A Sensor Network Approach for Violence Detection in Smart Cities Using Deep Learning.

Citizen safety in modern urban environments is an important aspect of life quality. Implementation of a smart city approach to video surveillance depends heavily on the capability of gathering and processing huge amounts of live urban data. Analyzing data from high bandwidth surveillance video streams provided by large size distributed sensor networks is particularly challenging. We propose here an efficient method for automatic violent behavior detection designed for video sensor networks. Known solutions to real-time violence detection are not suitable for implementation in a resource-constrained environment due to the high processing power requirements. Our algorithm achieves real-time processing on a Raspberry PI-embedded architecture. To ensure separation of temporal and spatial information processing we employ a computationally effective cascaded approach. It consists of a deep neural network followed by a time domain classifier. In contrast with current approaches, the deep neural network input is fed exclusively with motion vector features extracted directly from the MPEG encoded video stream. As proven by results, we achieve state-of-the-art performance, while running on a low computational resources embedded architecture.

Is Contrast Enhanced Ultrasonography a useful tool in a beginner's hand? How much can a Computer Assisted Diagnosis prototype help in characterizing the malignancy of focal liver lesions?

AIM: Contrast enhanced ultrasound (CEUS) improved the characterization of focal liver lesions (FLLs), but is an operatordependent method. The goal of this paper was to test a computer assisted diagnosis (CAD) prototype and to see its benefit in assisting a beginner in the evaluation of FLLs. MATERIAL AND METHOD: Our cohort included 97 good quality CEUS videos[34% hepatocellular carcinomas (HCC), 12.3% hypervascular metastases (HiperM), 11.3% hypovascular metastases (HipoM), 24.7% hemangiomas (HMG), 17.5% focal nodular hyperplasia (FNH)] that were used to develop a CAD prototype based on an algorithm that tested a binary decision based classifier. Two young medical doctors (1 year CEUS experience), two experts and the CAD prototype, reevaluated 50 FLLs CEUS videos (diagnosis of benign vs. malignant) first blinded to clinical data, in order to evaluate the diagnostic gap beginner vs. expert. RESULTS: The CAD classifier managed a 75.2% overall (benign vs. malignant) correct classification rate. The overall classification rates for the evaluators, before and after clinical data were: first beginner-78%; 94%; second beginner-82%; 96%; first expert-94%; 100%; second expert-96%; 98%. For both beginners, the malignant vs. benign diagnosis significantly improved after knowing the clinical data (p=0.005; p=0,008). The expert was better than the beginner (p=0.04) and better than the CAD (p=0.001). CAD in addition to the beginner can reach the expert diagnosis. CONCLUSIONS: The most frequent lesions misdiagnosed at CEUS were FNH and HCC. The CAD prototype is a good comparing tool for a beginner operator that can be developed to assist the diagnosis. In order to increase the classification rate, the CAD system for FLL in CEUS must integrate the clinical data.

Computer aided diagnosis method for steatosis rating in ultrasound images using random forests.

UNLABELLED: In this paper we discuss the problem of computer aided evaluation of the severity of steatosis disease using ultrasound images. The AIM of the study being to compare the automatic evaluation of liver steatosis using random forests (RF) and support vector machine (SVM) classifiers. MATERIAL AND METHOD: One hundred and twenty consecutive patients with steatosis or normal liver, assessed by ultrasound by the same expert, were enrolled. We graded steatosis in four stages and trained two classifiers to rate the severity of disease, based on a large set of labeled images and a large set of features, including several features obtained by robust estimation techniques. We compared RF and SVM classifiers. The classifiers were trained using cross-validation. There was 80% of data randomly selected for training and 20% for testing the classifier. This procedure was performed 20 times. The main measure of performance was the accuracy. RESULTS: From all cases, 10 were rated as normal liver, 70 as having mild, 33 moderate, and 7 severe steatosis. Our best experts' ratings were used as ground truth data. RF outperformed the SVM classifier and confirmed the ability of this classifier to perform well without feature selection. In contrast, the performance of the SVM classifier was poor without feature selection and improved significantly after feature selection. CONCLUSION: The ability and accuracy of RF to classify well the steatosis severity, without feature selection, were superior as compared to SVM.