Marco Carraro

Cost-efficient RGB-D smart camera for people detection and tracking

Abstract. We describe a software library we developed for efficiently using the Kinect v2, a time-of-flight RGB-D sensor, with an embedded system, the NVidia Jetson TK1, as a cost-efficient RGB-D smart camera for people detection and tracking. The speed-up needed for achieving real-time operation has been obtained using NVidia CUDA to concurrently generate and process the raw depth and infrared data and to create the three-dimensional point cloud. This library has been released as open source and the smart camera has been tested in real-world scenarios, as a people-detection node in an open-source multinode RGB-D tracking system (OpenPTrack) and onboard a service robot for endowing it with robust people-following capabilities. Moreover, we show that nonembedded computers also can benefit from our library in terms of people-detection frame rate.

Working toward precision medicine : Predicting phenotypes from exomes in the Critical Assessment of Genome Interpretation (CAGI) challenges

Precision medicine aims to predict a patients disease risk and best therapeutic options by using that individuals genetic sequencing data. The Critical Assessment of Genome Interpretation (CAGI) is a community experiment consisting of genotype–phenotype prediction challenges; participants build models, undergo assessment, and share key findings. For CAGI 4, three challenges involved using exome‐sequencing data: Crohns disease, bipolar disorder, and warfarin dosing. Previous CAGI challenges included prior versions of the Crohns disease challenge. Here, we discuss the range of techniques used for phenotype prediction as well as the methods used for assessing predictive models. Additionally, we outline some of the difficulties associated with making predictions and evaluating them. The lessons learned from the exome challenges can be applied to both research and clinical efforts to improve phenotype prediction from genotype. In addition, these challenges serve as a vehicle for sharing clinical and research exome data in a secure manner with scientists who have a broad range of expertise, contributing to a collaborative effort to advance our understanding of genotype–phenotype relationships.

Crohn Disease Risk Prediction-Best Practices and Pitfalls with Exome Data

The Critical Assessment of Genome Interpretation (CAGI) experiment is the first attempt to evaluate the state‐of‐the‐art in genetic data interpretation. Among the proposed challenges, Crohn disease (CD) risk prediction has become the most classic problem spanning three editions. The scientific question is very hard: can anybody assess the risk to develop CD given the exome data alone? This is one of the ultimate goals of genetic analysis, which motivated most CAGI participants to look for powerful new methods. In the 2016 CD challenge, we implemented all the best methods proposed in the past editions. This resulted in 10 algorithms, which were evaluated fairly by CAGI organizers. We also used all the data available from CAGI 11 and 13 to maximize the amount of training samples. The most effective algorithms used known genes associated with CD from the literature. No method could evaluate effectively the importance of unannotated variants by using heuristics. As a downside, all CD datasets were strongly affected by sample stratification. This affected the performance reported by assessors. Therefore, we expect that future datasets will be normalized in order to remove population effects. This will improve methods comparison and promote algorithms focused on causal variants discovery.

Performance of in silico tools for the evaluation of p16INK4a (CDKN2A) variants in CAGI

Correct phenotypic interpretation of variants of unknown significance for cancer‐associated genes is a diagnostic challenge as genetic screenings gain in popularity in the next‐generation sequencing era. The Critical Assessment of Genome Interpretation (CAGI) experiment aims to test and define the state of the art of genotype–phenotype interpretation. Here, we present the assessment of the CAGI p16INK4a challenge. Participants were asked to predict the effect on cellular proliferation of 10 variants for the p16INK4a tumor suppressor, a cyclin‐dependent kinase inhibitor encoded by the CDKN2A gene. Twenty‐two pathogenicity predictors were assessed with a variety of accuracy measures for reliability in a medical context. Different assessment measures were combined in an overall ranking to provide more robust results. The R scripts used for assessment are publicly available from a GitHub repository for future use in similar assessment exercises. Despite a limited test‐set size, our findings show a variety of results, with some methods performing significantly better. Methods combining different strategies frequently outperform simpler approaches. The best predictor, Yang&Zhou lab, uses a machine learning method combining an empirical energy function measuring protein stability with an evolutionary conservation term. The p16INK4a challenge highlights how subtle structural effects can neutralize otherwise deleterious variants.

Improved Skeleton Estimation by Means of Depth Data Fusion from Multiple Depth Cameras

In this work, we address the problem of human skeleton estimation when multiple depth cameras are available. We propose a system that takes advantage of the knowledge of the camera poses to create a collaborative virtual depth image of the person in the scene which consists of points from all the cameras and that represents the person in a frontal pose. This depth image is fed as input to the open-source body part detector in the Point Cloud Library. A further contribution of this work is the improvement of this detector obtained by introducing two new components: as a pre-processing, a people detector is applied to remove the background from the depth map before estimating the skeleton, while an alpha-beta tracking is added as a post-processing step for filtering the obtained joint positions over time. The overall system has been proven to effectively improve the skeleton estimation on two sequences of people in different poses acquired from two first-generation Microsoft Kinect.

Matching phenotypes to whole genomes: Lessons learned from four iterations of the personal genome project community challenges

The advent of next‐generation sequencing has dramatically decreased the cost for whole‐genome sequencing and increased the viability for its application in research and clinical care. The Personal Genome Project (PGP) provides unrestricted access to genomes of individuals and their associated phenotypes. This resource enabled the Critical Assessment of Genome Interpretation (CAGI) to create a community challenge to assess the bioinformatics communitys ability to predict traits from whole genomes. In the CAGI PGP challenge, researchers were asked to predict whether an individual had a particular trait or profile based on their whole genome. Several approaches were used to assess submissions, including ROC AUC (area under receiver operating characteristic curve), probability rankings, the number of correct predictions, and statistical significance simulations. Overall, we found that prediction of individual traits is difficult, relying on a strong knowledge of trait frequency within the general population, whereas matching genomes to trait profiles relies heavily upon a small number of common traits including ancestry, blood type, and eye color. When a rare genetic disorder is present, profiles can be matched when one or more pathogenic variants are identified. Prediction accuracy has improved substantially over the last 6 years due to improved methodology and a better understanding of features.

Lessons from the CAGI-4 Hopkins clinical panel challenge

The CAGI‐4 Hopkins clinical panel challenge was an attempt to assess state‐of‐the‐art methods for clinical phenotype prediction from DNA sequence. Participants were provided with exonic sequences of 83 genes for 106 patients from the Johns Hopkins DNA Diagnostic Laboratory. Five groups participated in the challenge, predicting both the probability that each patient had each of the 14 possible classes of disease, as well as one or more causal variants. In cases where the Hopkins laboratory reported a variant, at least one predictor correctly identified the disease class in 36 of the 43 patients (84%). Even in cases where the Hopkins laboratory did not find a variant, at least one predictor correctly identified the class in 39 of the 63 patients (62%). Each prediction group correctly diagnosed at least one patient that was not successfully diagnosed by any other group. We discuss the causal variant predictions by different groups and their implications for further development of methods to assess variants of unknown significance. Our results suggest that clinically relevant variants may be missed when physicians order small panels targeted on a specific phenotype. We also quantify the false‐positive rate of DNA‐guided analysis in the absence of prior phenotypic indication.

Skeleton estimation and tracking by means of depth data fusion from depth camera networks

Abstract In this work, we describe an approach for estimation and tracking of the skeleton of the human body from camera networks exploiting only depth data. The algorithm takes advantage of multiple views by building and merging together the 3D point clouds. The final skeleton is computed from a virtual depth image generated from this point cloud by means of back-projection to a reference camera image plane. Before the back-projection, the person point cloud is frontalized with respect to the reference camera, so that the virtual depth image represents the person from a frontal viewpoint and the accuracy of the skeleton estimation algorithm is maximized. Our experiments show how the proposed approach boosts the performance with respect to other state-of-the-art approaches. Moreover, the proposed algorithm requires low computational burden, thus running in real-time.

People tracking and re-identification by face recognition for RGB-D camera networks

This paper describes a face recognition-based people tracking and re-identification system for RGB-D camera networks. The system tracks people and learns their faces online to keep track of their identities even if they move out from the cameras field of view once. For robust people re-identification, the system exploits the combination of a deep neural network- based face representation and a Bayesian inference-based face classification method. The system also provides a predefined people identification capability: it associates the online learned faces with predefined people face images and names to know the peoples whereabouts, thus, allowing a rich human-system interaction. Through experiments, we validate the re-identification and the predefined people identification capabilities of the system and show an example of the integration of the system with a mobile robot. The overall system is built as a Robot Operating System (ROS) module. As a result, it simplifies the integration with the many existing robotic systems and algorithms which use such middleware. The code of this work has been released as open-source in order to provide a baseline for the future publications in this field.

The Origin of Personalized Medicine and the Systems Biology Revolution

The complete sequencing of the human genome has opened up many avenues of research. Among these, the notion of personalized medicine is becoming increasingly common. In this chapter, we review the major implications of the genomic era for improving the diagnosis and treatment of diseases. Medicine has witnessed several paradigm shifts in the course of the last two centuries, and personalized medicine is bound to be seen in the same way. Sequencing technology has evolved by orders of magnitude, coming into the range of