Multiaxial 3D MRI of the Ankle: Advanced High-Resolution Visualization of Ligaments, Tendons, and Articular Cartilage.

MRI is a valuable tool for diagnosing a broad spectrum of acute and chronic ankle disorders, including ligament tears, tendinopathy, and osteochondral lesions. Traditional two-dimensional (2D) MRI provides a high image signal and contrast of anatomic structures for accurately characterizing articular cartilage, bone marrow, synovium, ligaments, tendons, and nerves. However, 2D MRI limitations are thick slices and fixed slice orientations. In clinical practice, 2D MRI is limited to 2 to 3 mm slice thickness, which can cause blurred contours of oblique structures due to volume averaging effects within the image slice. In addition, image plane orientations are fixated and cannot be changed after the scan, resulting in 2D MRI lacking multiplanar and multiaxial reformation abilities for individualized image plane orientations along oblique and curved anatomic structures, such as ankle ligaments and tendons. In contrast, three-dimensional (3D) MRI is a newer, clinically available MRI technique capable of acquiring high-resolution ankle MRI data sets with isotropic voxel size. The inherently high spatial resolution of 3D MRI permits up to five times thinner (0.5 mm) image slices. In addition, 3D MRI can be acquired image voxel with the same edge length in all three space dimensions (isotropism), permitting unrestricted multiplanar and multiaxial image reformation and postprocessing after the MRI scan. Clinical 3D MRI of the ankle with 0.5 to 0.7 mm isotropic voxel size resolves the smallest anatomic ankle structures and abnormalities of ligament and tendon fibers, osteochondral lesions, and nerves. After acquiring the images, operators can align image planes individually along any anatomic structure of interest, such as ligaments and tendons segments. In addition, curved multiplanar image reformations can unfold the entire course of multiaxially curved structures, such as perimalleolar tendons, into one image plane. We recommend adding 3D MRI pulse sequences to traditional 2D MRI protocols to visualize small and curved ankle structures to better advantage. This article provides an overview of the clinical application of 3D MRI of the ankle, compares diagnostic performances of 2D and 3D MRI for diagnosing ankle abnormalities, and illustrates clinical 3D ankle MRI applications.

Conventional vs. endoscopic-assisted curettage of benign bone tumours. An experimental study.

BACKGROUND: This experimental study aimed at directly comparing conventional and endoscopic-assisted curettage towards (1) amount of residual tumour tissue (RTT) and (2) differences between techniques regarding surgical time and surgeons' experience level. METHODS: Three orthopaedic surgeons (trainee, consultant, senior consultant) performed both conventional (4x each) and endoscopic-assisted curettages (4x each) on specifically prepared cortical-soft cancellous femur and tibia sawbone models. "Tumours" consisted of radio-opaque polyurethane-based foam injected into prepared holes. Pre- and postinterventional CT-scans were carried out and RTT assessed on CT-scans. For statistical analyses, percentage of RTT in relation to total lesion's volume was used. T-tests, Wilcoxon rank-sum tests, and Kruskal-Wallis tests were applied to assess differences between surgeons and surgical techniques regarding RTT and timing. RESULTS: Median overall RTT was 1% (IQR 1 - 4%). Endoscopic-assisted curettage was associated with lower amount of RTT (median, 1%, IQR 0 - 5%) compared to conventional curettage (median, 4%, IQR 0 - 15%, p = 0.024). Mean surgical time was prolonged with endoscopic-assisted (9.2 ± 2.9 min) versus conventional curettage (5.9 ± 2.0 min; p = 0.004). No significant difference in RTT amount (p = 0.571) or curetting time (p = 0.251) depending on surgeons' experience level was found. CONCLUSIONS: Endoscopic-assisted curettage appears superior to conventional curettage regarding complete tissue removal, yet at expenses of prolonged curetting time. In clinical practice, this procedure may be reserved for cases at high risk of recurrence (e.g. anatomy, histology).

Fixed Subluxation of the Distal Radioulnar Joint following Malunion of the Ulnar Styloid Process.

Purpose Recent literature on significance and treatment of ulnar styloid fractures suggest that the ulnar styloid is not the central problem but the radioulnar ligaments and their impact on joint stability. However, specifically displaced ulnar styloid process fractures that secondarily heal in an ectopic position remain a rare situation for which diagnostics and treatment options remain a topic of discussion. Methods This case series presents four patients with limited supination due to a fixed dorsal subluxation of the distal radioulnar joint (DRUJ). The reason here fore was a significant malunion of ulnar styloid fracture that was addressed by corrective ulnar styloid osteotomy. Three of these osteotomies used three-dimensional (3D) preoperative planning and patient specific guides. Results All patients presented with a significant displacement of the malunited ulnar styloid fracture (average 32-degree rotation and 5-mm translation). In all four patients, the fixed subluxation of the ulnar head was resolved clinically and radiographically, and the forearm rotation restored after corrective osteotomy of the ulnar styloid and fixation in an anatomical position. Conclusion This case series presents a very specific subset of patients with nonanatomically healed ulnar styloid fractures responsible for a chronic DRUJ dislocation and limited prosupination and its treatment. Level of evidence This is a Level IV, therapeutic study.

Cerebrospinal Fluid Pressure Dynamics as a Bedside Test in Traumatic Spinal Cord Injury to Assess Surgical Spinal Cord Decompression: Safety, Feasibility, and Proof-of-Concept.

BACKGROUND: Sufficient and timely spinal cord decompression is a critical surgical objective for neurological recovery in spinal cord injury (SCI). Residual cord compression may be associated with disturbed cerebrospinal fluid pressure (CSFP) dynamics. OBJECTIVES: This study aims to assess whether intrathecal CSFP dynamics in SCI following surgical decompression are feasible and safe, and to explore the diagnostic utility. METHODS: Prospective cohort study. Bedside lumbar CSFP dynamics and cervical MRI were obtained following surgical decompression in N = 9 with mostly cervical acute-subacute SCI and N = 2 patients with non-traumatic SCI. CSFP measurements included mean CSFP, cardiac-driven CSFP peak-to-valley amplitudes (CSFPp), Valsalva maneuver, and Queckenstedt's test (firm pressure on jugular veins, QT). From QT, proxies for cerebrospinal fluid pulsatility curve were calculated (ie, relative pulse pressure coefficient; RPPC-Q). CSFP metrics were compared to spine-healthy patients. computer tomography (CT)-myelography was done in 3/8 simultaneous to CSFP measurements. RESULTS: Mean age was 45 ± 9 years (range 17-67; 3F), SCI was complete (AIS A, N = 5) or incomplete (AIS B-D, N = 6). No adverse events related to CSFP assessments. CSFP rise during QT was induced in all patients [range 9.6-26.6 mmHg]. However, CSFPp was reduced in 3/11 (0.1-0.3 mmHg), and in 3/11 RPPC-Q was abnormal (0.01-0.05). Valsalva response was reduced in 8/11 (2.6-23.4 mmHg). CSFP dynamics corresponded to CT-myelography. CONCLUSIONS: Comprehensive bedside lumbar CSFP dynamics in SCI following decompression are safe, feasible, and can reveal distinct patterns of residual spinal cord compression. Longitudinal studies are required to define critical thresholds of impaired CSFP dynamics that may impact neurological recovery and requiring surgical revisions.

Radiomics and Deep Learning for Disease Detection in Musculoskeletal Radiology: An Overview of Novel MRI- and CT-Based Approaches.

ABSTRACT: Radiomics and machine learning-based methods offer exciting opportunities for improving diagnostic performance and efficiency in musculoskeletal radiology for various tasks, including acute injuries, chronic conditions, spinal abnormalities, and neoplasms. While early radiomics-based methods were often limited to a smaller number of higher-order image feature extractions, applying machine learning-based analytic models, multifactorial correlations, and classifiers now permits big data processing and testing thousands of features to identify relevant markers. A growing number of novel deep learning-based methods describe magnetic resonance imaging- and computed tomography-based algorithms for diagnosing anterior cruciate ligament tears, meniscus tears, articular cartilage defects, rotator cuff tears, fractures, metastatic skeletal disease, and soft tissue tumors. Initial radiomics and deep learning techniques have focused on binary detection tasks, such as determining the presence or absence of a single abnormality and differentiation of benign versus malignant. Newer-generation algorithms aim to include practically relevant multiclass characterization of detected abnormalities, such as typing and malignancy grading of neoplasms. So-called delta-radiomics assess tumor features before and after treatment, with temporal changes of radiomics features serving as surrogate markers for tumor responses to treatment. New approaches also predict treatment success rates, surgical resection completeness, and recurrence risk. Practice-relevant goals for the next generation of algorithms include diagnostic whole-organ and advanced classification capabilities. Important research objectives to fill current knowledge gaps include well-designed research studies to understand how diagnostic performances and suggested efficiency gains of isolated research settings translate into routine daily clinical practice. This article summarizes current radiomics- and machine learning-based magnetic resonance imaging and computed tomography approaches for musculoskeletal disease detection and offers a perspective on future goals and objectives.

Extent of posterolateral tibial plateau impaction fracture correlates with anterolateral complex injury and has an impact on functional outcome after ACL reconstruction.

PURPOSE: The impact of posterolateral tibial plateau impaction fractures (TPIF) on posttraumatic knee stability in the setting of primary anterior cruciate ligament (ACL) tear is unknown. The main objective was to determine whether increased bone loss of the posterolateral tibial plateau is associated with residual rotational instability and impaired functional outcome after ACL reconstruction. METHODS: A cohort was identified in a prospective enrolled study of patients suffering acute ACL injury who underwent preoperative standard radiographic diagnostics and clinical evaluation. Patients were included when scheduled for isolated single-bundle hamstring autograft ACL reconstruction. Exclusion criteria were concurrent anterolateral complex (ALC) reconstruction (anterolateral tenodesis), previous surgery or symptoms in the affected knee, partial ACL tear, multi-ligament injury with an indication for additional surgical intervention, and extensive cartilage wear. On MRI, bony (TPIF, tibial plateau, and femoral condyle morphology) and ligament status (ALC, concomitant collateral ligament, and meniscus injuries) were assessed by a musculoskeletal radiologist. Clinical evaluation consisted of KT-1000, pivot-shift, and Lachman testing, as well as Tegner activity and IKDC scores. RESULTS: Fifty-eight patients were included with a minimum follow-up of 12 months. TPIF was identified in 85% of ACL injuries (n = 49). The ALC was found to be injured in 31 of 58 (53.4%) cases. Pearson analysis showed a positive correlation between TPIF and the degree of concomitant ALC injury (p < 0.001). Multiple regression analysis revealed an increased association of high-grade TPIF with increased lateral tibial convexity (p = 0.010). The high-grade TPIF group showed worse postoperative Tegner scores 12 months postoperatively (p = 0.035). CONCLUSION: Higher degrees of TPIFs are suggestive of a combined ACL/ALC injury. Moreover, patients with increased posterolateral tibial plateau bone loss showed lower Tegner activity scores 12 months after ACL reconstruction. LEVEL OF EVIDENCE: III.

Diagnostic Performance of Advanced Metal Artifact Reduction MRI for Periprosthetic Shoulder Infection.

BACKGROUND: The diagnosis of periprosthetic shoulder infection (PSI) in patients with a painful arthroplasty is challenging. Magnetic resonance imaging (MRI) may be helpful, but shoulder implant-induced metal artifacts degrade conventional MRI. Advanced metal artifact reduction (MARS) improves the visibility of periprosthetic bone and soft tissues. The purpose of our study was to determine the reliability, repeatability, and diagnostic performance of advanced MARS-MRI findings for diagnosing PSI. METHODS: Between January 2015 and December 2019, we enrolled consecutive patients suspected of having PSI at our academic hospital. All 89 participants had at least 1-year clinical follow-up and underwent standardized clinical, radiographic, and laboratory evaluations and advanced MARS-MRI. Two fellowship-trained musculoskeletal radiologists retrospectively evaluated the advanced MARS-MRI studies for findings associated with PSI in a blinded and independent fashion. Both readers repeated their evaluations after a 2-month interval. Interreader reliability and intrareader repeatability were assessed with κ coefficients. The diagnostic performance of advanced MARS-MRI for PSI was quantified using sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC). When applying the International Consensus Meeting (ICM) 2018 criteria, of the 89 participants, 22 (25%) were deemed as being infected and 67 (75%) were classified as being not infected (unlikely to have PSA and not requiring a surgical procedure during 1-year follow-up). RESULTS: The interreader reliability and intrareader repeatability of advanced MARS-MRI findings, including lymphadenopathy, joint effusion, synovitis, extra-articular fluid collection, a sinus tract, rotator cuff muscle edema, and periprosthetic bone resorption, were good (κ = 0.61 to 0.80) to excellent (κ > 0.80). Lymphadenopathy, complex joint effusion, and edematous synovitis had sensitivities of >85%, specificities of >90%, odds ratios of >3.6, and AUC values of >0.90 for diagnosing PSI. The presence of all 3 findings together yielded a PSI probability of >99%, per logistic regression analysis. CONCLUSIONS: Our study shows the clinical utility of advanced MARS-MRI for diagnosing PSI when using the ICM 2018 criteria as the reference standard. Although the reliability and diagnostic accuracy were high, these conclusions are based on our specific advanced MARS-MRI protocol interpreted by experienced musculoskeletal radiologists. Investigations with larger sample sizes are needed to confirm these results. LEVEL OF EVIDENCE: Diagnostic Level III . See Instructions for Authors for a complete description of levels of evidence.

Bone marrow edema of the medioplantar talar head is associated with severe ligamentous injury in ankle sprain.

PURPOSE: To investigate the predictive value of talar head edema (THE) in acute ankle sprain for the presence of concomitant ligament injuries. METHODS: This retrospective study was approved by the ethics committee and informed consent was obtained. One hundred patients (mean age: 37 years ± 14 [standard deviation], range 13-77 years) with MRI of the ankle after acute trauma were included. The cohort in this matched-pair study consisted of 50 patients with THE (group 1) and 50 patients without THE (group 2). Two readers independently evaluated presence and size of bone marrow edema of the talus head and injuries of the lateral, medial, talonavicular, and spring ligament complex. Statistics included intraclass correlation coefficient (ICC) and Kappa statistics as well as parametric and non-parametric tests. RESULTS: On average, patients with THE demonstrated significantly more ligament injuries in comparison to patients without THE (3.7 vs. 1.3, p ≤ 0.01). Also, in patients with THE, the number of injured ligaments was significantly higher at the lateral (p = 0.03), medial (p ≤ 0.01), and talonavicular (p ≤ 0.01) compartment in comparison to patients without THE. The most frequently injured ligaments in patients with THE were the anterior talofibular ligament (60%) and the anterior tibiotalar ligament (42%). There was no significant correlation between edema size and the number of injured ligaments or compartments (p = 0.5). CONCLUSION: THE is associated with more extensive ligamentous ankle injury, in particular to the medial and lateral collateral ligament complex, and therefore indicative of severe ankle trauma.

Artificial intelligence for MRI diagnosis of joints: a scoping review of the current state-of-the-art of deep learning-based approaches.

Deep learning-based MRI diagnosis of internal joint derangement is an emerging field of artificial intelligence, which offers many exciting possibilities for musculoskeletal radiology. A variety of investigational deep learning algorithms have been developed to detect anterior cruciate ligament tears, meniscus tears, and rotator cuff disorders. Additional deep learning-based MRI algorithms have been investigated to detect Achilles tendon tears, recurrence prediction of musculoskeletal neoplasms, and complex segmentation of nerves, bones, and muscles. Proof-of-concept studies suggest that deep learning algorithms may achieve similar diagnostic performances when compared to human readers in meta-analyses; however, musculoskeletal radiologists outperformed most deep learning algorithms in studies including a direct comparison. Earlier investigations and developments of deep learning algorithms focused on the binary classification of the presence or absence of an abnormality, whereas more advanced deep learning algorithms start to include features for characterization and severity grading. While many studies have focused on comparing deep learning algorithms against human readers, there is a paucity of data on the performance differences of radiologists interpreting musculoskeletal MRI studies without and with artificial intelligence support. Similarly, studies demonstrating the generalizability and clinical applicability of deep learning algorithms using realistic clinical settings with workflow-integrated deep learning algorithms are sparse. Contingent upon future studies showing the clinical utility of deep learning algorithms, artificial intelligence may eventually translate into clinical practice to assist detection and characterization of various conditions on musculoskeletal MRI exams.

External snapping hip syndrome is associated with an increased femoral offset.

BACKGROUND: External snapping hip syndrome (ESH) is postulated to be one of the causes of greater trochanteric pain syndrome, which also includes greater trochanteric bursitis and tendinopathy or tears of the hip abductor mechanism. However, it was not yet described what kind of bony morphology can cause the snapping and whether symptomatic and asymptomatic individuals have different imaging features. PURPOSE: It was the purpose of this study to look for predisposing morphological factors for ESH and to differentiate between painful and asymptomatic snapping. METHODS: A consecutive cohort with ESH and available magnetic resonance imaging (MRI) between 2014 and 2019 was identified. The control group consisted of patients that underwent corrective osteotomies around the knee for mechanical axis correction and never complained of hip symptoms nor had undergone previous hip procedures. The following parameters were blindly assessed for determination of risk factors for ESH: CCD (corpus collum diaphysis) angle; femoral and global offset; femoral antetorsion; functional femoral antetorsion; translation of the greater trochanter (GT); posterior tilt of the GT; pelvic width/anterior pelvic length; intertrochanteric width. Hip and pelvic offset indexes were calculated as ratios of femoral/global offset and intertrochanteric/pelvic width, respectively. For the comparison of symptomatic and asymptomatic snapping, the following soft-tissue signs were investigated: presence of trochanteric bursitis or gluteal tendinopathy; presence of surface bony irregularities on trochanter major and ITB (Iliotibial band) thickness. RESULTS: A total of 31 hips with ESH were identified. The control group (n = 29) consisted of patients matched on both age (± 1) and gender. Multiple regression analysis determined an increased hip offset index to be independent predictor of ESH (r = + 0.283, p = 0.025), most likely due to the higher femoral offset in the ESH group (p = 0.031). Pearson correlation analysis could not identify any significant secondary factors. No differences were found between painful and asymptomatic snapping on MRI. CONCLUSIONS: A high hip offset index was found as an independent predictor for external snapping hip in our cohort, mainly due to increased femoral offset. No imaging soft-tissue related differences could be outlined between symptomatic and asymptomatic external snapping. LEVEL OF EVIDENCE III: This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors   www.springer.com/00590 .

Sternoclavicular joint injections and their predictive value for the outcome of surgery.

BACKGROUND: Sternoclavicular joint injections are one of the first-line treatment options for painful non-infectious pathologies of the sternoclavicular joint; however, their long-term effects and predictive value in decision-making for surgery are yet to be defined. METHODS: 27/32 Patients who received in total 36 computed tomography-guided sternoclavicular joint injections in 2012-2017 replied the questionnaire with a mean follow-up of 38 months after the first sternoclavicular joint injection. Of those patients, seven underwent subsequent surgery. We evaluated pain response after sternoclavicular joint injection and American Shoulder and Elbow Surgeons-score at later follow-up. RESULTS: Directly after sternoclavicular joint injection, pain decreased from Visual Analog Scale 5.3 ± 2.4 to 3.8 ± 3 (p = 0.001). In the seven patients who underwent surgery for degenerative changes, definitive outcome correlated with pain relief after the last infiltration (r = 0.86, p = 0.012). Also, the final American Shoulder and Elbow Surgeons-score was lower in patients with multiple injections compared to those who were satisfied after the first injection (p = 0.019). DISCUSSION: Sternoclavicular injections are a useful tool in the context of degenerative sternoclavicular joint disorders as the amount of pain reduction is, in case the short-term effect is not long-lasting, at least a strong indicator for the future success of operative treatment.

Posttraumatic Carpal Instability Nondissociative.

Background Posttraumatic midcarpal instability nondissociative (CIND) is an exceptional rare condition, therefore the outcome after different treatment options remains unknown. Questions The purpose of this study was to investigate the different treatment options for posttraumatic CIND. We also describe the different radiological and magnetic resonance imaging (MRI) findings in this patient cohort. Patients and Methods We present outcomes of 10 patients who developed CIND following acute wrist trauma between 2007 and 2018, 3 with dorsal intercalated segment instability pattern (CIND-DISI) and 7 with volar intercalated segment instability (CIND-VISI) radiographically. Results Three patients with CIND-VISI had satisfactory outcomes with conservative treatment. Two patients with irreducible CIND-DISI and one with CIND-VISI underwent proximal row carpectomy (PRC), two with reducible CIND-VISI had radiolunate fusion, and two with secondary osteoarthritis had total wrist fusion. All patients with CIND-DISI needed surgery, whereas only four of the seven patients with CIND-VISI needed surgery. On MRI, all three patients with CIND-DISI had rupture of the radiolunate ligament. Conclusions The data collected in this study may provide the first step toward better understanding of the pathology for this exceptionally rare finding. In CIND-VISI, we have not seen any ligament injury in four patients. Therefore, conservative therapy is more likely to be the first step. In CIND-DISI, we recommend an operative procedure: if detected early, with ligament suture, otherwise by radiolunate fusion, PRC, or total wrist fusion. Level of Evidence This is a Level IV study.

Impact of 30 years' high-level rock climbing on the shoulder: an magnetic resonance imaging study of 31 climbers.

BACKGROUND: Rock climbers are particularly susceptible to shoulder injuries due to repetitive upper-limb movements on vertical or overhanging terrain. However, the long-term effects of prolonged climbing on the shoulder joints are still unknown. PURPOSE: The purpose of this study was to analyze the prevalence of pain and degenerative changes in the shoulder joints after high-level rock climbing over at least 25 years. We hypothesized that specific climber-associated patterns of degeneration would be found. METHODS: Thirty-one adult male high-level rock climbers were compared to an age- and sex-matched control group of 31 nonclimbers. All participants underwent a detailed interview, standardized clinical examination, and bilateral (climbers) or unilateral (nonclimbers, dominant side) magnetic resonance imaging (MRI) scans. Clinical and MRI findings of the groups were compared. RESULTS: The lifetime prevalence of shoulder pain in the rock climbers was 77%. The rock climbers had significantly more abnormalities in the labrum (82% vs. 52%; P = .002), long biceps tendon (53% vs. 23%; P = .006), and cartilage (28% vs. 3%; P = .005). These increased changes positively correlated with climbing intensity. There were no differences between the 2 groups with respect to rotator cuff tendon pathology (68% vs. 58%; P = .331) and acromioclavicular joint degeneration (88% vs. 90%; P = .713). Despite the increased degenerative changes in the rock climbers, their Constant score (CS) was still better than that of the nonclimbers (CS 94, interquartile range [IQR] 92-97, vs. CS 93, IQR 91-95; P = .019). CONCLUSIONS: Prolonged high-level rock climbing leads to a high prevalence of shoulder pain and increased degenerative changes to the labrum, long biceps tendon, and cartilage. However, it is not related to any restriction in shoulder function.

Osseous defect of the anteroinferior femoral head: is it associated with femoroacetabular impingement (FAI)?

OBJECTIVE: To evaluate the prevalence, morphology, and clinical significance of a repeatedly observed yet not examined circumscript osseous defect at the anteroinferior aspect of the femoral head, termed femoral head defect. MATERIALS AND METHODS: Retrospective study with approval of the institutional review board. There was informed consent by all individuals. Magnetic resonance imaging (MRI) hip examinations of 970 individuals (age 15 to 55) were analyzed for femoral head defect. Patients with femoral head defect were matched for age and gender with patients without defect. Two readers independently assessed MRI images regarding presence, location, and morphology of the defect. MR images and radiographs were analyzed for findings of femoroacetabular impingement (FAI). Femoral torsion was measured. Independent t test and chi-square test were used for statistics. RESULTS: Sixty-eight (7%) of 970 MRI examinations exhibited a femoral head defect in an anteroinferior location of the femoral head (29/400 men, 7.3%; 39/570 women, 6.8%; p = 0.8). The most frequent morphology of femoral head defect was type I, dent-like (34; 50%), followed by type II, crater-like (27; 40%), and III, cystic (7; 10%). Femoral head defect was slightly more common on the right hip (39 individuals; 57%) compared to left (29 individuals; 43%), non-significantly (p = 0.115). There was no association between FAI or its subtypes and the presence of femoral head defect (p = 0.890). Femoral antetorsion was reduced in patients with femoral head defect (12.9° ± 8.6) compared to patients without defect (15.2° ± 8.5), without statistical significance (p = 0.121). CONCLUSION: The femoral head defect is a common finding in MRI examinations of the hip and is situated in the anteroinferior location. There was no association with FAI yet a non-significant trend towards lower femoral antetorsion in patients with femoral head defects.

Correction to: Deep convolutional neural network-based detection of meniscus tears: comparison with radiologists and surgery as standard of reference.

The article "Deep convolutional neural network-based detection of meniscus tears: comparison with radiologists and surgery as standard of reference.

3D-printed anatomic models of the knee for evaluation of patellofemoral dysplasia in comparison to standard radiographs and computed tomography.

PURPOSE: To evaluate 3D-printed anatomic models of the distal femur and patella for diagnosis and classification of patellofemoral dysplasia in comparison to conventional radiographs (CR) and CT. METHOD: Following local ethics committee approval, CT-datasets of 50 patients were segmented and 3D-anatomic models of the distal femur and patella were printed. An expert panel reviewed CR, CT, 3D-models and patient history and classified the femoral trochleas into normal or Dejour type A-D and the patellas into Wiberg type A-C, which served as the standard of reference. The same classifications were performed by two readers independently, first based on 3D-models and after 3 weeks based on CR and CT. Descriptive statistics, ROC-analysis and inter-reader reliability were performed. RESULTS: Trochlear dysplasia was present in 28/50 patients. Evaluations of 3D-models vs. CR/CT for trochlear dysplasia showed a sensitivity/specificity of 89.3 %/100 % vs. 96.4 %/68.2 % for reader 1 and 96.4 %/100 % vs. 96.4 %/90.9 % for reader 2, and an area under the curve of 0.946 vs. 0.823 for reader 1 (p = 0.029) and 0.982 vs. 0.937 for reader 2 (p = 0.147). Evaluations of 3D-models vs. CR/CT for the Dejour classification showed a sensitivity/specificity of 32.1 %/100 % vs. 57.1 %/68.2 % for reader 1 and 46.4 %/100 % vs. 50 %/90.9 % for reader 2 without significant differences. No significant differences existed for Wiberg-classification (50-66 % exact matches) or inter-reader reliabilities between 3D-models and CR/CT for all assessments (Kappa 0.428-0.92). CONCLUSION: In comparison to radiographs and CT, 3D-models achieve similar diagnostic accuracy for detection of patellofemoral dysplasia and have the potential to improve diagnosis for less experienced physicians.

Deep convolutional neural network-based detection of meniscus tears: comparison with radiologists and surgery as standard of reference.

OBJECTIVE: To clinically validate a fully automated deep convolutional neural network (DCNN) for detection of surgically proven meniscus tears. MATERIALS AND METHODS: One hundred consecutive patients were retrospectively included, who underwent knee MRI and knee arthroscopy in our institution. All MRI were evaluated for medial and lateral meniscus tears by two musculoskeletal radiologists independently and by DCNN. Included patients were not part of the training set of the DCNN. Surgical reports served as the standard of reference. Statistics included sensitivity, specificity, accuracy, ROC curve analysis, and kappa statistics. RESULTS: Fifty-seven percent (57/100) of patients had a tear of the medial and 24% (24/100) of the lateral meniscus, including 12% (12/100) with a tear of both menisci. For medial meniscus tear detection, sensitivity, specificity, and accuracy were for reader 1: 93%, 91%, and 92%, for reader 2: 96%, 86%, and 92%, and for the DCNN: 84%, 88%, and 86%. For lateral meniscus tear detection, sensitivity, specificity, and accuracy were for reader 1: 71%, 95%, and 89%, for reader 2: 67%, 99%, and 91%, and for the DCNN: 58%, 92%, and 84%. Sensitivity for medial meniscus tears was significantly different between reader 2 and the DCNN (p = 0.039), and no significant differences existed for all other comparisons (all p ≥ 0.092). The AUC-ROC of the DCNN was 0.882, 0.781, and 0.961 for detection of medial, lateral, and overall meniscus tear. Inter-reader agreement was very good for the medial (kappa = 0.876) and good for the lateral meniscus (kappa = 0.741). CONCLUSION: DCNN-based meniscus tear detection can be performed in a fully automated manner with a similar specificity but a lower sensitivity in comparison with musculoskeletal radiologists.

Deep Convolutional Neural Network-Based Diagnosis of Anterior Cruciate Ligament Tears: Performance Comparison of Homogenous Versus Heterogeneous Knee MRI Cohorts With Different Pulse Sequence Protocols and 1.5-T and 3-T Magnetic Field Strengths.

OBJECTIVES: The aim of this study was to clinically validate a Deep Convolutional Neural Network (DCNN) for the detection of surgically proven anterior cruciate ligament (ACL) tears in a large patient cohort and to analyze the effect of magnetic resonance examinations from different institutions, varying protocols, and field strengths. MATERIALS AND METHODS: After ethics committee approval, this retrospective analysis of prospectively collected data was performed on 512 consecutive subjects, who underwent knee magnetic resonance imaging (MRI) in a total of 59 different institutions followed by arthroscopic knee surgery at our institution. The DCNN and 3 fellowship-trained full-time academic musculoskeletal radiologists evaluated the MRI examinations for full-thickness ACL tears independently. Surgical reports served as the reference standard. Statistics included diagnostic performance metrics, including sensitivity, specificity, area under the receiver operating curve ("AUC ROC"), and kappa statistics. P values less than 0.05 were considered to represent statistical significance. RESULTS: Anterior cruciate ligament tears were present in 45.7% (234/512) and absent in 54.3% (278/512) of the subjects. The DCNN had a sensitivity of 96.1%, which was not significantly different from the readers (97.5%-97.9%; all P ≥ 0.118), but significantly lower specificity of 93.1% (readers, 99.6%-100%; all P < 0.001) and "AUC ROC" of 0.935 (readers, 0.989-0.991; all P < 0.001) for the entire cohort. Subgroup analysis showed a significantly lower sensitivity, specificity, and "AUC ROC" of the DCNN for outside MRI (92.5%, 87.1%, and 0.898, respectively) than in-house MRI (99.0%, 94.4%, and 0.967, respectively) examinations (P = 0.026, P = 0.043, and P < 0.05, respectively). There were no significant differences in DCNN performance for 1.5-T and 3-T MRI examinations (all P ≥ 0.753, respectively). CONCLUSIONS: Deep Convolutional Neural Network performance of ACL tear diagnosis can approach performance levels similar to fellowship-trained full-time academic musculoskeletal radiologists at 1.5 T and 3 T; however, the performance may decrease with increasing MRI examination heterogeneity.

Editorial Commentary: Is the Lateral Center-Edge Angle Sufficient for the Approximation of Acetabular Coverage?

The lateral center-edge angle is a robust technique with proven clinical applicability for quantifying acetabular coverage. However, it measures only the lateral coverage of the femoral head without consideration of other important portions of the acetabulum such as the anterior or posterior wall. Three-dimensional acetabular coverage measurement techniques capable of quantifying the entire acetabulum have become available, thus posing the question of whether we should still rely on the lateral center-edge angle for the assessment of acetabular coverage.