Principal component analysis-T1ρ voxel based relaxometry of the articular cartilage: a comparison of biochemical patterns in osteoarthritis and anterior cruciate ligament subjects
Abstract:Background: Quantitative MR, including T 1ρ mapping, has been extensively used to probe early biochemical changes in knee articular cartilage of subjects with osteoarthritis (OA) and others at risk for cartilage degeneration, such as those with anterior cruciate ligament (ACL) injury and reconstruction.However, limited studies have been performed aimed to assess the spatial location and patterns of T 1ρ . In this study we used a novel voxel-based relaxometry (VBR) technique coupled with principal component ana… Show more
“…In contrast, in this study, we applied automated feature learning of CNN in addition to the classical feature extraction techniques, which allowed us to gain the understanding of the relaxation time features. The PCA-based pattern analysis approach applied in this study provided insights on the role of the different layers of cartilage T 2 in characterizing OA; similar results were previously observed in a much smaller pilot study on a separate dataset (N ¼ 40) 29 . Our results suggest that, in addition to the expected global average T 2 prolongation, OA subjects show a localized prolongation just in the deep layer of the cartilage which ultimately results in the T 2 differences between the two layers being different in subjects with radiographic OA compared to controls.…”
Objective: We aim to study to what extent conventional and deep-learning-based T 2 relaxometry patterns are able to distinguish between knees with and without radiographic osteoarthritis (OA). Methods: T 2 relaxation time maps were analyzed for 4,384 subjects from the baseline Osteoarthritis Initiative (OAI) Dataset. Voxel Based Relaxometry (VBR) was used for automatic quantification and voxelbased analysis of the differences in T 2 between subjects with and without radiographic OA. A Densely Connected Convolutional Neural Network (DenseNet) was trained to diagnose OA from T 2 data. For comparison, more classical feature extraction techniques and shallow classifiers were used to benchmark the performance of our algorithm's results. Deep and shallow models were evaluated with and without the inclusion of risk factors. Sensitivity and Specificity values and McNemar test were used to compare the performance of the different classifiers. Results: The best shallow model was obtained when the first ten Principal Components, demographics and pain score were included as features (AUC ¼ 77.77%, Sensitivity ¼ 67.01%, Specificity ¼ 71.79%). In comparison, DenseNet trained on raw T 2 data obtained AUC ¼ 83.44%, Sensitivity ¼ 76.99%, Specificity ¼ 77.94%. McNemar test on two misclassified proportions form the shallow and deep model showed that the boost in performance was statistically significant (McNemar's chi-squared ¼ 10.33, degree of freedom (DF) ¼ 1, P-value ¼ 0.0013). Conclusion: In this study, we presented a Magnetic Resonance Imaging (MRI)-based data-driven platform using T 2 measurements to characterize radiographic OA. Our results showed that feature learning from T 2 maps has potential in uncovering information that can potentially better diagnose OA than simple averages or linear patterns decomposition.
“…In contrast, in this study, we applied automated feature learning of CNN in addition to the classical feature extraction techniques, which allowed us to gain the understanding of the relaxation time features. The PCA-based pattern analysis approach applied in this study provided insights on the role of the different layers of cartilage T 2 in characterizing OA; similar results were previously observed in a much smaller pilot study on a separate dataset (N ¼ 40) 29 . Our results suggest that, in addition to the expected global average T 2 prolongation, OA subjects show a localized prolongation just in the deep layer of the cartilage which ultimately results in the T 2 differences between the two layers being different in subjects with radiographic OA compared to controls.…”
Objective: We aim to study to what extent conventional and deep-learning-based T 2 relaxometry patterns are able to distinguish between knees with and without radiographic osteoarthritis (OA). Methods: T 2 relaxation time maps were analyzed for 4,384 subjects from the baseline Osteoarthritis Initiative (OAI) Dataset. Voxel Based Relaxometry (VBR) was used for automatic quantification and voxelbased analysis of the differences in T 2 between subjects with and without radiographic OA. A Densely Connected Convolutional Neural Network (DenseNet) was trained to diagnose OA from T 2 data. For comparison, more classical feature extraction techniques and shallow classifiers were used to benchmark the performance of our algorithm's results. Deep and shallow models were evaluated with and without the inclusion of risk factors. Sensitivity and Specificity values and McNemar test were used to compare the performance of the different classifiers. Results: The best shallow model was obtained when the first ten Principal Components, demographics and pain score were included as features (AUC ¼ 77.77%, Sensitivity ¼ 67.01%, Specificity ¼ 71.79%). In comparison, DenseNet trained on raw T 2 data obtained AUC ¼ 83.44%, Sensitivity ¼ 76.99%, Specificity ¼ 77.94%. McNemar test on two misclassified proportions form the shallow and deep model showed that the boost in performance was statistically significant (McNemar's chi-squared ¼ 10.33, degree of freedom (DF) ¼ 1, P-value ¼ 0.0013). Conclusion: In this study, we presented a Magnetic Resonance Imaging (MRI)-based data-driven platform using T 2 measurements to characterize radiographic OA. Our results showed that feature learning from T 2 maps has potential in uncovering information that can potentially better diagnose OA than simple averages or linear patterns decomposition.
“…Forty-seven studies were included in the meta-analysis, including data from 3079 participants. Articles included in the systematic review but excluded from the meta-analysis either examined incomparable regions of interest (ROI), or had insufficient data to be included in the meta-analyses [54, 66, 68, 69, 77, 85, 89, 90]. Fig.…”
Background
Magnetic resonance imaging (MRI) T2 and T1ρ relaxation are increasingly being proposed as imaging biomarkers potentially capable of detecting biochemical changes in articular cartilage before structural changes are evident. We aimed to: 1) summarize MRI methods of published studies investigating T2 and T1ρ relaxation time in participants at risk for but without radiographic knee OA; and 2) compare T2 and T1ρ relaxation between participants at-risk for knee OA and healthy controls.
Methods
We conducted a systematic review of studies reporting T2 and T1ρ relaxation data that included both participants at risk for knee OA and healthy controls. Participant characteristics, MRI methodology, and T1ρ and T2 relaxation data were extracted. Standardized mean differences (SMDs) were calculated within each study. Pooled effect sizes were then calculated for six commonly segmented knee compartments.
Results
55 articles met eligibility criteria. There was considerable variability between scanners, coils, software, scanning protocols, pulse sequences, and post-processing. Moderate risk of bias due to lack of blinding was common. Pooled effect sizes indicated participants at risk for knee OA had lengthened T2 relaxation time in all compartments (SMDs from 0.33 to 0.74;
p
< 0.01) and lengthened T1ρ relaxation time in the femoral compartments (SMD from 0.35 to 0.40;
p
< 0.001).
Conclusions
T2 and T1ρ relaxation distinguish participants at risk for knee OA from healthy controls. Greater standardization of MRI methods is both warranted and required for progress towards biomarker validation.
Electronic supplementary material
The online version of this article (10.1186/s12891-019-2547-7) contains supplementary material, which is available to authorized users.
“…We limited this initial study to the T 2 analysis of average values across compartments, however, previous studies reported that spatial assessment of the knee cartilage relaxation times using laminar and sub-compartmental analyses could lead to better and probably earlier identification of cartilage matrix abnormalities 36,37 Extraction of second-order statistical information or texture analysis 38,39 has been widely used to overcome the limitation of the average-based approaches. Voxelbased relaxometry techniques have also been previously proposed 40 ; this technique allows for the investigation of local cartilage composition differences, through voxel-based statistics as statistical parametric mapping, 41 principal component analysis, 42 or deep learning feature extractions. 34 TKR is the most effective intervention for end-stage OA 12 ;…”
While substantial work has been done to understand the relationships between cartilage T 2 relaxation times and osteoarthritis (OA), diagnostic and prognostic abilities of T 2 on a large population yet need to be established. Using 3921 manually annotated 2D multi-slice multi-echo spin-echo magnetic resonance imaging volume, a segmentation model for automatic knee cartilage segmentation was built and evaluated. The optimized model was then used to calculate T 2 values on the entire osteoarthritis initiative (OAI) dataset composed of longitudinal acquisitions of 4796 unique patients, 25 729 magnetic resonance imaging studies in total. Cross-sectional relationships between T 2 values, OA risk factors, radiographic OA, and pain were analyzed in the entire OAI dataset. The performance of T 2 values in predicting the future incidence of radiographic OA as well as total knee replacement (TKR) were also explored. Automatic T 2 values were comparable with manual ones. Significant associations between T 2 relaxation times and demographic and clinical variables were found. Subjects in the highest 25% quartile of tibio-femoral T 2 values had a five times higher risk of radiographic OA incidence 2 years later. Elevation of medial femur T 2 values was significantly associated with TKR after 5 years (coeff = 0.10; P = .036; CI = [0.01,0.20]). Our investigation reinforces the predictive value of T 2 for future incidence OA and TKR. The inclusion of T 2 averages from the automatic segmentation model improved several evaluation metrics when compared to only using demographic and clinical variables. K E Y W O R D S deep learning, early osteoarthritis, imaging biomarkers, T 2 relaxometry 1 | INTRODUCTION Osteoarthritis (OA), a multifactorial disease that causes joint degeneration, affects 27 million U.S. adults 1,2 and often leads to severe disability. 3 The typical treatment for late-stage, symptomatic knee OA is joint replacement, such as total knee replacement (TKR). This surgically invasive remedy offers temporary relief, but replacements often fail after 10 to 15 years-with even shorter life spans in obese individuals. Current interventions for OA mostly target the late stages of the disease, often after the irreversible joint damage has already occurred. Proposed interventions and therapies are generally reactive rather than proactive. The development of proactive therapies is contingent upon detecting the early stages of OA. Preventative strategies such as weight reduction 4,5 and various levels of exercise 6,7 decrease OA progression, and should be rigorously enforced if individual assessment demonstrates a high long-term risk for OA.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
