Knee Cartilage Segmentation Research Articles

The unified extreme learning machines and discriminative random fields for automatic knee cartilage and meniscus segmentation from multi-contrast MR images

Segmenting articular cartilage and meniscus from magnetic resonance (MR) images is an essential task for the assessment of knee pathology. Most of the previous classification-based works for cartilage and meniscus segmentation only rely on independent labellings by a classifier, but do not consider the spatial context interaction. The labels of most image voxels are actually dependent upon their neighbours. In this study, we present an automatic knee segmentation system working on multi-contrast MR images where a novel classification model unifying an extreme learning machine (ELM)-based association potential and a discriminative random field (DRF)-based interaction potential is proposed. The DRF model introduces spatial dependencies between neighbouring voxels to the independent ELM classification. We exploit a rich set of features From multi-contrast MR images to train the proposed classification model and perform the loopy belief propagation for the inference. The proposed model is evaluated on multi-contrast MR datasets acquired from 11 subjects with results outperforming the independent classifiers in terms of segmentation accuracy of both cartilages and menisci.

Optimization of local shape and appearance probabilities for segmentation of knee cartilage in 3-D MR images

We propose a fully automatic method for segmenting knee cartilage in 3-D MR images which consists of bone segmentation, bone-cartilage interface (BCI) classification, and cartilage segmentation. For bone segmentation, we propose a modified version of the recently presented branch-and-mincut method, and for classifying the BCI, we propose a voxel classification method based on binary classifiers of position and local appearance. The core contribution of this paper is the cartilage segmentation method where localized Markov random fields (MRF) are separately constructed and optimized for local image patches. The region and boundary potentials of the MRFs are computed from the retrieved segmentation results of training images that are relevant to each local patch. Here, local shape and appearance cues are adaptively combined depending on the local image characteristics. For experimentation, a dataset comprising MR images of ten different subjects and another comprising the baseline and two-year follow-up scans for nine different subjects are constructed. Both qualitative and quantitative comparisons of the results of the proposed method with semi-automatic segmentation methods demonstrate the potential of the proposed method for clinical application.

A novel fast knee cartilage segmentation technique for T 2 measurements at MR imaging – data from the Osteoarthritis Initiative

Magnetic resonance imaging (MRI) T(2) relaxation time assesses non-invasively cartilage composition and can be used as early biomarker for knee osteoarthritis. Most knee cartilage segmentation techniques were primarily developed for volume measurements in DESS or SPGR sequences. For T(2) quantifications, these segmentations need to be superimposed on T(2) maps. However, given that these procedures are time consuming and require manual alignment, using them for analysis of T(2) maps in large clinical trials like the Osteoarthritis Initiative (OAI) is challenging. A novel direct segmentation technique (DST) for T(2) maps was therefore developed. Using the DST, T(2) measurements were performed and compared with those determined with an established segmentation superimposition technique (SST). MR images of five OAI participants were analysed with both techniques three times by one reader and five different images sets additionally with DST three times by two readers. Segmentations and T(2) measurements of one knee required on average 63±3min with DST (vs302±13min for volume and T(2) measurements with SST). Bland-Altman plots indicated good agreement between the two segmentation techniques, respectively the two readers. Reproducibility errors of both techniques (DST vs SST) were similar (P>0.05) for whole knee cartilage mean T(2) (1.46% vs 2.18%), laminar (up to 2.53% vs 3.19%) and texture analysis (up to 8.34% vs 9.45%). Inter-reader reproducibility errors of DST were higher for texture analysis (up to 15.59%) than for mean T(2) (1.57%) and laminar analysis (up to 2.17%). Due to these results, the novel DST can be recommended for T(2) measurements in large clinical trials like the OAI.

Intra- and inter-observer reproducibility of volume measurement of knee cartilage segmented from the OAI MR image set using a novel semi-automated segmentation method

We developed a semi-automated method based on a graph-cuts algorithm for segmentation and volumetric measurements of the cartilage from high-resolution knee magnetic resonance (MR) images from the Osteoarthritis Initiative (OAI) database and assessed the intra- and inter-observer reproducibility of measurements obtained via this method. MR image sets from 20 subjects of varying Kellgren-Lawrence (KL) grades (from 0 to IV) on fixed flexion knee radiographs were selected from the baseline double-echo and steady-state (DESS) knee MR images in the OAI database (0.B.1 Imaging Data set). Two trained radiologists independently performed the segmentation of knee cartilage twice using the semi-automated method. The volumes of segmented cartilage were computed and compared. The intra- and inter-observer reproducibility were determined by means of the coefficient of variation (CV%) of repeated cartilage segmented volume measurements. The subjects were also divided into the low- (0, I or II) and high-KL (III or IV) groups. The differences in cartilage volume measurements and CV% within and between the observers were tested with t tests. The mean (+/-SD) intra-observer CV% for the 20 cases was 1.29 (+/-1.05)% for observer 1 and 1.67 (+/-1.14)% for observer 2, while the mean (+/-SD) inter-observer CV% was 1.31 (+/-1.26)% for session 1 and 1.79 (+/-1.72)% for session 2. There was no significant difference between the two intra-observer CV%'s (P=0.272) and between the two inter-observer CV%'s (P=0.353). The mean intra-observer CV% of the low-KL group was significantly smaller than that for the high-KL group for observer 1 (0.83 vs 1.86%: P=0.025). The segmentation processing times used by the two observers were significantly different (observer 1 vs 2): (mean 49+/-12 vs 33+/-6min) for session 1 and (49+/-8 vs 32+/-8min) for session 2. The semi-automated graph-cuts method allowed us to segment and measure cartilage from high-resolution 3T MR images of the knee with high intra- and inter-observer reproducibility in subjects with varying severity of OA.

Knee Cartilage: Efficient and Reproducible Segmentation on High-Spatial-Resolution MR Images with the Semiautomated Graph-Cut Algorithm Method

This HIPAA-compliant study was exempt from institutional review board approval because the 10 image data sets were deidentified in the Osteoarthritis Initiative database, and they were processed and analyzed without any clinical information being accessed. The purpose of this study was to prospectively evaluate the efficiency and reproducibility of the semiautomated graph-cut method (SA method) in the segmentation of knee cartilage and to compare its performance with that of the conventional manual delineation segmentation method (M method). Two radiologists independently performed segmentation with each method in two separate sessions: They performed the M method (M1 and M2 for the first and second sessions, respectively) for every third section and the SA method (SA1 and SA2 for the first and second sessions, respectively) for every section. The SA method was significantly more efficient (mean processing time, 53 minutes vs 156 minutes for SA1 vs M1 and 53 minutes vs 118 minutes for SA2 vs M2; P < .001) and reproducible (mean volume overlap, 94.3% vs 87.8% for the SA method vs the M method; P < .001) than the M method.

2.1 Maintaining walking direction when combining gaitwith vision distraction in young and elderly

We present a method for the automated segmentation of knee bones and cartilage from magnetic resonance imaging (MRI) that combines a priori knowledge of anatomical shape with Convolutional Neural Networks (CNNs). The proposed approach incorporates 3D Statistical Shape Models (SSMs) as well as 2D and 3D CNNs to achieve a robust and accurate segmentation of even highly pathological knee structures. The shape models and neural networks employed are trained using data from the Osteoarthritis Initiative (OAI) and the MICCAI grand challenge “Segmentation of Knee Images 2010” (SKI10), respectively. We evaluate our method on 40 validation and 50 submission datasets from the SKI10 challenge. For the first time, an accuracy equivalent to the inter-observer variability of human readers is achieved in this challenge. Moreover, the quality of the proposed method is thoroughly assessed using various measures for data from the OAI, i.e. 507 manual segmentations of bone and cartilage, and 88 additional manual segmentations of cartilage. Our method yields sub-voxel accuracy for both OAI datasets. We make the 507 manual segmentations as well as our experimental setup publicly available to further aid research in the field of medical image segmentation. In conclusion, combining localized classification via CNNs with statistical anatomical knowledge via SSMs results in a state-of-the-art segmentation method for knee bones and cartilage from MRI data.

Characterisation of the guinea pig model of osteoarthritis by in vivo three-dimensional magnetic resonance imaging

Objective: To characterise longitudinal changes in joint integrity and cartilage volume in vivo in the guinea pig spontaneous osteoarthritis (OA) model by magnetic resonance imaging (MRI). Methods: Guinea pigs knee were imaged in vivo by high-resolution three-dimensional (3D) MRI between the ages of 3 and 12 months. Image analysis was performed to assess qualitative knee joint changes between 3 and 12 months ( n=16) and quantitative volumetric changes of the medial tibial cartilage between 9 and 12 months ( n=7). After imaging, animals were killed and knees were assessed macroscopically and histologically. Results: From 3 to 6 months qualitative observation by MRI and histopathology indicated localised cartilage swelling on the medial tibial plateau. At 6 months, bone cysts had developed in the epiphysis. At 9 months, we observed by MRI and histopathology, fragmentation of the medial tibial cartilage in areas not protected by the meniscus. Cartilage degeneration had intensified at 12 months with evidence of widespread loss of cartilage throughout the tibial plateau. Segmentation of the MR cartilage images showed a 36% loss of volume between 9 and 12 months. Conclusions: We have achieved 3D image acquisition and segmentation of knee cartilage in a guinea pig model of chronic OA, which permits measurements previously only possible in man. High resolution and short acquisition time allowed qualitative longitudinal characterisation of the entire knee joint and enabled us to quantify for the first time longitudinal tibial cartilage volume loss associated with disease progression.