Image processing in detection of knee joints injuries based on MRI images

Abstract

This paper presents image processing methods for visualization and classification of medial meniscus tears. The first method uses watershed with a threshold segmentation approach. The algorithm was tested on a number of images of the knee obtained with a use of the magnetic resonance imaging technique (MR). Images of the knee were collected from healthy subjects and patients with a clinically diagnosed meniscal pathology. Then, watershed technique was compared with other popular methods of image segmentation, i.e. simple thresholding and region growing. For this purpose, the execution speed and the efficiency of the methods were analyzed. Additionally, an automatic detection of the meniscus based on MRI of the knee joint was developed. The solutions were implemented using classical image processing methods in the MATLAB environment with an application of the Image Processing Toolbox and MVtec Halcon vision libraries. The presented methods will have a practical value for the referring physicians and the diagnostic imaging specialists.

Using marker-controlled watershed transform to detect Baker's cyst in magnetic resonance imaging images: A pilot study

“…The results show the superiority of the watershed segmentation from the point of view of effectiveness and better distinguishing of some irregularities. [ 19 ]…”

“…The results demonstrated that watershed segmentation has the potential to be used as a screening method for meniscal injuries and diseases, with the aim of improving care. [ 17 18 19 20 ] We could not find an article that has used this algorithm for the detection of Baker's cyst in MRI. Moreover, due to more accuracy detection of lesions and delineation lymphomas, a nominated algorithm was performed in mammography and computed tomography.…”