Search

Background: MRI acceleration using deep learning (DL) convolutional neural networks (CNNs) is a novel technique with great promise. Increasing the number of convolutional layers may allow for more accurate image reconstruction. Studies on evaluating the diagnostic interchangeability of DL reconstructed knee magnetic resonance (MR) images are scarce. The purpose of this study was to develop a deep CNN (DCNN) with an optimal number of layers for accelerating knee magnetic resonance imaging (MRI) acquisition by 6-fold and to test the diagnostic interchangeability and image quality of nonaccelerated images versus images reconstructed with a 15-layer DCNN or 3-layer CNN. Methods: For the feasibility portion of this study, 10 patients were randomly selected from the Osteoarthritis Initiative (OAI) cohort. For the interchangeability portion of the study, 40 patients were randomly selected from the OAI cohort. Three readers assessed meniscal and anterior cruciate ligament (ACL) tears and cartilage defects using DCNN, CNN, and nonaccelerated images. Image quality was subjectively graded as nondiagnostic, poor, acceptable, or excellent. Interchangeability was tested by comparing the frequency of agreement when readers used both accelerated and nonaccelerated images to frequency of agreement when readers only used nonaccelerated images. A noninferiority margin of 0.10 was used to ensure type I error <= 5% and power >= 80%. A logistic regression model using generalized estimating equations was used to compare proportions; 95% confidence intervals (CIs) were constructed. Results: DCNN and CNN images were interchangeable with nonaccelerated images for all structures, with excess disagreement values ranging from -2.5% [95% CI: (-6.1, 1.1)] to 3.0% [95% CI: (-0.1, 6.1)]. The quality of DCNN images was graded higher than that of CNN images but less than that of nonaccelerated images [excellent/acceptable quality: DCNN, 95% of cases (114/120); CNN, 60% (72/120); nonaccelerated, 97.5%

The purpose of this study, is to compare the sequences: 1) proton density (PD) BLADE (BLADE is a PROPELLER-equivalent implementation of the Siemens Medical System) with fat saturation (FS) coronal (COR), 2) PD FS COR, 3) multi-planar reconstruction (MPR) with 3 mm slice thickness and 4) multi-planar reconstruction (MPR) with 1.5 mm slice thickness, both from the T2 3D-double-echo steady state (DESS) with water excitation (WE) sagittal (SAG), regarding their abilities to identify changes in the femorotibial condyle cartilage in knee MRI examinations. Thirty three consecutive patients with osteoarthritis (18 females, 15 males; mean age 56 years, range 37-71 years), who had been routinely scanned for knee examination using the previously mentioned image acquisition techniques, participated in the study. A quantitative analysis was performed based on the relative contrast (ReCON) measurements, which were taken both on normal tissues as well as on pathologies. Additionally, a qualitative analysis was performed by two radiologists. Motion and pulsatile flow artifacts were evaluated. The PD BLADE FS COR sequence produced images of higher contrast between Menisci and Cartilage, Fluid and Cartilage, Pathologies and Cartilage as well as of the Conspicuousness Superficial Cartilage and it was found to be superior to the other sequences (p < 0.001). The sequences T2 3D DESS 1.5 mm and T2 3D DESS 3 mm were significantly superior to the PD BLADE FS COR and the PD FS COR sequences in the visualization of Bone and Cartilage and the Conspicuousness Deep Surface Cartilage. This pattern of results is also confirmed by the quantitative analysis. PD FS BLADE sequences are ideal for the depiction of the cartilage pathologies compared to the conventional PD FS and 12 3D DESS sequences.

This study investigates whether measures of knee cartilage thickness can predict future loss of knee cartilage. A slow and a rapid progressor group was determined using longitudinal data, and anatomically aligned cartilage thickness maps were extracted from MRI at baseline. A novel machine learning framework was then trained using these maps. Compared to measures of mean cartilage plate thickness, group separation was increased by focusing on local cartilage differences. This result is central for clinical trials where inclusion of rapid progressors may help reduce the period needed to study effects of new disease-modifying drugs for osteoarthritis. (C) 2013 Elsevier Ltd. All rights reserved.

This study investigates whether measures of knee cartilage thickness can predict future loss of knee cartilage. A slow and a rapid progressor group was determined using longitudinal data, and anatomically aligned cartilage thickness maps were extracted from MRI at baseline. A novel machine learning framework was then trained using these maps. Compared to measures of mean cartilage plate thickness, group separation was increased by focusing on local cartilage differences. This result is central for clinical trials where inclusion of rapid progressors may help reduce the period needed to study effects of new disease-modifying drugs for osteoarthritis. (C) 2013 Elsevier Ltd. All rights reserved.

PurposeStandard knee MRI is performed under unloading (ULC) conditions and not much is known about changes of the meniscus, ligaments or cartilage under loading conditions (LC). The aim is to study the influence of loading of different knee structures at 3Tesla (T) in subjects with osteoarthritis (OA) and normal controls.Materials and methods30 subjects, 10 healthy and 20 with radiographic evidence of OA (10 mild and 10 moderate) underwent 3T MRI under ULC and LC at 50% body weight. All images were analyzed by two musculoskeletal radiologists identifying and grading cartilage, meniscal, ligamentous abnormalities. The changes between ULC and LC were assessed. For meniscus, cartilage and ligaments the changes of lesions, signal and shape were evaluated. In addition, for the meniscus changes in extrusion were examined. A multivariate regression model was used for correlations to correct the data for the impact of age, gender, BMI. A paired T-Test was performed to calculate the differences in meniscus extrusion.ResultsSubjects with degenerative knee abnormalities demonstrated significantly increased meniscus extrusion under LC when compared to normal subjects (p=0.0008-0.0027). Subjects with knee abnormalities and higher KL scores showed significantly more changes in lesion, signal and shape of the meniscus (80% (16/20) vs. 20% (2/10); p=0.0025), ligaments and cartilage during LC.ConclusionThe study demonstrates that axial loading has an effect on articular cartilage, ligament, and meniscus morphology, which is more significant in subjects with degenerative disease and may serve as an additional diagnostic tool for disease diagnosis and assessing progression in subjects with knee OA.

Background: Traditionally, assessing the degree of ossification in the epiphyseal plate for growth plate development relies on manual evaluation, which can be inefficient due to the complexities of the distal femoral epiphysis anatomy. Existing methods lack efficient detection techniques. Method: This study proposes an AI-based decision support system, designed within a product-service system (PSS) framework, to automate ossification assessment and detection of the distal femoral epiphysis in knee magnetic resonance imaging (MRI) data. The system leverages advanced machine learning techniques, specifically two Convolutional Neural Networks (CNNs), combined with computer vision techniques. This intelligent system analyzes MRI slices to predict the optimal slice for analysis and identify variations in the degree of ossification within individual datasets. Results: The proposed method's effectiveness is demonstrated using a set of T2-weighted gradient echo grayscale knee MRI data. The system successfully detects the complex anatomy of the distal femoral epiphysis, revealing variations in the degree of ossification ranging from completely closed/open to fully open/closed regions. Conclusions: This study presents a robust and efficient AI-based method, integrated within a PSS framework, for measuring the degree of ossification in the distal femoral epiphysis. This approach automates ossification assessment, providing valuable insights for clinical decision-making by clinicians and forensic practitioners. The PSS framework ensures seamless integration of the AI technology into existing workflows.