Performance comparison of medical image classification systems using TensorFlow Keras, PyTorch, and JAX

TL;DR

This study compares the inference speed and classification accuracy of TensorFlow Keras, PyTorch, and JAX frameworks in blood cell image classification, highlighting performance differences influenced by image resolution and framework optimizations.

Contribution

It provides a detailed performance analysis of three popular deep learning frameworks specifically for medical blood image classification tasks.

Findings

JAX and PyTorch achieved classification accuracy comparable to benchmarks.

Performance varies across frameworks depending on image size and resolution.

Inference times differ significantly among the frameworks.

Abstract

Medical imaging plays a vital role in early disease diagnosis and monitoring. Specifically, blood microscopy offers valuable insights into blood cell morphology and the detection of hematological disorders. In recent years, deep learning-based automated classification systems have demonstrated high potential in enhancing the accuracy and efficiency of blood image analysis. However, a detailed performance analysis of specific deep learning frameworks appears to be lacking. This paper compares the performance of three popular deep learning frameworks, TensorFlow with Keras, PyTorch, and JAX, in classifying blood cell images from the publicly available BloodMNIST dataset. The study primarily focuses on inference time differences, but also classification performance for different image sizes. The results reveal variations in performance across frameworks, influenced by factors such as image resolution and framework-specific optimizations. Classification accuracy for JAX and PyTorch was comparable to current benchmarks, showcasing the efficiency of these frameworks for medical image classification.