Summary Deep learning has gained a significant popularity in recent years thanks to its tremendous success across a wide range of relevant fields of applications, including medical image analysis domain in particular. Although convolutional neural networks (CNNs) based medical applications have been providing powerful solutions and revolutionizing medicine, efficiently training of CNNs models is a tedious and challenging task. It is a computationally intensive process taking long time and rare system resources, which represents a significant hindrance to scientific research progress. In order to address this challenge, we propose in this article, R2D2, a scalable intuitive deep learning toolkit for medical imaging semantic segmentation. To the best of our knowledge, the present work is the first that aims to tackle this issue by offering a novel distributed versions of two well‐known and widely used CNN segmentation architectures [ie, fully convolutional network (FCN) and U‐Net]. We introduce the design and the core building blocks of R2D2. We further present and analyze its experimental evaluation results on two different concrete medical imaging segmentation use cases. R2D2 achieves up to 17.5× and 10.4× speedup than single‐node based training of U‐Net and FCN, respectively, with a negligible, though still unexpected segmentation accuracy loss. R2D2 offers not only an empirical evidence and investigates in‐depth the latest published works but also it facilitates and significantly reduces the effort required by researchers to quickly prototype and easily discover cutting‐edge CNN configurations and architectures.

Effectively training of Convolutional Neural Networks (CNNs) is a computationally intensive and time-consuming task. Therefore, scaling up the training of CNNs has become a key approach to decrease the training duration and train CNN models in a reasonable time. Nevertheless, introducing parallelism to CNNs is a laborious task in practice. It is a manual, repetitive and error-prone process. In this paper, we present Auto-CNNp, a novel framework that aims to address this challenge by automating CNNs training parallelization task. To achieve this goal, the Auto-CNNp introduces a key component which is called CNN-Parallelism-Generator. The latter component aims to streamline routine tasks throughout (1) capturing cumbersome CNNs parallelization tasks within a backbone structure while (2) keeping the framework flexible enough and extensible for user-specific personalization. Our proposed reference implementation provides a high level of abstraction over MPI-based CNNs parallelization process, despite the CNN-based imaging task and its related architecture and training dataset. We introduce the design and the core building blocks of Auto-CNNp. We further conduct an extensive assessment of our proposal that not only shows its effectiveness in accelerating the process of scaling up CNNs training, but also its generalization for a wider variety of use cases.

This paper compares two statistical models of location within a smart flat during the day. The location is then identified with a task executed normally or repeated pathologically, e.g., in case of Alzheimer disease (AD), whereas a task persistence parameter assesses tendency to perseverate. Compared with a Pólya's urns derived approach, the Markovian one is more effective and offers up to 98 percent of good prediction using only the last known location but distinguishing days of week. To extend these results to a multisensor context, some difficulties must be overcome. An external knowledge is made from a set of observable random variables provided by body sensors and organized either in a Bayesian network or in a reference knowledge base system (KBS) containing the person's actimetric profile. When data missed or errors occurred, an estimate of the joint probabilities of these random variables and hence the probability of all events appearing in the network or the KBS was developed and corrects the bias of the Lancaster and Zentgraf classical approach which in certain circumstances provides negative estimates. Finally, we introduce a correction corresponding to a possible loss of the person's synchronization with the nycthemeral (day versus night) zeitgebers (synchronizers) to avoid false alarms.

Free of most social constraints, elderly people tend to perform activities of daily living following the same sequence. This paper proposes a method for medical telesurveillance to detect and quantify a nycthemeral shift in this daily routine. While this common phenomenon is mostly mild, in acute cases, however, it may reveal a pathological behavior requiring a rapid medical examination. This method allows to compare two sequences of activities using the Hamming distance and to interpret the result according to the Gumbel distribution. It may be used to compare either consecutive sequences thereby taking into account evolution in the habits or a sequence to the person's individual activity profile to detect dementia's onset. In this early stage, only elementary activities were considered. That is the reason why location data were used to monitor the person's nycthemeral rhythm of activity. IR sensors placed in her own flat allowed us to follow-up the inhabitant's successive activities. Improvements of this method have already been planned. They include the use of a multi-sensors network to monitor both actimetric (location, movement, posture) and physiological nycthemeral rhythms (ECG, respiratory frequency) and to detect the use of particular items (fridge, chairs, bed). This more sophisticated sensors network will allow us to monitor more complex tasks execution and then to detect pathological behaviors such as perseveration in a task or wandering. On the other hand, multiplying sensors will require more storage capacities and the use of time-consuming data fusion tools. Therefore, a classification phase will be necessary to reduce as possible the number of relevant sensors.

This paper presents an approach of location model deriving from language models existing in speech recognition research. The purpose is to applicate existing model in speech recognition to predict location of an elderly person. Using statistics, the model offers up to 98.03% of good prediction location, considering only the last second of location but distinguishing days of week. Simpler than Polya's urns derived approach, this approach seems to be a good way of location modelling.

If we watch elderly people at home for health or security purposes, we need a reliable pervasive information from a domestic network of localizing sensors allowing to follow the different locations at which the dependent persons can be detected. The data recorded can be treated as the sequence of color coded numbers of balls (symbolizing rooms) taken in a Polya's urn, in which the persistence of the presence in a room is taken into account by adding a certain number of balls of the same color as the ball just drawn. We discuss the pertinency of such a procedure to early detect sudden or chronic changes in the parameters values of the random process made of the succession of ball numbers and we use it to trigger alarms, whose level depends on the real biological age of the elderly people watched at home.

A biofeedback system whose underlying principle consists in supplying the user with supplementary sensory information related to foot sole pressure distribution through a tongue-placed tactile output device was recently developed for improving balance. The purpose of this study was to unravel the underlying control mechanisms involved in the regulation of the centre of foot pressure (CoP) displacements during upright quiet standing with this biofeedback system. Ten young healthy adults were asked to stand as immobile as possible with their eyes closed in two conditions of No-biofeedback and Biofeedback. CoP displacements, recorded using a force platform, were processed through a space-time domain analysis and modeled as fractional Brownian motions according to the procedure of stabilogram-diffusion analysis (SDA). The space-time domain analysis showed decreased CoP displacements in the Biofeedback relative to No-biofeedback condition. Complementary, the SDA showed decreased spatiotemporal threshold at which corrective mechanisms are called into play and an increased degree of control of the CoP displacements in the Biofeedback relative to No-biofeedback condition. The present findings evidence that the effectiveness of the biofeedback in decreasing the CoP displacements during upright quiet standing stems from an increased contribution and efficiency of anti-persistent mechanisms (feedback control) involved in the regulation of the CoP displacements.