Deep learning research has generated widespread interest leading to emergence of a large variety of technological innovations and applications. As significant proportion of deep learning research focuses on vision based applications, there exists a potential for using some of these techniques to enable low-power portable health-care diagnostic support solutions. In this paper, we propose an embedded-hardware-based implementation of microscopy diagnostic support system for PoC case study on: (a) Malaria in thick blood smears, (b) Tuberculosis in sputum samples, and (c) Intestinal parasite infection in stool samples. We use a Squeeze-Net based model to reduce the network size and computation time. We also utilize the Trained Quantization technique to further reduce memory footprint of the learned models. This enables microscopy-based detection of pathogens that classifies with laboratory expert level accuracy as a standalone embedded hardware platform. The proposed implementation is 6x more power-efficient compared to conventional CPU-based implementation and has an inference time of $\sim$ 3 ms/sample.
A Prospective Observational Study to Investigate Performance of a Chest X-ray Artificial Intelligence Diagnostic Support Tool Across 12 U.S. Hospitals
University of Minnesota(明尼苏达大学);Emory University(埃默里大学);M Health Fairview Informatics(M Health Fairview 信息学);Indiana University(印第安纳大学);North Memorial Health Hospital(North Memorial Health 医院)
Peri-Diagnostic Decision Support Through Cost-Efficient Feature Acquisition at Test-Time
Technical University of Munich (TUM)(慕尼黑工业大学);German Center for Vertigo and Balance Disorders (DSGZ)(德国眩晕与平衡障碍中心);Ludwig-Maximilians-Universität (LMU)(路德维希-马克西米利安大学);Whiting School of Engineering, Johns Hopkins University(约翰斯·霍普金斯大学怀廷工程学院)
CGX: Adaptive System Support for Communication-Efficient Deep Learning
Institute of Science and Technology Austria(奥地利科学技术研究所);University of British Columbia(不列颠哥伦比亚大学)
扩展训练工作负载的能力一直是深度学习的关键性能推动因素之一。主要的扩展方法是基于GPU的数据并行训练,这是通过硬件和软件支持高效的点对点通信,特别是通过硬件带宽预留来推动的。过度配置是有代价的:相对于流行的“消费级”服务器,具有这种支持的“云级”服务器之间存在数量级的价格差异,尽管单个服务器级和消费级GPU可以具有类似的计算包络。在本文中,我们证明了昂贵的硬件过度配置方法可以通过算法和系统设计来取代,并提出了一个名为CGX的框架,该框架为ML应用程序中的压缩通信提供了有效的软件支持,用于多GPU单节点训练以及更大规模的多节点训练。CGX基于两项技术进步:\emph{At the system level},它依赖于重新开发的ML框架通信堆栈,为压缩通信提供灵活,高效的支持。\emph{At the application level},它提供了\emph{seamless, parameter-free}与流行框架的集成,因此最终用户不必修改培训食谱,也不必修改重要的培训代码。这由\emph{layer-wise adaptive compression}技术补充,该技术动态地平衡压缩增益与精度保持。CGX与流行的ML框架集成,为基于商用硬件的多GPU节点提供高达3倍的加速,并在多节点设置中进行了数量级改进,对准确性的影响可以忽略不计。