Optimizing Medical Image Classification Models for Edge Devices
Optimizing Medical Image Classification Models for Edge Devices
复制标题
优化边缘设备的医学图像分类模型
DOI：
10.1007/978-3-030-86261-9_8
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Purkayastha, S.
中科院分区：
文献类型：
--
作者：
Abid, A.;Sinha, P.;Harpale, A.;Gichoya, J.;Purkayastha, S.
用于医疗诊断的机器学习算法通常需要运行资源密集型环境，例如昂贵的云服务器或高端GPU，这使得这些模型不适用于现场使用。我们研究了模型量化和GPU加速在边缘设备上用于胸片分类的应用。我们使用了三种类型的量化(动态范围、浮点数-16和全INT-8)，我们在胸片-X光14数据集上训练的模型上进行了测试。我们实现了模型尺寸减少2-4倍，被平均AUC-ROC评分0.0%-0.9%的小幅下降所抵消。在ARM架构上，整数量化可以将推理延迟提高高达57%。然而，我们也观察到x86处理器上的延迟显著增加。GPU加速也改善了推理延迟，但内核启动开销超过了这一点。我们表明，诊断模型的优化有可能将其用途扩展到患者和医护人员使用的日常设备；然而，这些改进取决于环境和架构，应该在低资源环境中部署之前在相关设备上进行测试。
Machine learning algorithms for medical diagnostics often require resource-intensive environments to run, such as expensive cloud servers or high-end GPUs, making these models impractical for use in the field. We investigate the use of model quantization and GPU-acceleration for chest X-ray classification on edge devices. We employ 3 types of quantization (dynamic range, float-16, and full int8) which we tested on models trained on the Chest-XRay14 Dataset. We achieved a 2–4x reduction in model size, offset by small decreases in the mean AUC-ROC score of 0.0%–0.9%. On ARM architectures, integer quantization was shown to improve inference latency by up to 57%. However, we also observe significant increases in latency on x86 processors. GPU acceleration also improved inference latency, but this was outweighed by kernel launch overhead. We show that optimization of diagnostic models has the potential to expand their utility to day-to-day devices used by patients and healthcare workers; however, these improvements are context- and architecture-dependent and should be tested on the relevant devices before deployment in low-resource environments.
使用逐层训练的卷积神经网络中的快速整数近似
DOI：
--
发表时间：
2017
期刊：
International Conference on Machine Vision
影响因子：
--
作者：
D. Ilin;E. Limonova;V. Arlazarov;D. Nikolaev
通讯作者：
D. Nikolaev