ShmCaffe: A Distributed Deep Learning Platform with Shared Memory Buffer for HPC Architecture

2018 ◽  
Author(s):  
Shinyoung Ahn ◽  
Joongheon Kim ◽  
Eunji Lim ◽  
Wan Choi ◽  
Aziz Mohaisen ◽  
...  
Keyword(s):  
Deep Learning ◽  
Shared Memory ◽  
Learning Platform ◽  
Memory Buffer

scholarly journals Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Ling-Ping Cen ◽  
Jie Ji ◽  
Jian-Wei Lin ◽  
Si-Tong Ju ◽  
Hong-Jie Lin ◽  
...  
Keyword(s):  
Deep Learning ◽  
High Efficiency ◽  
Characteristic Curve ◽  
Weighted Average ◽  
Age Related Macular Degeneration ◽  
Learning Platform ◽  
Age Related ◽  
Public Data ◽  
Primary Test ◽  
Retinal Fundus

AbstractRetinal fundus diseases can lead to irreversible visual impairment without timely diagnoses and appropriate treatments. Single disease-based deep learning algorithms had been developed for the detection of diabetic retinopathy, age-related macular degeneration, and glaucoma. Here, we developed a deep learning platform (DLP) capable of detecting multiple common referable fundus diseases and conditions (39 classes) by using 249,620 fundus images marked with 275,543 labels from heterogenous sources. Our DLP achieved a frequency-weighted average F1 score of 0.923, sensitivity of 0.978, specificity of 0.996 and area under the receiver operating characteristic curve (AUC) of 0.9984 for multi-label classification in the primary test dataset and reached the average level of retina specialists. External multihospital test, public data test and tele-reading application also showed high efficiency for multiple retinal diseases and conditions detection. These results indicate that our DLP can be applied for retinal fundus disease triage, especially in remote areas around the world.

QuantMed: Component-based deep learning platform for translational research

2020 ◽  
Author(s):  
Jan Klein ◽  
Markus Wenzel ◽  
Daniel Romberg ◽  
Alexander Köhn ◽  
Peter Kohlmann ◽  
...  
Keyword(s):  
Deep Learning ◽  
Translational Research ◽  
Learning Platform

Violence recognition using convolutional neural network: A survey

2020 ◽  
Vol 39 (5) ◽  
pp. 7931-7952
Author(s):  
Gaurav Tripathi ◽  
Kuldeep Singh ◽  
Dinesh Kumar Vishwakarma
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Law And Order ◽  
Linear Features ◽  
Automatic Feature Extraction ◽  
Learning Platform ◽  
Violence Detection ◽  
The World ◽  
Crowd Behaviour ◽  
Behaviour Changes

Violence detection is a challenging task in the computer vision domain. Violence detection framework depends upon the detection of crowd behaviour changes. Violence erupts due to disagreement of an idea, injustice or severe disagreement. The aim of any country is to maintain law and order and peace in the area. Violence detection thus becomes an important task for authorities to maintain peace. Traditional methods have existed for violence detection which are heavily dependent upon hand crafted features. The world is now transitioning in to Artificial Intelligence based techniques. Automatic feature extraction and its classification from images and videos is the new norm in surveillance domain. Deep learning platform has provided us the platter on which non-linear features can be extracted, self-learnt and classified as per the appropriate tool. One such tool is the Convolutional Neural Networks, also known as ConvNets, which has the ability to automatically extract features and classify them in to their respective domain. Till date there is no survey of deciphering violence behaviour techniques using ConvNets. We hope that this survey becomes an exclusive baseline for future violence detection and analysis in the deep learning domain.

scholarly journals DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models

2020 ◽  
Vol 253 ◽  
pp. 107206 ◽  
Author(s):  
Yuzhi Zhang ◽  
Haidi Wang ◽  
Weijie Chen ◽  
Jinzhe Zeng ◽  
Linfeng Zhang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Potential Energy ◽  
Concurrent Learning ◽  
Energy Models ◽  
Learning Platform

BCI cinematics – A pre-release analyser for movies using H2O deep learning platform

2019 ◽  
Vol 74 ◽  
pp. 547-556 ◽  
Author(s):  
Sundhara Kumar K.B ◽  
Krishna G ◽  
Bhalaji N ◽  
Chithra S
Keyword(s):  
Deep Learning ◽  
Learning Platform

Cross Hardware-Software Boundary Exploration for Scalable and Optimized Deep Learning Platform Design

2018 ◽  
Vol 10 (4) ◽  
pp. 107-110 ◽  
Author(s):  
Baozi Chen ◽  
Lei Wang ◽  
Qingbo Wu ◽  
Yusong Tan ◽  
Peng Zou
Keyword(s):  
Deep Learning ◽  
Platform Design ◽  
Learning Platform

scholarly journals Knowledge sharing on deep learning in physics research using VISPA

2020 ◽  
Vol 245 ◽  
pp. 05040
Author(s):  
Max Beer ◽  
Niclas Eich ◽  
Martin Erdmann ◽  
Peter Fackeldey ◽  
Benjamin Fischer ◽  
...  
Keyword(s):  
Deep Learning ◽  
Knowledge Sharing ◽  
Work Environment ◽  
Central Place ◽  
Current Status ◽  
Web Interface ◽  
Software Environment ◽  
Web Browser ◽  
Learning Platform ◽  
Hands On

The VISPA (VISual Physics Analysis) project provides a streamlined work environment for physics analyses and hands-on teaching experiences with a focus on deep learning. VISPA has already been successfully used in HEP analyses and teaching and is now being further developed into an interactive deep learning platform. One specific example is to meet knowledge sharing needs in deep learning by combining paper, code and data at a central place. Additionally the possibility to run it directly from the web browser is a key feature of this development. Any SSH reachable resource can be accessed via the VISPA web interface. This enables a flexible and experiment agnostic computing experience. The user interface is based on JupyterLab and is extended with analysis specific tools, such as a parametric file browser and TensorBoard. Our VISPA instance is backed by extensive GPU resources and a rich software environment. We present the current status of the VISPA project and its upcoming new features.

scholarly journals A Hybrid Catheter Localisation Framework in Echocardiography Based on Electromagnetic Tracking and Deep Learning Segmentation

2020 ◽  
Author(s):  
Fei. Jia ◽  
Shu. Wang
Keyword(s):  
Deep Learning ◽  
Medical Device ◽  
Ultrasound Imaging ◽  
Image Method ◽  
Ultrasound Images ◽  
Electromagnetic Tracking ◽  
External Device ◽  
Low Resolution ◽  
Learning Platform ◽  
Catheter Tip

AbstractInterventional cardiology procedure is an important type of minimally invasive surgery that deals with the catheter-based treatment of cardiovascular diseases, such as coronary artery diseases, strokes, peripheral arterial diseases and aortic diseases. Ultrasound imaging, also called echocardiography, is a typical imaging tool that monitors catheter puncturing. Localising a medical device accurately during cardiac interventions can help improve the procedure’s safety and reliability under ultrasound imaging. However, external device tracking and image-based tracking methods can only provide a partial solution. Thus, we proposed a hybrid framework, with the combination of both methods to localise the catheter tip target in an automatic way. The external device used was an electromagnetic tracking system from North Digital Inc (NDI) and the ultrasound image analysis was based on UNet, a deep learning network for semantic segmentation. From the external method, the tip’s location was determined precisely, and the deep learning platform segmented the exact catheter tip automatically. This novel hybrid localisation framework combines the advantages of external electromagnetic (EM) tracking and deep-learning-based image method, which offers a new solution to identify the moving medical device in low-resolution ultrasound images.Featured ApplicationThis framework can be applied to other medical-device localisation fields to help doctors identify a moving target in low-resolution ultrasound images.

Sign in / Sign up

Export Citation Format

Share Document