Your browser doesn't support javascript.
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 12.424
Filtrar
2.
PLoS One ; 15(2): e0228434, 2020.
Artigo em Inglês | MEDLINE | ID: mdl-32027668

RESUMO

The service quality and system dependability of real-time communication networks strongly depends on the analysis of monitored data, to identify concrete problems and their causes. Many of these can be described by either their structural or temporal properties, or a combination of both. As current research is short of approaches sufficiently addressing both properties simultaneously, we propose a new feature space specifically suited for this task, which we analyze for its theoretical properties and its practical relevance. We evaluate its classification performance when used on real-world data sets of structural-temporal mobile communication data, and compare it to the performance achieved of feature representations used in related work. For this purpose we propose a system which allows the automatic detection and prediction of classes of pre-defined sequence behavior, greatly reducing costs caused by the otherwise required manual analysis. With our proposed feature spaces this system achieves a precision of more than 93% at recall values of 100%, with an up to 6.7% higher effective recall than otherwise similarly performing alternatives, notably outperforming alternative deep learning, kernel learning and ensemble learning approaches of related work. Furthermore the supported system calibration allows separating reliable from unreliable predictions more effectively, which is highly relevant for any practical application.


Assuntos
Comunicação , Aprendizado Profundo , Aprendizado de Máquina , Redes Neurais de Computação , Sistemas Computacionais/normas , Confiabilidade dos Dados , Mineração de Dados/métodos , Mineração de Dados/normas , Conjuntos de Dados como Assunto/normas , Humanos , Aplicativos Móveis/normas , Aplicativos Móveis/estatística & dados numéricos , Reprodutibilidade dos Testes , Fatores de Tempo , Estudos de Validação como Assunto
3.
PLoS One ; 15(2): e0227805, 2020.
Artigo em Inglês | MEDLINE | ID: mdl-32045936

RESUMO

A force sensor system was developed to give real-time visual feedback on a range of force. In a prospective observational cross-section study, twenty-two anaesthesia nurses applied cricoid pressure at a target range of 30-40 Newtons for 60 seconds in three sequential steps on manikin: Group A (step 1 blinded, no sensor), Group B (step 2 blinded sensor), Group C (step 3 sensor feedback). A weighing scale was placed below the manikin. This procedure was repeated once again at least 1 week apart. The feedback system used 3 different colours to indicate the force range achieved as below target, achieve target, above target. Significantly higher proportion of target cricoid pressure was achieved with the use of sensor feedback in Group C; 85.9% (95%CI: 82.7%-88.7%) compared to when blinded from sensor in Group B; 31.3% (95%CI: 27.4-35.4%). Cricoid force achieved blind (Group B) exceeded force achieved with feedback (Group C) by a mean of 8.0 (95%CI: 5.9-10.2, p<0.0001) and 6.2 (95%CI:4.1-8.3, p< 0.0001) Newtons in round 1 and 2 respectively. Weighing scale read lower than corresponding force sensor by a mean of 8.4 Newtons (95% CI: 7.1-9.7, p<0.0001) in group B and 5.8 Newtons (95% CI: 4.5-7.1, p<0.0001) in Group C. Force sensor visual feedback system enabled application of reproducible target cricoid pressure with less variability and has potential value in clinical use. Using weighing scale to quantify and train cricoid pressure requires a review. Understanding the force applied is the first step to make cricoid pressure a safe procedure.


Assuntos
Anestesiologia , Sistemas Computacionais , Manequins , Adulto , Fenômenos Biomecânicos , Cartilagem Cricoide/fisiologia , Estudos Transversais , Retroalimentação , Feminino , Humanos , Estudos Prospectivos
4.
J Surg Oncol ; 121(6): 964-966, 2020 May.
Artigo em Inglês | MEDLINE | ID: mdl-32103507

RESUMO

BACKGROUND: Near-infrared (NIR) fluorescence imaging has recently been introduced to the sentinel lymph node (SLN) mapping because of the benefits of the SLN biopsy, such as providing real-time and high-resolution optical guidance. Methylene blue is available and less expensive as an SLN mapping tracer. Our study aims to identify SLN through the NIR fluorescence imaging system mediated by blue dye. METHODS: Early-stage breast cancer patients were prospectively enrolled. All participants received a subareolar or peritumoral injection of 1 mL methylene blue (MB) before surgery. The MB fluorescence system was set immediately after injection. SLNs were searched and removed under the guidance of fluorescence and blue dye. RESULTS: We identified SLN adequately with the help of real-time lymphography and blue dye. Symbolic lymphatic drainage patterns were also observed. CONCLUSION: NIR fluorescence imaging mediated by blue dye has benefits on the identification of lymph vessels, the location of SLN, and the patterns of breast lymphatic flow.


Assuntos
Neoplasias da Mama/diagnóstico por imagem , Azul de Metileno , Linfonodo Sentinela/diagnóstico por imagem , Espectroscopia de Luz Próxima ao Infravermelho/métodos , Neoplasias da Mama/patologia , Corantes , Sistemas Computacionais , Feminino , Humanos , Linfografia/métodos , Estadiamento de Neoplasias , Linfonodo Sentinela/patologia , Espectrometria de Fluorescência/métodos
5.
PLoS One ; 15(2): e0229431, 2020.
Artigo em Inglês | MEDLINE | ID: mdl-32092113

RESUMO

BACKGROUND: Current guidelines underline the importance of high-quality chest compression during cardiopulmonary resuscitation (CPR), to improve outcomes. Contrary to this many studies show that chest compression is often carried out poorly in clinical practice, and long interruptions in compression are observed. This prospective cohort study aimed to analyse whether chest compression quality changes when a real-time feedback system is used to provide simultaneous audiovisual feedback on chest compression quality. For this purpose, pauses in compression, compression frequency and compression depth were compared. METHODS: The study included 292 out-of-hospital cardiac arrests in three consecutive study groups: first group, conventional resuscitation (no-sensor CPR); second group, using a feedback sensor to collect compression depth data without real-time feedback (sensor-only CPR); and third group, with real-time feedback on compression quality (sensor-feedback CPR). Pauses and frequency were analysed using compression artefacts on electrocardiography, and compression depth was measured using the feedback sensor. With this data, various parameters were determined in order to be able to compare the chest compression quality between the three consecutive groups. RESULTS: The compression fraction increased with sensor-only CPR (group 2) in comparison with no-sensor CPR (group 1) (80.1% vs. 87.49%; P < 0.001), but there were no further differences belonging compression fraction after activation of sensor-feedback CPR (group 3) (P = 1.00). Compression frequency declined over the three study groups, reaching the guideline recommendations (127.81 comp/min vs. 122.96 comp/min, P = 0.02 vs. 119.15 comp/min, P = 0.008) after activation of sensor-feedback CPR (group 3). Mean compression depth only changed minimally with sensor-feedback (52.49 mm vs. 54.66 mm; P = 0.16), but the fraction of compressions with sufficient depth (at least 5 cm) and compressions within the recommended 5-6 cm increased significantly with sensor-feedback CPR (56.90% vs. 71.03%; P = 0.003 and 28.74% vs. 43.97%; P < 0.001). CONCLUSIONS: The real-time feedback system improved chest compression quality regarding pauses in compression and compression frequency and facilitated compliance with the guideline recommendations. Compression depth did not change significantly after activation of the real-time feedback. Even the sole use of a CPR-feedback-sensor ("sensor-only CPR") improved performance regarding pauses in compression and compression frequency, a phenomenon known as the 'Hawthorne effect'. Based on this data real-time feedback systems can be expected to raise the quality level in some parts of chest compression quality. TRIAL REGISTRATION: International Clinical Trials Registry Platform of the World Health Organisation and German Register of Clinical Trials (DRKS00009903), Registered 09 February 2016 (retrospectively registered).


Assuntos
Reanimação Cardiopulmonar/normas , Sistemas Computacionais , Retroalimentação Sensorial , Massagem Cardíaca/normas , Parada Cardíaca Extra-Hospitalar/terapia , Adulto , Idoso , Idoso de 80 Anos ou mais , Reanimação Cardiopulmonar/instrumentação , Reanimação Cardiopulmonar/métodos , Estudos de Coortes , Cardioversão Elétrica , Eletrocardiografia/instrumentação , Eletrocardiografia/métodos , Retroalimentação Sensorial/fisiologia , Feminino , Alemanha , Massagem Cardíaca/instrumentação , Massagem Cardíaca/métodos , Humanos , Masculino , Pessoa de Meia-Idade , Pressão , Melhoria de Qualidade , Tórax , Fatores de Tempo
6.
Nat Med ; 26(1): 52-58, 2020 01.
Artigo em Inglês | MEDLINE | ID: mdl-31907460

RESUMO

Intraoperative diagnosis is essential for providing safe and effective care during cancer surgery1. The existing workflow for intraoperative diagnosis based on hematoxylin and eosin staining of processed tissue is time, resource and labor intensive2,3. Moreover, interpretation of intraoperative histologic images is dependent on a contracting, unevenly distributed, pathology workforce4. In the present study, we report a parallel workflow that combines stimulated Raman histology (SRH)5-7, a label-free optical imaging method and deep convolutional neural networks (CNNs) to predict diagnosis at the bedside in near real-time in an automated fashion. Specifically, our CNNs, trained on over 2.5 million SRH images, predict brain tumor diagnosis in the operating room in under 150 s, an order of magnitude faster than conventional techniques (for example, 20-30 min)2. In a multicenter, prospective clinical trial (n = 278), we demonstrated that CNN-based diagnosis of SRH images was noninferior to pathologist-based interpretation of conventional histologic images (overall accuracy, 94.6% versus 93.9%). Our CNNs learned a hierarchy of recognizable histologic feature representations to classify the major histopathologic classes of brain tumors. In addition, we implemented a semantic segmentation method to identify tumor-infiltrated diagnostic regions within SRH images. These results demonstrate how intraoperative cancer diagnosis can be streamlined, creating a complementary pathway for tissue diagnosis that is independent of a traditional pathology laboratory.


Assuntos
Neoplasias Encefálicas/diagnóstico , Sistemas Computacionais , Monitorização Intraoperatória , Redes Neurais de Computação , Análise Espectral Raman , Algoritmos , Neoplasias Encefálicas/diagnóstico por imagem , Ensaios Clínicos como Assunto , Aprendizado Profundo , Humanos , Processamento de Imagem Assistida por Computador , Probabilidade
8.
Int J Radiat Oncol Biol Phys ; 106(2): 413-421, 2020 02 01.
Artigo em Inglês | MEDLINE | ID: mdl-31655198

RESUMO

PURPOSE: The transition from frame-based brain stereotactic radiosurgery (SRS) to frameless delivery is supported by real-time intrafraction monitoring to ensure accurate delivery. The purpose of this study is to characterize these real-time motion traces in a large cohort of patients treated with frameless gated brain SRS and to develop patient-specific predictions of tolerance violations. METHODS AND MATERIALS: SRS patients treated on the Gamma Knife Icon were immobilized using a device-specific thermoplastic head mask. A motion marker was fixed to the patient's nose, with gating and cone beam computed tomography (CBCT)-based corrections to the treatment at excursions from baseline exceeding 1.5 mm. The traces of 1446 fractions were analyzed according to magnitude (932 unique treatment plans for 462 unique individual patients), directional distribution of displacement, and stability. A neural network model was developed to predict interruptions based on a subset of trace data. RESULTS: The average displacement of the marker in the first fraction of all patients was 0.62 ± 0.25 mm with beam CBCT corrections, which would otherwise be modeled at 0.96 ± 0.96 mm without intrafraction motion correction (P < .0001). Twenty-nine percent of fractions delivered were interrupted, of which the Z-axis (superoinferior) motion was the largest contributor to excursion. Baseline corrections significantly compensated for the magnitude of motion in all 3 dimensions (P < .01). The motion relative to the first acquired CBCT was on average seen to consistently increase with treatment time, with the minimum P value occurring at 61.3 minutes. The neural network prediction model was able to predict treatment interruptions with 84% sensitivity on the first 5-minute sample of the trace. CONCLUSIONS: Corrections to marker position significantly decreased marker excursions in all 3 axes compared with a single CBCT alignment. Patient-specific modeling may aid in the optimization of cases selected for frameless radiosurgery to increase the accuracy of planned delivery.


Assuntos
Neoplasias Encefálicas/radioterapia , Sistemas Computacionais , Movimentos dos Órgãos , Radiocirurgia/métodos , Radioterapia Guiada por Imagem/métodos , Neoplasias Encefálicas/diagnóstico por imagem , Tomografia Computadorizada de Feixe Cônico , Marcadores Fiduciais , Humanos , Imobilização/instrumentação , Raios Infravermelhos , Máscaras , Nariz , Radiocirurgia/instrumentação
9.
Neural Netw ; 121: 366-386, 2020 Jan.
Artigo em Inglês | MEDLINE | ID: mdl-31593842

RESUMO

Machine learning is yielding unprecedented interest in research and industry, due to recent success in many applied contexts such as image classification and object recognition. However, the deployment of these systems requires huge computing capabilities, thus making them unsuitable for embedded systems. To deal with this limitation, many researchers are investigating brain-inspired computing, which would be a perfect alternative to the conventional Von Neumann architecture based computers (CPU/GPU) that meet the requirements for computing performance, but not for energy-efficiency. Therefore, neuromorphic hardware circuits that are adaptable for both parallel and distributed computations need to be designed. In this paper, we focus on Spiking Neural Networks (SNNs) with a comprehensive study of neural coding methods and hardware exploration. In this context, we propose a framework for neuromorphic hardware design space exploration, which allows to define a suitable architecture based on application-specific constraints and starting from a wide variety of possible architectural choices. For this framework, we have developed a behavioral level simulator for neuromorphic hardware architectural exploration named NAXT. Moreover, we propose modified versions of the standard Rate Coding technique to make trade-offs with the Time Coding paradigm, which is characterized by the low number of spikes propagating in the network. Thus, we are able to reduce the number of spikes while keeping the same neuron's model, which results in an SNN with fewer events to process. By doing so, we seek to reduce the amount of power consumed by the hardware. Furthermore, we present three neuromorphic hardware architectures in order to quantitatively study the implementation of SNNs. One of these architectures integrates a novel hybrid structure: a highly-parallel computation core for most solicited layers, and time-multiplexed computation units for deeper layers. These architectures are derived from a novel funnel-like Design Space Exploration framework for neuromorphic hardware.


Assuntos
Potenciais de Ação/fisiologia , Inteligência Artificial , Encéfalo/fisiologia , Computadores , Redes Neurais de Computação , Neurônios/fisiologia , Sistemas Computacionais , Humanos , Aprendizado de Máquina
10.
BMC Bioinformatics ; 20(Suppl 16): 591, 2019 Dec 02.
Artigo em Inglês | MEDLINE | ID: mdl-31787090

RESUMO

BACKGROUND: Supercomputers have become indispensable infrastructures in science and industries. In particular, most state-of-the-art scientific results utilize massively parallel supercomputers ranked in TOP500. However, their use is still limited in the bioinformatics field due to the fundamental fact that the asynchronous parallel processing service of Grid Engine is not provided on them. To encourage the use of massively parallel supercomputers in bioinformatics, we developed middleware called Virtual Grid Engine, which enables software pipelines to automatically perform their tasks as MPI programs. RESULT: We conducted basic tests to check the time required to assign jobs to workers by VGE. The results showed that the overhead of the employed algorithm was 246 microseconds and our software can manage thousands of jobs smoothly on the K computer. We also tried a practical test in the bioinformatics field. This test included two tasks, the split and BWA alignment of input FASTQ data. 25,055 nodes (2,000,440 cores) were used for this calculation and accomplished it in three hours. CONCLUSION: We considered that there were four important requirements for this kind of software, non-privilege server program, multiple job handling, dependency control, and usability. We carefully designed and checked all requirements. And this software fulfilled all the requirements and achieved good performance in a large scale analysis.


Assuntos
Algoritmos , Simulação por Computador , Sistemas Computacionais , Interface Usuário-Computador , Humanos , Software
11.
Sensors (Basel) ; 20(1)2019 Dec 24.
Artigo em Inglês | MEDLINE | ID: mdl-31878128

RESUMO

Information about an approaching vehicle is helpful for pedestrians to avoid traffic accidents while most of the past studies related to collision avoidance systems have focused on alerting drivers and controlling vehicles. This paper proposes a technique to detect an approaching vehicle aiming at alerting a pedestrian by observing the variation of the received signal strength indicator (RSSI) of the repeatedly radiated beacons from a vehicle, called the alert beacons. A linear regression algorithm is first applied to RSSI samples. The decision about whether a vehicle is approaching or not is made by the Student's t-test for the linear regression coefficient. A passive method, where the pedestrian's device behaves only as a receiver, is first described. The neighbor-discovery-based (ND-based) method, in which the pedestrian's device repeatedly broadcasts advertising beacons and the moving vehicle in the vicinity returns the alert beacon when it receives the advertising beacon, is then proposed to improve the detection performance as well as reduce the device's energy consumption. The theoretical detection error rate under Rayleigh fading is derived. It is revealed that the proposed ND-based method achieves a lower detection error rate when compared with the passive method under the same delay.


Assuntos
Acidentes de Trânsito/prevenção & controle , Algoritmos , Automóveis , Sistemas Computacionais , Humanos , Pedestres
12.
J Bus Contin Emer Plan ; 13(2): 120-135, 2019 Jan 01.
Artigo em Inglês | MEDLINE | ID: mdl-31779740

RESUMO

The hype surrounding cloud-based disaster recovery (DR) continues to gain steam - and with good reason. That said, due to cost, compatibility and/or organisational restrictions, cloud-based DR is not necessarily the best fit for all systems. This paper presents case studies and a structured approach to evaluate the various options, including public/private cloud DR, DR as a service (DRaaS) and hybrid solutions. For example, the pay-per-use model for cloud DR saves money while the cloud environment is dormant, but results in higher run-time costs than co-location or on-premises solutions when DR is executed. Similarly, organisations need to understand the range of DRaaS options to make an informed decision. Where cloud is not a good fit for all systems, a hybrid solution can satisfy conflicting requirements while leveraging the benefits of cloud where appropriate. Finally, organisational constraints can tip the balance away from what may appear to be the best fit on paper.


Assuntos
Planejamento em Desastres , Desastres , Computação em Nuvem , Sistemas Computacionais
14.
Sensors (Basel) ; 19(18)2019 Sep 10.
Artigo em Inglês | MEDLINE | ID: mdl-31509999

RESUMO

Freezing of gait (FoG) is a common motor symptom in patients with Parkinson's disease (PD). FoG impairs gait initiation and walking and increases fall risk. Intelligent external cueing systems implementing FoG detection algorithms have been developed to help patients recover gait after freezing. However, predicting FoG before its occurrence enables preemptive cueing and may prevent FoG. Such prediction remains challenging given the relative infrequency of freezing compared to non-freezing events. In this study, we investigated the ability of individual and ensemble classifiers to predict FoG. We also studied the effect of the ADAptive SYNthetic (ADASYN) sampling algorithm and classification cost on classifier performance. Eighteen PD patients performed a series of daily walking tasks wearing accelerometers on their ankles, with nine experiencing FoG. The ensemble classifier formed by Support Vector Machines, K-Nearest Neighbors, and Multi-Layer Perceptron using bagging techniques demonstrated highest performance (F1 = 90.7) when synthetic FoG samples were added to the training set and class cost was set as twice that of normal gait. The model identified 97.4% of the events, with 66.7% being predicted. This study demonstrates our algorithm's potential for accurate prediction of gait events and the provision of preventive cueing in spite of limited event frequency.


Assuntos
Sistemas Computacionais , Marcha/fisiologia , Doença de Parkinson/diagnóstico , Doença de Parkinson/fisiopatologia , Acelerometria , Idoso , Algoritmos , Feminino , Humanos , Masculino , Processamento de Sinais Assistido por Computador
15.
Comput Methods Programs Biomed ; 178: 247-263, 2019 Sep.
Artigo em Inglês | MEDLINE | ID: mdl-31416553

RESUMO

BACKGROUND AND OBJECTIVE: Conventional information systems are built on top of a relational database. The main weakness of these systems is impossibility to define stable data schema ahead when the knowledge of the system is evolving and dynamic. The widely accepted alternatives to relational databases are ontologies that can be used for designing information systems. Many research papers describe various methods for improving reliability and precision in generating the type of the Lenke classification based on the image processing techniques or a computer program, but all of them require radiograph images. The main objective of this paper is to demonstrate the development of an ontology-based module of the information system ScolioMedIS for adolescent idiopathic scoliosis (AIS) diagnosis and monitoring, which uses optical 3D methods to determine the Lenke classification of AIS and to avoid harmful effects of traditional radiation diagnosis. METHODS: For creating an ontology-based module of the ScolioMedIS we used the following steps: specification, conceptualization, formalization and implementation. In the specification and conceptualization phase we performed data collection and analysis to define domain, concepts and relationships for ontology design. In the formalization and implementation stage we developed the OBR-Scolio ontology and the ontology-based module of the ScolioMedIS. The module employs the Protégé-OWL API, as a collection of Java interfaces for the OBR-Scolio ontology, which enables the creating, deleting, and editing of the basic elements of the OBR-Scolio ontology, as well as the querying of the ontology. RESULTS: The ontology-based module of ScolioMedIS is tested on the datasets of 20 female and 15 male patients with AIS between the ages of 11 and 18, to categorize spinal curvatures and to automatically generate statistical indicators about the frequency of the basic spinal curvatures, degree of progression or regression of deformity and statistical indicators about curvature characteristics according to the Lenke classification system and Lenke scoliosis types. Results are then compared with analysis of the Lenke classification of 315 observed patients, performed using traditional radiation techniques. CONCLUSIONS: This part of the system allows continuous monitoring of the progression/regression of spinal curvatures for each registered patient, which may provide a better management of scoliosis (diagnosis and treatment).


Assuntos
Diagnóstico por Computador/métodos , Imageamento Tridimensional/métodos , Escoliose/diagnóstico por imagem , Adolescente , Algoritmos , Criança , Gráficos por Computador , Sistemas Computacionais , Feminino , Humanos , Processamento de Imagem Assistida por Computador , Vértebras Lombares/diagnóstico por imagem , Masculino , Informática Médica , Linguagens de Programação , Reprodutibilidade dos Testes , Software , Vértebras Torácicas/diagnóstico por imagem , Interface Usuário-Computador
16.
Comput Methods Programs Biomed ; 178: 31-39, 2019 Sep.
Artigo em Inglês | MEDLINE | ID: mdl-31416558

RESUMO

BACKGROUND AND OBJECTIVE: Convolutional neural networks (CNNs) offer human experts-like performance and in the same time they are faster and more consistent in their prediction. However, most of the proposed CNNs require an expensive state-of-the-art hardware which substantially limits their use in practical scenarios and commercial systems, especially for clinical, biomedical and other applications that require on-the-fly analysis. In this paper, we investigate the possibility of making CNNs lighter by parametrizing the architecture and decreasing the number of trainable weights of a popular CNN: U-Net. METHODS: In order to demonstrate that comparable results can be achieved with substantially less trainable weights than the original U-Net we used a challenging application of a pixel-wise virus classification in Transmission Electron Microscopy images with minimal annotations (i.e. consisting only of the virus particle centers or centerlines). We explored 4 U-Net hyper-parameters: the number of base feature maps, the feature maps multiplier, the number of the encoding-decoding levels and the number of feature maps in the last 2 convolutional layers. RESULTS: Our experiments lead to two main conclusions: 1) the architecture hyper-parameters are pivotal if less trainable weights are to be used, and 2) if there is no restriction on the trainable weights number using a deeper network generally gives better results. However, training larger networks takes longer, typically requires more data and such networks are also more prone to overfitting. Our best model achieved an accuracy of 82.2% which is similar to the original U-Net while using nearly 4 times less trainable weights (7.8 M in comparison to 31.0 M). We also present a network with  < 2 M trainable weights that achieved an accuracy of 76.4%. CONCLUSIONS: The proposed U-Net hyper-parameter exploration can be adapted to other CNNs and other applications. It allows a comprehensive CNN architecture designing with the aim of a more efficient trainable weight use. Making the networks faster and lighter is crucial for their implementation in many practical applications. In addition, a lighter network ought to be less prone to over-fitting and hence generalize better.


Assuntos
Microscopia Eletrônica de Transmissão/métodos , Vírus/ultraestrutura , Algoritmos , Sistemas Computacionais , Bases de Dados Factuais , Aprendizado Profundo , Processamento de Imagem Assistida por Computador , Reprodutibilidade dos Testes
17.
Comput Intell Neurosci ; 2019: 1939171, 2019.
Artigo em Inglês | MEDLINE | ID: mdl-31396269

RESUMO

The threat to people's lives and property posed by fires has become increasingly serious. To address the problem of a high false alarm rate in traditional fire detection, an innovative detection method based on multifeature fusion of flame is proposed. First, we combined the motion detection and color detection of the flame as the fire preprocessing stage. This method saves a lot of computation time in screening the fire candidate pixels. Second, although the flame is irregular, it has a certain similarity in the sequence of the image. According to this feature, a novel algorithm of flame centroid stabilization based on spatiotemporal relation is proposed, and we calculated the centroid of the flame region of each frame of the image and added the temporal information to obtain the spatiotemporal information of the flame centroid. Then, we extracted features including spatial variability, shape variability, and area variability of the flame to improve the accuracy of recognition. Finally, we used support vector machine for training, completed the analysis of candidate fire images, and achieved automatic fire monitoring. Experimental results showed that the proposed method could improve the accuracy and reduce the false alarm rate compared with a state-of-the-art technique. The method can be applied to real-time camera monitoring systems, such as home security, forest fire alarms, and commercial monitoring.


Assuntos
Cor , Sistemas Computacionais , Fogo , Máquina de Vetores de Suporte , Algoritmos , Metodologias Computacionais , Humanos
19.
Bone Joint J ; 101-B(8): 960-969, 2019 08.
Artigo em Inglês | MEDLINE | ID: mdl-31362543

RESUMO

AIMS: The aim of this study was to give estimates of the incidence of component incompatibility in hip and knee arthroplasty and to test the effect of an online, real-time compatibility check. MATERIALS AND METHODS: Intraoperative barcode registration of arthroplasty implants was introduced in Denmark in 2013. We developed a compatibility database and, from May 2017, real-time compatibility checking was implemented and became part of the registration. We defined four classes of component incompatibility: A-I, A-II, B-I, and B-II, depending on an assessment of the level of risk to the patient (A/B), and on whether incompatibility was knowingly accepted (I/II). RESULTS: A total of 26 524 arthroplasties were analyzed. From 12 307 procedures that were undertaken before implementation of the compatibility check, 21 class A incompatibilities were identified (real- or high-risk combinations; 0.17%; 95% confidence interval (CI) 0.11 to 0.26). From 5692 hip and 6615 knee procedures prior to implementation of the compatibility check, we found rates of class A-I incompatibility (real- or high-risk combinations unknowingly inserted) of 0.14% (95% CI 0.06 to 0.28) and 0.17% (95% CI 0.08 to 0.30), respectively. From 14 217 procedures after the introduction of compatibility checking (7187 hips and 7030 knees), eight class A incompatibilities (0.06%; 95% CI 0.02 to 0.11) were identified. This difference was statistically significant (p = 0.008). CONCLUSION: Our data presents validated estimates of the baseline incidence of incompatibility events for hip and knee arthroplasty procedures and shows that a significant reduction in class A incompatibility events is possible using a web-based recording system. Cite this article: Bone Joint J 2019;101-B:960-969.


Assuntos
Artroplastia de Quadril/instrumentação , Artroplastia do Joelho/instrumentação , Prótese de Quadril/efeitos adversos , Prótese do Joelho/efeitos adversos , Erros Médicos/prevenção & controle , Desenho de Prótese/efeitos adversos , Falha de Prótese/etiologia , Estudos de Coortes , Sistemas Computacionais , Dinamarca , Humanos , Erros Médicos/efeitos adversos , Erros Médicos/estatística & dados numéricos , Falha de Prótese/efeitos adversos , Sistema de Registros
20.
Stud Health Technol Inform ; 265: 74-79, 2019 Aug 09.
Artigo em Inglês | MEDLINE | ID: mdl-31431580

RESUMO

There has been an acknowledged need for the integration of health technologies such as the electronic health record system (EHR) into health professional education. At the University of Victoria we have been experimenting with different models, architectures and applications of educational EHRs in the context of training health informatics, medical, and nursing students who will ultimately use this technology in their daily practice upon graduation. Our initial work involved the development of a Web-based portal that contained a number of open source EHRs and is described in this paper. In addition to the technical side, considerations around pedagogy and how best to integrate such technology into the classroom and educational experience are discussed. Finally, challenges and lessons learned from our decade of work in this area are discussed.


Assuntos
Registros Eletrônicos de Saúde , Informática Médica , Sistemas Computacionais , Humanos , Estudantes de Enfermagem
SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA