Evaluation of Extra Pixel Interpolation with Mask Processing for Medical Image Segmentation with Deep Learning

TL;DR

This paper evaluates the impact of different pixel interpolation methods, including bicubic and bilinear, on mask processing for deep learning-based medical image segmentation, comparing their effects on model performance.

Contribution

It introduces a comparative analysis of BIC and NN interpolation methods for mask processing, highlighting their influence on segmentation model accuracy.

Findings

BIC-BIC model improved performance by ~9% at 256x256 resolution.

BIC-BIC model showed a 1% increase at 384x384 resolution.

NN-NN model had lower performance gains compared to BIC-BIC.

Abstract

Current mask processing operations rely on interpolation algorithms that do not produce extra pixels, such as nearest neighbor (NN) interpolation, as opposed to algorithms that do produce extra pixels, like bicubic (BIC) or bilinear (BIL) interpolation. In our previous study, the author proposed an alternative approach to NN-based mask processing and evaluated its effects on deep learning training outcomes. In this study, the author evaluated the effects of both BIC-based image and mask processing and BIC-and-NN-based image and mask processing versus NN-based image and mask processing. The evaluation revealed that the BIC-BIC model/network was an 8.9578 % (with image size 256 x 256) and a 1.0496 % (with image size 384 x 384) increase of the NN-NN network compared to the NN-BIC network which was an 8.3127 % (with image size 256 x 256) and a 0.2887 % (with image size 384 x 384) increase of the NN-NN network.