Adaptive segmentation of knee radiographs for selecting the optimal ROI in texture analysis

Abstract

Objective:The purposes of this study were to investigate: 1) the effect of placement of region-of-interest (ROI) for texture analysis of subchondral bone in knee radiographs, and 2) the ability of several texture descriptors to distinguish between the knees with and without radiographic osteoarthritis (OA). Design:Bilateral posterior-anterior knee radiographs were analyzed from the baseline of Osteoarthritis Initiative (OAI) and Multicenter Osteoarthritis Study (MOST) datasets. A fully automatic method to locate the most informative region from subchondral bone using adaptive segmentation was developed. We used an oversegmentation strategy for partitioning knee images into the compact regions that follow natural texture boundaries. Local Binary Patterns (LBP), Fractal Dimension (FD), Haralick features, Shannon entropy, and Histogram of Oriented Gradients (HOG) methods were computed within the standard ROI and within the proposed adaptive ROIs. Subsequently, we built logistic regression models to identify and compare the performances of each texture descriptor and each ROI placement method using 5-fold cross validation setting. Importantly, we also investigated the generalizability of our approach by training the models on OAI and testing them on MOST dataset. We used area under the receiver operating characteristic (ROC) curve (AUC) and average precision (AP) obtained from the precision-recall (PR) curve to compare the results. Results:We found that the adaptive ROI improves the classification performance (OA vs. non-OA) over the commonly-used standard ROI (up to 9% percent increase in AUC). We also observed that, from all texture parameters, LBP yielded the best performance in all settings with the best AUC of 0.840 [0.825, 0.852] and associated AP of 0.804 [0.786, 0.820]. Conclusion:Compared to the current state-of-the-art approaches, our results suggest that the proposed adaptive ROI approach in texture analysis of subchondral bone can increase the diagnostic performance for detecting the presence of radiographic OA.

A Multiparametric MR-Based RadioFusionOmics Model with Robust Capabilities of Differentiating Glioblastoma Multiforme from Solitary Brain Metastasis

“…Tumor heterogeneity from different VOI components has been associated with poor survival rates in GBM [ 16 , 32 ]. Besides, a radiomics model’s predictive performance also depends on the image features as previously demonstrated [ 15 , 21 , 33 ]. The previous studies did not, however, resolve the dilemma of whether to derive the lesion VOI from the peritumoral edema region [ 12 , 15 , 21 ], or from the enhanced tumor area as a whole [ 19 , 34 ].…”

Detection of Pediatric Femur Configuration on X-ray Images

“…Interestingly, several deep learning algorithms had been used on adult X-ray images [13][14][15][16]. Meanwhile, very little research was conducted for medical image data collected for children [17].…”

Image Texture Analysis for Medical Image Mining: A Comparative Study Direct to Osteoarthritis Classification using Knee X-ray Image

“…In recent years there has been substantial research work in OA detection and classification [3]- [7]. The early detection of OA can be applied to medical imaging and used in conjunction with a professional clinician to classify the OA.…”

“…Texture plays as one of the most important properties in images as it shows the arrangement of pixels in objects to analyze. Therefore, the texture is one of the useful solutions in medical image processing for diagnosis and detection of OA in a clinical setting [6], [7]. In [4], the research focused on the RoI of tibia texture for the analysis of OA.…”

“…In addition, the texture of an object can be analyzed using texture descriptors, which is a technique to represent and handle texture in a numeric form. Texture descriptors include: LBP [7], [11]; CLBP [12]; LBP ri ; LBP-HF [13] and LTP [14], [15].…”