional simulation tools. The present work reports on a specific test series of the aforementioned experimental program, the so-called Free-Plume Series featuring vertical injection of buoyant steam driving the gas transport. The experimental data is presented showing its quality and the trends in the experiments are analyzed. The phenomena whose prediction might be challenging for codes are emphasized. After the expiration of the confidentiality period, the experimental data is now available for the broad scientific community to be used for code validation purposes.