Faculty Bibliography

OBJECTIVE:To determine if MRI-based radiomics from hamstring muscles are related to injury and if the features could be used to perform a time to return to sport (RTS) classification. We hypothesize that radiomics from hamstring muscles, especially T2-weighted and diffusion tensor imaging-based features, are related to injury and can be used for RTS classification. SUBJECTS AND METHODS/METHODS:MRI data from 32 athletes at the University of Wisconsin-Madison that sustained a hamstring strain injury were collected. Diffusion tensor imaging and T1- and T2-weighted images were processed, and diffusion maps were calculated. Radiomics features were extracted from the four hamstring muscles in each limb and for each MRI modality, individually. Feature selection was performed and multiple support vector classifiers were cross-validated to differentiate between involved and uninvolved limbs and perform binary (≤ or > 25 days) and multiclass (< 14 vs. 14-42 vs. > 42 days) classification of RTS. RESULT/RESULTS:The combination of radiomics features from all diffusion tensor imaging and T2-weighted images provided the most accurate differentiation between involved and uninvolved limbs (AUC ≈ 0.84 ± 0.16). For the binary RTS classification, the combination of all extracted radiomics offered the most accurate classification (AUC ≈ 0.95 ± 0.15). While for the multiclass RTS classification, the combination of features from all the diffusion tensor imaging maps provided the most accurate classification (weighted one vs. rest AUC ≈ 0.81 ± 0.16). CONCLUSION/CONCLUSIONS:This pilot study demonstrated that radiomics features from hamstring muscles are related to injury and have the potential to predict RTS.

OBJECTIVE:To identify the region of interest (ROI) to represent injury and observe between-limb diffusion tensor imaging (DTI) microstructural differences in muscle following hamstring strain injury. MATERIALS AND METHODS/METHODS:Participants who sustained a hamstring strain injury prospectively underwent 3T-MRI of bilateral thighs using T1, T2, and diffusion-weighted imaging at time of injury (TOI), return to sport (RTS), and 12 weeks after RTS (12wks). ROIs were using the hyperintense region on a T2-weighted sequence: edema, focused edema, and primary muscle injured excluding edema (no edema). Linear mixed-effects models were used to compare diffusion parameters between ROIs and timepoints and limbs and timepoints. RESULTS:(p-values = 0.058-0.12), and compared to 12wks (p-values < 0.02). In the no edema ROI, differences in diffusivity measures were not observed (p-values > 0.82). At TOI, no edema ROI diffusivity measures were lower than the edema ROI (p-values < 0.001) but not at RTS or 12wks (p-values > 0.69). A significant limb-by-timepoint interaction was detected for all diffusivity measures with increased diffusion in the involved limb at TOI (p-values < 0.001) but not at RTS or 12wks (p-values > 0.42). Significant differences in fractional anisotropy over time or between limbs were not detected. CONCLUSION/CONCLUSIONS:Hyperintensity on T2-weighted imaging used to define the injured region holds promise in describing muscle microstructure following hamstring strain injury by demonstrating between-limb differences at TOI but not at follow-up timepoints.

Hamstring strain injuries (HSI) are a common occurrence in athletics and complicated by limited prognostic indicators and high rates of reinjury. Assessment of injury characteristics at the time of injury (TOI) may be used to manage athlete expectations for time to return to sport (RTS) and mitigate reinjury risk. Magnetic resonance imaging (MRI) is routinely used in soft tissue injury management, but its prognostic value for HSI is widely debated. Recent advancements in musculoskeletal MRI, such as diffusion tensor imaging (DTI), have allowed for quantitative measures of muscle microstructure assessment. The purpose of this study was to determine the association of TOI MRI-based measures, including the British Athletic Muscle Injury Classification (BAMIC) system, edema volume, and DTI metrics, with time to RTS and reinjury incidence. Negative binomial regressions and generalized estimating equations were used to determine relationships between imaging measures and time to RTS and reinjury, respectively. Twenty-six index injuries were observed, with five recorded reinjuries. A significant association was not detected between BAMIC score and edema volume at TOI with days to RTS (p-values ≥ 0.15) or reinjury (p-values ≥ 0.13). Similarly, a significant association between DTI metrics and days to RTS was not detected (p-values ≥ 0.11). Although diffusivity metrics are expected to increase following injury, decreased values were observed in those who reinjured (mean diffusivity, p = 0.016; radial diffusivity, p = 0.02; principal effective diffusivity eigenvalues, p-values = 0.007-0.057). Additional work to further understand the directional relationship observed between DTI metrics and reinjury status and the influence of external factors is warranted.

Objective: To identify the region of interest (ROI) to represent injury and observe between-limb diffusion tensor imaging (DTI) microstructural differences in muscle following hamstring strain injury. Materials and methods: Participants who sustained a hamstring strain injury prospectively underwent 3T-MRI of bilateral thighs using T1, T2, and diffusion-weighted imaging at time of injury (TOI), return to sport (RTS), and 12 weeks after RTS (12wks). ROIs were using the hyperintense region on a T2-weighted sequence: edema, focused edema, and primary muscle injured excluding edema (no edema). Linear mixed-effects models were used to compare diffusion parameters between ROIs and timepoints and limbs and timepoints. Results: Twenty-four participants (29 injuries) were included. A significant ROI-by-timepoint interaction was detected for all diffusivity measures. The edema and focused edema ROIs demonstrated increased diffusion at TOI compared to RTS for all diffusivity measures (p-values < 0.006), except λ 1 (p-values = 0.058"“0.12), and compared to 12wks (p-values < 0.02). In the no edema ROI, differences in diffusivity measures were not observed (p-values > 0.82). At TOI, no edema ROI diffusivity measures were lower than the edema ROI (p-values < 0.001) but not at RTS or 12wks (p-values > 0.69). A significant limb-by-timepoint interaction was detected for all diffusivity measures with increased diffusion in the involved limb at TOI (p-values < 0.001) but not at RTS or 12wks (p-values > 0.42). Significant differences in fractional anisotropy over time or between limbs were not detected. Conclusion: Hyperintensity on T2-weighted imaging used to define the injured region holds promise in describing muscle microstructure following hamstring strain injury by demonstrating between-limb differences at TOI but not at follow-up timepoints.

Different pathologies of the hip are characterized by the abnormal shape of the bony structures of the joint, namely the femur and the acetabulum. Three-dimensional (3D) models of the hip can be used for diagnosis, biomechanical simulation, and planning of surgical treatments. These models can be generated by building 3D surfaces of the joint's structures segmented on magnetic resonance (MR) images. Deep learning can avoid time-consuming manual segmentations, but its performance depends on the amount and quality of the available training data. Data augmentation and transfer learning are two approaches used when there is only a limited number of datasets. In particular, data augmentation can be used to artificially increase the size and diversity of the training datasets, whereas transfer learning can be used to build the desired model on top of a model previously trained with similar data. This study investigates the effect of data augmentation and transfer learning on the performance of deep learning for the automatic segmentation of the femur and acetabulum on 3D MR images of patients diagnosed with femoroacetabular impingement. Transfer learning was applied starting from a model trained for the segmentation of the bony structures of the shoulder joint, which bears some resemblance to the hip joint. Our results suggest that data augmentation is more effective than transfer learning, yielding a Dice similarity coefficient compared to ground-truth manual segmentations of 0.84 and 0.89 for the acetabulum and femur, respectively, whereas the Dice coefficient was 0.78 and 0.88 for the model based on transfer learning. The Accuracy for the two anatomical regions was 0.95 and 0.97 when using data augmentation, and 0.87 and 0.96 when using transfer learning. Data augmentation can improve the performance of deep learning models by increasing the diversity of the training dataset and making the models more robust to noise and variations in image quality. The proposed segmentation model could be combined with radiomic analysis for the automatic evaluation of hip pathologies.

While musculoskeletal imaging volumes are increasing, there is a relative shortage of subspecialized musculoskeletal radiologists to interpret the studies. Will artificial intelligence (AI) be the solution? For AI to be the solution, the wide implementation of AI-supported data acquisition methods in clinical practice requires establishing trusted and reliable results. This implementation will demand close collaboration between core AI researchers and clinical radiologists. Upon successful clinical implementation, a wide variety of AI-based tools can improve the musculoskeletal radiologist's workflow by triaging imaging examinations, helping with image interpretation, and decreasing the reporting time. Additional AI applications may also be helpful for business, education, and research purposes if successfully integrated into the daily practice of musculoskeletal radiology. The question is not whether AI will replace radiologists, but rather how musculoskeletal radiologists can take advantage of AI to enhance their expert capabilities.

PURPOSE/OBJECTIVE:To characterize the safety, efficacy, and potential role of genicular artery embolization (GAE) as a disease-modifying treatment for symptomatic knee osteoarthritis (OA). MATERIALS AND METHODS/METHODS:This is an interim analysis of a prospective, single-arm clinical trial of patients with symptomatic knee OA who failed conservative therapy for greater than 3 months. Sixteen patients who underwent GAE using 250-μm microspheres and had at least 1 month of follow-up were included. Six patients completed the 12-month follow-up, and 10 patients remain enrolled. Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was evaluated at baseline and at 1, 3, and 12 months. Serum and plasma samples were collected for biomarker analysis. The primary end point was the percentage of patients who achieved the minimal clinically important difference (MCID) for WOMAC pain score at 12 months. Baseline and follow-up outcomes were analyzed using the Wilcoxon matched-pairs signed-rank test. RESULTS:Technical success of the procedure was 100%, with no major adverse events. The MCID was achieved in 5 of the 6 (83%) patients at 12 months. The mean WOMAC pain score decreased from 8.6 ± 2.7 at baseline to 4.9 ± 2.7 (P = .001), 4.4 ± 2.8 (P < .001), and 4.7 ± 2.7 (P = .094) at 1, 3, and 12 months, respectively. There was a statistically significant decrease in nerve growth factor (NGF) levels at 12 months. The remaining 8 biomarkers showed no significant change at 12 months. CONCLUSIONS:GAE is a safe and efficacious treatment for symptomatic knee OA. Decreased NGF levels after GAE may contribute to pain reduction and slowing of cartilage degeneration.

Deep learning (DL) is one of the most exciting new areas in medical imaging. This article will provide a review of current applications of DL in osteoarthritis (OA) imaging, including methods used for cartilage lesion detection, OA diagnosis, cartilage segmentation, and OA risk assessment. DL techniques have been shown to have similar diagnostic performance as human readers for detecting and grading cartilage lesions within the knee on MRI. A variety of DL methods have been developed for detecting and grading the severity of knee OA and various features of knee OA on X-rays using standardized classification systems with diagnostic performance similar to human readers. Multiple DL approaches have been described for fully automated segmentation of cartilage and other knee tissues and have achieved higher segmentation accuracy than currently used methods with substantial reductions in segmentation times. Various DL models analyzing baseline X-rays and MRI have been developed for OA risk assessment. These models have shown high diagnostic performance for predicting a wide variety of OA outcomes, including the incidence and progression of radiographic knee OA, the presence and progression of knee pain, and future total knee replacement. The preliminary results of DL applications in OA imaging have been encouraging. However, many DL techniques require further technical refinement to maximize diagnostic performance. Furthermore, the generalizability of DL approaches needs to be further investigated in prospective studies using large image datasets acquired at different institutions with different imaging hardware before they can be implemented in clinical practice and research studies.

BACKGROUND:Multiparameter characterization using MR fingerprinting (MRF) can quantify multiple relaxation parameters of intervertebral disc (IVD) simultaneously. These parameters may vary by age and sex. PURPOSE/OBJECTIVE:To investigate age- and sex-related differences in the relaxation parameters of the IVD of the lumbar spine using a multiparameter MRF technique. STUDY TYPE/METHODS:Prospective. SUBJECTS/METHODS:17 healthy subjects (8 male; mean age = 34 ± 10 years, range 20-60 years). FIELD STRENGTH/SEQUENCE/UNASSIGNED:maps at 3.0T. ASSESSMENT/RESULTS:maps. STATISTICAL TESTS/METHODS:of IVD. Statistical significance was defined as P-value <0.05. RESULTS:contrast (R = 0.709). CONCLUSION/CONCLUSIONS:of IVD in healthy subjects. LEVEL OF EVIDENCE/METHODS:2 TECHNICAL EFFICACY: Stage 1.