A Neural Template Matching Method to Detect Knee Joint Areas

TL;DR

This paper introduces a neural template matching approach using a trainable optimizer to accurately detect knee joint areas in X-ray images, reducing manual annotation requirements and improving localization precision.

Contribution

It presents a novel trainable optimizer for template matching that enhances knee joint detection accuracy with minimal manual annotations.

Findings

The trainable optimizer outperforms other optimization methods in finding minimal distance values.

Single-side manual annotation suffices for precise knee joint localization.

The method effectively correlates minimal matching scores with localization quality.

Abstract

In this paper, new methods are considered to detect knee joint areas in bilateral PA fixed flexion knee X-ray images. The methods are of template matching type where the distance criterion is based on the negative normalized cross-correlation. The manual annotations are made on only one side of a single bilateral image when the templates are selected. The best matching patch search is formulated as an unconstrained continuous domain minimization problem. For the minimization problem different optimization methods are considered. The main method of the paper is a trainable optimizer where the method is taught to take zoomed and possibly rotated patches from its input images which look like the template. In the experiments, we compare the minimum values found by different optimization methods. We also look at some test images to examine the correspondence between the minimum value and how well the knee area is localized. It seems that making annotations only to a single image enables to detect knee joint areas quite precisely.