OpenMedIA: Open-Source Medical Image Analysis Toolbox and Benchmark under Heterogeneous AI Computing Platforms

TL;DR

OpenMedIA is an open-source toolbox that offers diverse deep learning methods for medical image analysis, supporting heterogeneous AI platforms like NVIDIA and Huawei Ascend, with implementations in PyTorch and MindSpore.

Contribution

It introduces the first open-source library providing comparable PyTorch and MindSpore implementations for medical image analysis on heterogeneous AI platforms.

Findings

Provides a comprehensive set of methods for classification, segmentation, localisation, detection

Includes benchmark results on multiple datasets

Supports both NVIDIA and Huawei Ascend systems

Abstract

In this paper, we present OpenMedIA, an open-source toolbox library containing a rich set of deep learning methods for medical image analysis under heterogeneous Artificial Intelligence (AI) computing platforms. Various medical image analysis methods, including 2D/3D medical image classification, segmentation, localisation, and detection, have been included in the toolbox with PyTorch and/or MindSpore implementations under heterogeneous NVIDIA and Huawei Ascend computing systems. To our best knowledge, OpenMedIA is the first open-source algorithm library providing compared PyTorch and MindSpore implementations and results on several benchmark datasets. The source codes and models are available at https://git.openi.org.cn/OpenMedIA.