Osteoarthritis (OA) develops through heterogenous pathophysiologic pathways. As a result, no regulatory agency approved disease modifying OA drugs are available to date. Stratifying knees into MRI-based morphological phenotypes may provide insight into predicting future OA incidence, leading to improved inclusion criteria and efficacy of therapeutics. We trained convolutional neural networks to classify bone, meniscus/cartilage, inflammatory, and hypertrophy phenotypes in knee MRIs from participants in the Osteoarthritis Initiative (n = 4791). We investigated cross-sectional association between baseline morphological phenotypes and baseline structural OA (Kellgren Lawrence grade > 1) and symptomatic OA. Among participants without baseline OA, we evaluated association of baseline phenotypes with 48-month incidence of structural OA and symptomatic OA. The area under the curve of bone, meniscus/cartilage, inflammatory, and hypertrophy phenotype neural network classifiers was 0.89 ± 0.01, 0.93 ± 0.03, 0.96 ± 0.02, and 0.93 ± 0.02, respectively (mean ± standard deviation). Among those with no baseline OA, bone phenotype (OR: 2.99 (95%CI: 1.59-5.62)) and hypertrophy phenotype (OR: 5.80 (95%CI: 1.82-18.5)) each respectively increased odds of developing incident structural OA and symptomatic OA at 48 months. All phenotypes except meniscus/cartilage increased odds of undergoing total knee replacement within 96 months. Artificial intelligence can rapidly stratify knees into structural phenotypes associated with incident OA and total knee replacement, which may aid in stratifying patients for clinical trials of targeted therapeutics.

Purpose

To evaluate the diagnostic utility of two convolutional neural networks (CNNs) for severity staging of anterior cruciate ligament (ACL) injuries.

Materials and methods

In this retrospective study, 1243 knee MR images (1008 intact, 18 partially torn, 77 fully torn, and 140 reconstructed ACLs) from 224 patients (mean age, 47 years ± 14 [standard deviation]; 54% women) were analyzed. The MRI examinations were performed between 2011 and 2014. A modified scoring metric was used. Classification of ACL injuries using deep learning involved use of two types of CNN, one with three-dimensional (3D) and the other with two-dimensional (2D) convolutional kernels. Performance metrics included sensitivity, specificity, weighted Cohen κ, and overall accuracy, and the McNemar test was used to compare the performance of the CNNs.

Results

The overall accuracies for ACL injury classification using the 3D CNN and 2D CNN were 89% (225 of 254) and 92% (233 of 254), respectively (P = .27), and both CNNs had a weighted Cohen κ of 0.83. The 2D CNN and 3D CNN performed similarly in classifying intact ACLs (2D CNN, sensitivity of 93% [188 of 203] and specificity of 90% [46 of 51] vs 3D CNN, sensitivity of 89% [180 of 203] and specificity of 88% [45 of 51]). Classification of full tears by both networks was also comparable (2D CNN, sensitivity of 82% [14 of 17] and specificity of 94% [222 of 237] vs 3D CNN, sensitivity of 76% [13 of 17] and specificity of 100% [236 of 237]). The 2D CNN classified all reconstructed ACLs correctly.

Conclusion

Two-dimensional and 3D CNNs applied to ACL lesion classification had high sensitivity and specificity, suggesting that these networks could be used to help nonexperts grade ACL injuries. Supplemental material is available for this article. © RSNA, 2020.

Purpose

To learn bone shape features from spherical bone map of knee MRI images using established convolutional neural networks (CNN) and use these features to diagnose and predict osteoarthritis (OA).

Methods

A bone segmentation model was trained on 25 manually annotated 3D MRI volumes to segment the femur, tibia, and patella from 47 078 3D MRI volumes. Each bone segmentation was converted to a 3D point cloud and transformed into spherical coordinates. Different fusion strategies were performed to merge spherical maps obtained by each bone. A total of 41 822 merged spherical maps with corresponding Kellgren-Lawrence grades for radiographic OA were used to train a CNN classifier model to diagnose OA using bone shape learned features. Several OA Diagnosis models were tested and the weights for each trained model were transferred to the OA Incidence models. The OA incidence task consisted of predicting OA from a healthy scan within a range of eight time points, from 1 y to 8 y. The validation performance was compared and the test set performance was reported.

Results

The OA Diagnosis model had an area-under-the-curve (AUC) of 0.905 on the test set with a sensitivity and specificity of 0.815 and 0.839. The OA Incidence models had an AUC ranging from 0.841 to 0.646 on the test set for the range from 1 y to 8 y.

Conclusion

Bone shape was successfully used as a predictive imaging biomarker for OA. This approach is novel in the field of deep learning applications for musculoskeletal imaging and can be expanded to other OA biomarkers.

Background

Accurate interpretation of hip MRI is time-intensive and difficult, prone to inter- and intrareviewer variability, and lacks a universally accepted grading scale to evaluate morphological abnormalities.

Purpose

To 1) develop and evaluate a deep-learning-based model for binary classification of hip osteoarthritis (OA) morphological abnormalities on MR images, and 2) develop an artificial intelligence (AI)-based assist tool to find if using the model predictions improves interreader agreement in hip grading.

Study type

Retrospective study aimed to evaluate a technical development.

Population

A total of 764 MRI volumes (364 patients) obtained from two studies (242 patients from LASEM [FORCe] and 122 patients from UCSF), split into a 65-25-10% train, validation, test set for network training.

Field strength/sequence

3T MRI, 2D T₂ FSE, PD SPAIR.

Assessment

Automatic binary classification of cartilage lesions, bone marrow edema-like lesions, and subchondral cyst-like lesions using the MRNet, interreader agreement before and after using network predictions.

Statistical tests

Receiver operating characteristic (ROC) curve, area under curve (AUC), specificity and sensitivity, and balanced accuracy.

Results

For cartilage lesions, bone marrow edema-like lesions and subchondral cyst-like lesions the AUCs were: 0.80 (95% confidence interval [CI] 0.65, 0.95), 0.84 (95% CI 0.67, 1.00), and 0.77 (95% CI 0.66, 0.85), respectively. The sensitivity and specificity of the radiologist for binary classification were: 0.79 (95% CI 0.65, 0.93) and 0.80 (95% CI 0.59, 1.02), 0.40 (95% CI -0.02, 0.83) and 0.72 (95% CI 0.59, 0.86), 0.75 (95% CI 0.45, 1.05) and 0.88 (95% CI 0.77, 0.98). The interreader balanced accuracy increased from 53%, 71% and 56% to 60%, 73% and 68% after using the network predictions and saliency maps.

Data conclusion

We have shown that a deep-learning approach achieved high performance in clinical classification tasks on hip MR images, and that using the predictions from the deep-learning model improved the interreader agreement in all pathologies.

Level of evidence

3 TECHNICAL EFFICACY STAGE: 1 J. Magn. Reson. Imaging 2020;52:1163-1172.

[This corrects the article DOI: 10.1148/ryai.2021200165.].

Radiologists today play a central role in making diagnostic decisions and labeling images for training and benchmarking artificial intelligence (AI) algorithms. A key concern is low inter-reader reliability (IRR) seen between experts when interpreting challenging cases. While team-based decisions are known to outperform individual decisions, inter-personal biases often creep up in group interactions which limit nondominant participants from expressing true opinions. To overcome the dual problems of low consensus and interpersonal bias, we explored a solution modeled on bee swarms. Two separate cohorts, three board-certified radiologists, (cohort 1), and five radiology residents (cohort 2) collaborated on a digital swarm platform in real time and in a blinded fashion, grading meniscal lesions on knee MR exams. These consensus votes were benchmarked against clinical (arthroscopy) and radiological (senior-most radiologist) standards of reference using Cohen's kappa. The IRR of the consensus votes was then compared to the IRR of the majority and most confident votes of the two cohorts. IRR was also calculated for predictions from a meniscal lesion detecting AI algorithm. The attending cohort saw an improvement of 23% in IRR of swarm votes (k = 0.34) over majority vote (k = 0.11). Similar improvement of 23% in IRR (k = 0.25) in 3-resident swarm votes over majority vote (k = 0.02) was observed. The 5-resident swarm had an even higher improvement of 30% in IRR (k = 0.37) over majority vote (k = 0.07). The swarm consensus votes outperformed individual and majority vote decision in both the radiologists and resident cohorts. The attending and resident swarms also outperformed predictions from a state-of-the-art AI algorithm.

Purpose

To test the hypothesis that artificial intelligence (AI) techniques can aid in identifying and assessing lesion severity in the cartilage, bone marrow, meniscus, and anterior cruciate ligament (ACL) in the knee, improving overall MRI interreader agreement.

Materials and methods

This retrospective study was conducted on 1435 knee MRI studies (n = 294 patients; mean age, 43 years ± 15 [standard deviation]; 153 women) collected within three previous studies (from 2011 to 2014). All MRI studies were acquired using high-spatial-resolution three-dimensional fast-spin-echo CUBE sequence. Three-dimensional convolutional neural networks were developed to detect the regions of interest within MRI studies and grade abnormalities of the cartilage, bone marrow, menisci, and ACL. Evaluation included sensitivity, specificity, and Cohen linear-weighted ĸ. The impact of AI-aided grading in intergrader agreement was assessed on an external dataset.

Results

Binary lesion sensitivity reported for all tissues was between 70% and 88%. Specificity ranged from 85% to 89%. The area under the receiver operating characteristic curve for all tissues ranged from 0.83 to 0.93. Deep learning-assisted intergrader Cohen ĸ agreement significantly improved in 10 of 16 comparisons among two attending physicians and two trainees for all tissues.

Conclusion

The three-dimensional convolutional neural network had high sensitivity, specificity, and accuracy for knee-lesion-severity scoring and also increased intergrader agreement when used on an external dataset.Supplemental material is available for this article. Keywords: Bone Marrow, Cartilage, Computer Aided Diagnosis (CAD), Computer Applications-3D, Computer Applications-Detection/Diagnosis, Knee, Ligaments, MR-Imaging, Neural Networks, Observer Performance, Segmentation, Statistics © RSNA, 2021See also the commentary by Li and Chang in this issue.: An earlier incorrect version of this article appeared online. This article was corrected on April 16, 2021.