Automatic Deep Learning-assisted Detection and Grading of Abnormalities in Knee MRI Studies.

SAT0435 Tibial Spine Morphology and Association with Anterior Cruciate Ligament Degeneration – A Combined MRI, X-rays and Histology Study of Early Knee Osteoarthritis

New books

The aim of this work was to study anterior cruciate ligament (ACL) degeneration in relation to MRI-based morphological knee abnormalities and cartilage T2 relaxation times in subjects with symptomatic osteoarthritis. Two radiologists screened the right knee MRI of 304 randomly selected participants in the Osteoarthritis Initiative cohort with symptomatic OA, for ACL abnormalities. Of the 52 knees with abnormalities, 28 had mucoid degeneration, 12 had partially torn ACLs, and 12 had completely torn ACLs. Fifty-three randomly selected subjects with normal ACLs served as controls. Morphological knee abnormalities were graded using the WORMS score. Cartilage was segmented and compartment-specific T2 values were calculated. Compared to normal ACL knees, those with ACL abnormalities had a greater prevalence of, and more severe, cartilage, meniscal, bone marrow, subchondral cyst, and medial collateral ligament lesions (all p < 0.05). T2 measurements did not significantly differ by ACL status. ACL abnormalities were associated with more severe degenerative changes, likely because of greater joint instability. T2 measurements may not be well suited to assess advanced cartilage degeneration.

Bone contusions correlate with severity and location of articular injuries after knee sprains

Engineering Stem and Stromal Cell Therapies for Musculoskeletal Tissue Repair.

MRI study of the ligamentization of ACL grafts in children with open growth plates

The knee joint is very important for everyone as it helps us in movements, which is essential for everyone.One of the most diseases that injure the knee is the anterior cruciate ligament (ACL).This work has developed a computer aided diagnosis (CAD) system for examining the given knee magnetic resonance imaging (MRI) image and automatically determining if there is a torn in ACL or not.The region growing based segmentation algorithm is used to get the region of interest (ROI) from MRI image, e.g., extract ACL region from knee image then CNN-XGBoost model is used for knee ACL classification.The model is divided into two main parts: the first part extract the feature uses CNN and the second part using XGBoost for feature classification.The designed model gives us an accuracy of 91%.

BackgroundBone marrow concentrate (BMC) has shown promise in the treatment of several orthopedic conditions. This registry study investigated the use of autologous BMC and platelet products for percutaneous anterior cruciate ligament (ACL) treatment.MethodsTwenty-nine patients presenting to a single outpatient interventional musculoskeletal and pain practice with symptomatic grade 1, 2, or 3 ACL tears with less than 1 cm retraction were enrolled. Patients were treated with a percutaneous ACL injection of autologous BMC and platelet products using fluoroscopic guidance. Pre- and post-treatment magnetic resonance imaging analysis was completed for 23 patients using ImageJ software for an objective quantitative analysis of pixel density as a proxy for ACL integrity. Subjective clinical outcome measures collected pre-treatment and at 1, 3, 6, 12, 18, 24, and 36 months post-treatment include the Numerical Pain Scale (NPS), the Lower Extremity Functional Scale (LEFS), the International Knee Documentation Committee (IKDC) form, and a modified version of the Single Assessment Numeric Evaluation.ResultsSeventy-seven percent of patients treated with BMC injections into the ACL showed significant improvement (p < 0.01) in objective measures of ACL integrity at an average of 8.8 months (median 4.7 months). The mean of last patient-reported improvement was 72% (SD = 35) at an average of 23 (SD = 10) months post-treatment. Mean scores were found to be significantly different (p < 0.05) for the NPS at 6, 18, and 24 months, and LEFS and IKDC at all time points (i.e. 1, 3, 6, 12, 18, 24, and 36 months) relative to baseline.ConclusionIn symptomatic patients with grade 1, 2, or even grade 3 tears with minimal retraction, ACL treatment with percutaneous injection of BMC and platelet products shows promise as a non-surgical alternative. However, a larger randomized controlled trial is warranted to confirm these findings.Trial registration NCT03011398. A Clinical Registry of Orthobiologics Procedures. https://clinicaltrials.gov/ct2/show/NCT03011398?term=orthobiologics&rank=1. Registered 29 December 2016. Enrollment 1 December 2011-retrospectively registered

Vertebral fractures are the most common osteoporotic fractures in older individuals. Recent studies suggest that the performance of artificial intelligence is equal to humans in detecting osteoporotic fractures, such as fractures of the hip, distal radius, and proximal humerus. However, whether artificial intelligence performs as well in the detection of vertebral fractures on plain lateral spine radiographs has not yet been reported. (1) What is the accuracy, sensitivity, specificity, and interobserver reliability (kappa value) of an artificial intelligence model in detecting vertebral fractures, based on Genant fracture grades, using plain lateral spine radiographs compared with values obtained by human observers? (2) Do patients' clinical data, including the anatomic location of the fracture (thoracic or lumbar spine), T-score on dual-energy x-ray absorptiometry, or fracture grade severity, affect the performance of an artificial intelligence model? (3) How does the artificial intelligence model perform on external validation? Between 2016 and 2018, 1019 patients older than 60 years were treated for vertebral fractures in our institution. Seventy-eight patients were excluded because of missing CT or MRI scans (24% [19]), poor image quality in plain lateral radiographs of spines (54% [42]), multiple myeloma (5% [4]), and prior spine instrumentation (17% [13]). The plain lateral radiographs of 941 patients (one radiograph per person), with a mean age of 76 ± 12 years, and 1101 vertebral fractures between T7 and L5 were retrospectively evaluated for training (n = 565), validating (n = 188), and testing (n = 188) of an artificial intelligence deep-learning model. The gold standard for diagnosis (ground truth) of a vertebral fracture is the interpretation of the CT or MRI reports by a spine surgeon and a radiologist independently. If there were any disagreements between human observers, the corresponding CT or MRI images would be rechecked by them together to reach a consensus. For the Genant classification, the injured vertebral body height was measured in the anterior, middle, and posterior third. Fractures were classified as Grade 1 (< 25%), Grade 2 (26% to 40%), or Grade 3 (> 40%). The framework of the artificial intelligence deep-learning model included object detection, data preprocessing of radiographs, and classification to detect vertebral fractures. Approximately 90 seconds was needed to complete the procedure and obtain the artificial intelligence model results when applied clinically. The accuracy, sensitivity, specificity, interobserver reliability (kappa value), receiver operating characteristic curve, and area under the curve (AUC) were analyzed. The bootstrapping method was applied to our testing dataset and external validation dataset. The accuracy, sensitivity, and specificity were used to investigate whether fracture anatomic location or T-score in dual-energy x-ray absorptiometry report affected the performance of the artificial intelligence model. The receiver operating characteristic curve and AUC were used to investigate the relationship between the performance of the artificial intelligence model and fracture grade. External validation with a similar age population and plain lateral radiographs from another medical institute was also performed to investigate the performance of the artificial intelligence model. The artificial intelligence model with ensemble method demonstrated excellent accuracy (93% [773 of 830] of vertebrae), sensitivity (91% [129 of 141]), and specificity (93% [644 of 689]) for detecting vertebral fractures of the lumbar spine. The interobserver reliability (kappa value) of the artificial intelligence performance and human observers for thoracic and lumbar vertebrae were 0.72 (95% CI 0.65 to 0.80; p < 0.001) and 0.77 (95% CI 0.72 to 0.83; p < 0.001), respectively. The AUCs for Grades 1, 2, and 3 vertebral fractures were 0.919, 0.989, and 0.990, respectively. The artificial intelligence model with ensemble method demonstrated poorer performance for discriminating normal osteoporotic lumbar vertebrae, with a specificity of 91% (260 of 285) compared with nonosteoporotic lumbar vertebrae, with a specificity of 95% (222 of 234). There was a higher sensitivity 97% (60 of 62) for detecting osteoporotic (dual-energy x-ray absorptiometry T-score ≤ -2.5) lumbar vertebral fractures, implying easier detection, than for nonosteoporotic vertebral fractures (83% [39 of 47]). The artificial intelligence model also demonstrated better detection of lumbar vertebral fractures compared with detection of thoracic vertebral fractures based on the external dataset using various radiographic techniques. Based on the dataset for external validation, the overall accuracy, sensitivity, and specificity on bootstrapping method were 89%, 83%, and 95%, respectively. The artificial intelligence model detected vertebral fractures on plain lateral radiographs with high accuracy, sensitivity, and specificity, especially for osteoporotic lumbar vertebral fractures (Genant Grades 2 and 3). The rapid reporting of results using this artificial intelligence model may improve the efficiency of diagnosing vertebral fractures. The testing model is available at http://140.113.114.104/vght_demo/corr/. One or multiple plain lateral radiographs of the spine in the Digital Imaging and Communications in Medicine format can be uploaded to see the performance of the artificial intelligence model. Level II, diagnostic study.

This Letter presents a stable polyp-scene classification method with low false positive (FP) detection. Precise automated polyp detection during colonoscopies is essential for preventing colon-cancer deaths. There is, therefore, a demand for a computer-assisted diagnosis (CAD) system for colonoscopies to assist colonoscopists. A high-performance CAD system with spatiotemporal feature extraction via a three-dimensional convolutional neural network (3D CNN) with a limited dataset achieved about 80% detection accuracy in actual colonoscopic videos. Consequently, further improvement of a 3D CNN with larger training data is feasible. However, the ratio between polyp and non-polyp scenes is quite imbalanced in a large colonoscopic video dataset. This imbalance leads to unstable polyp detection. To circumvent this, the authors propose an efficient and balanced learning technique for deep residual learning. The authors’ method randomly selects a subset of non-polyp scenes whose number is the same number of still images of polyp scenes at the beginning of each epoch of learning. Furthermore, they introduce post-processing for stable polyp-scene classification. This post-processing reduces the FPs that occur in the practical application of polyp-scene classification. They evaluate several residual networks with a large polyp-detection dataset consisting of 1027 colonoscopic videos. In the scene-level evaluation, their proposed method achieves stable polyp-scene classification with 0.86 sensitivity and 0.97 specificity.

The anterior cruciate ligament (ACL) is essential for the stability of the knee joint and ACL tears are one of the most common sports injuries with ahigh incidence, especially in sports that require rotational movements and abrupt changes in direction. Injuries of the ACL are rarely isolated and are often accompanied by meniscal and other internal knee injuries, which increase the risk of osteoarthritis. The spectrum of ACL injuries includes strains, partial tears and complete tears. Magnetic resonance imaging (MRI) plays apivotal role in the diagnostics as it can accurately depict not only the ACL but also accompanying injuries. Proton density and T2-weighted sequences are particularly suitable for evaluating the ACL, which is usually well visible and assessable in all planes. In addition to depicting fiber disruption as adirect sign and central diagnostic indicator of an ACL tear, there are numerous other direct and indirect signs of an ACL injury in MRI. These include abnormal fiber orientations, signal increases and an anterior subluxation of the tibia relative to the femur. The bone marrow edema patterns often associated with ACL tears are indicative of the underlying injury mechanism. The treatment of ACL tears can be conservative or surgical depending on various factors, such as the patient's activity level and the presence of accompanying injuries. The precise and comprehensive description of ACL injuries by radiology is crucial for optimal treatment planning. Anterolateral rotational instability (ALRI) of the knee joint characterizes acondition of excessive lateral and rotational mobility of the tibia in relation to the femur in the anterolateral knee region. This instability is primarily caused by arupture of the ACL, with the anterolateral ligament (ALL) that was rediscovered about 10 years ago, also being attributed arole in stabilizing the knee. Although ALRI is primarily diagnosed through clinical examinations, MRI is indispensable for detecting injuries to the ACL, ALL, and other internal knee structures, which is essential for developing an optimal treatment strategy.

The aim of the present study was to investigate the validity and reliability of the deep lateral femoral notch sign (DLFNS) in identifying a concomitant anterior cruciate ligament (ACL)/anterolateral ligament (ALL) rupture and predicting the clinical outcomes following an anatomical single-bundle ACL reconstruction. It was hypothesized that patients with a concomitant ACL/ALL rupture would have an increased DLFNS compared to patients without a concomitant ACL/ALL rupture. The lateral preoperative radiographs and MRI images of 100 patients with an ACL rupture and 100 control subjects were evaluated for the presence of a DLFNS and ACL/ALL rupture, respectively. The patients were evaluated clinically preoperatively and at a minimum 1 year following the ACL reconstruction. A receiver operator curve (ROC) analysis was performed to define the optimal cut-off value of the DLFNS for identifying a concomitant ACL/ALL injury. The relative risk (RR) was also calculated to determine whether the presence of the DLFNS was a risk factor for residual instability or ACL graft rupture following an ACL reconstruction. The prevalence of DLFNS was 52% in the ACL-ruptured patients and 15% in the control group. At a minimum 1-year follow-up, 35% (6/17) of the patients with DLFNS > 1.8mm complained of persistent instability, and an MRI evaluation demonstrated a graft re-rupture rate of 12% (2/17). In patients with a DLFNS < 1.8mm, 8% (7/83) reported a residual instability, and the graft rupture rate was 2.4% (2/83). A DLFNS > 1.8mm demonstrated a sensitivity of 89%, a specificity of 95%, a negative predictive value of 98%, and a positive predictive value of 89% in identifying a concomitant ACL/ALL rupture. Patients with a DLFNS > 1.8mm had 4.2 times increased risk for residual instability and graft rupture compared to patients with a DLFNS ≤ 1.8mm. A DLFNS > 1.8mm could be a clinically relevant diagnostic tool for identifying a concomitant ACL/ALL rupture with high sensitivity and PPV. Patients with a DLFNS > 1.8mm should be carefully evaluated for clinical and radiological signs of a concomitant ACL/ALL rupture and treated when needed with a combined intra-articular ACL reconstruction and extra-articular tenodesis to avoid a residual rotational instability and ACL graft rupture. III.

Due to the characteristics of cruciate ligaments, knee ligament tears are not suitable to heal by remedy. Anterior cruciate ligament (ACL) reconstruction is one of the most common and reliable surgeries to handle knee ligament injuries. To date, most knee ligament reconstructions can be done by using endoscopic surgery. Although single bundle ACL reconstructions could improve the recovery of knee function, yet it was still not enough for the stability of the knee joint. In some cases instability might even develop to be osteoarthritis. Via some vitro tests, such as varus, valgus and rotation tests, the test results showed that the ligament after single bundle ACL reconstruction was not able to bear the rotational, and varus/valgus torque at the knee joint. That means using conventional arthrometer such as KT-1000 could lead to misdiagnosis in the above mentioned situation. We developed a new instrument to measure the extent of tibia's varus, valgus and rotation relative to the femur. Via the test data, the instrument could quantify the knee laxity and help surgeon to diagnose the injured knee more accurately and verify whether an ACL reconstruction was successful. The main function of the instrument is to test knee ligament laxity by measuring the varus/valgus angulation, and internal/external rotation with high sensitivity and data reproducibility. We verified the instrument's performance via experiments. The results of the experiments demonstrated the usability of the arthrometer.

Simple SummaryDistinguishing between benign vs. malignant bone lesions is often difficult on imaging. Many bone lesions are infrequent or rarely seen, and often only specialist radiologists have sufficient expertise to provide an accurate diagnosis. In addition, some benign bone tumours may exhibit potentially aggressive features that mimic malignant bone tumours, making the diagnosis even more difficult. The rapid development of artificial intelligence (AI) techniques has led to remarkable progress in image-recognition tasks, including the classification and characterization of various tumours. This study will review the most recent evidence for AI techniques in differentiating bone lesions on various imaging modalities using a systematic approach. Potential clinical applications of AI techniques for clinical diagnosis and management of bone lesions will also be discussed.An accurate diagnosis of bone tumours on imaging is crucial for appropriate and successful treatment. The advent of Artificial intelligence (AI) and machine learning methods to characterize and assess bone tumours on various imaging modalities may assist in the diagnostic workflow. The purpose of this review article is to summarise the most recent evidence for AI techniques using imaging for differentiating benign from malignant lesions, the characterization of various malignant bone lesions, and their potential clinical application. A systematic search through electronic databases (PubMed, MEDLINE, Web of Science, and clinicaltrials.gov) was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 34 articles were retrieved from the databases and the key findings were compiled and summarised. A total of 34 articles reported the use of AI techniques to distinguish between benign vs. malignant bone lesions, of which 12 (35.3%) focused on radiographs, 12 (35.3%) on MRI, 5 (14.7%) on CT and 5 (14.7%) on PET/CT. The overall reported accuracy, sensitivity, and specificity of AI in distinguishing between benign vs. malignant bone lesions ranges from 0.44–0.99, 0.63–1.00, and 0.73–0.96, respectively, with AUCs of 0.73–0.96. In conclusion, the use of AI to discriminate bone lesions on imaging has achieved a relatively good performance in various imaging modalities, with high sensitivity, specificity, and accuracy for distinguishing between benign vs. malignant lesions in several cohort studies. However, further research is necessary to test the clinical performance of these algorithms before they can be facilitated and integrated into routine clinical practice.

To compare the biological characteristics of bone marrow mesenchymal stem cells (BMSCs) and anterior cruciate ligament derived mesenchymal stem cells (ACL-MSCs) from rats in vitro. Ten male SPF-level BN rats, weighing 200-220 g, were selected to obtain anterior cruciate ligaments and bone marrows, and ACL-MSCs and BMSCs were isolated for passage culture respectively under sterile condition. The cell morphology was observed, and the cells at passage 3 were used to detect the surface markers of CD34, CD45, CD90, and CD29 by flow cytometry, the ability of cell proliferation by cell counting kit 8 (CCK-8), and colony formation ability by clone forming test. The mRNA levels of differentiation related genes [alkaline phosphatas (ALP), bone gamma-carboxyglutamate protein, runt related transcription factor 2, bone morphogenetic protein 2 (BMP-2), secreted phosphoprotein 1 (Spp1), collagen type II α1 (Col2α1), Aggrecan (Acan), Sox9, peroxisome proliferator activated receptor γ2 (PPARγ2), and CCAAT-enhancer-binding protein-α] were also determined by real-time fluorescent quantitative PCR. BMSCs and ACL-MSCs had similar morphology, adherent cells displaying long fusiform. The immunoprofile of ACL-MSCs and BMSCs at passage 3 was positive for CD29 and CD90 and was negative for CD45 and CD34. The absorbance ( A) value of ACL-MSCs (1.11±0.08) was significantly higher than that of BMSCs (0.78±0.05) ( t=3.599, P=0.023); the number of colonies of ACL-MSCs [(53.00±5.51)/hole] was significantly more than that of BMSCs [(30.67±4.84)/hole] ( t=3.045, P=0.038). The results of toluidine blue staining, alizarin red staining, and oil red O staining were positive in BMSCs and ACL-MSCs at 21 days after osteogenic, chondrogenic, and adipogenic induction. The mRNA expressions of BMP-2, Spp1, Col2α1, Acan, Sox9, and PPARγ2 in ACL-MSCs were significantly higher than those in BMSCs ( P<0.01). The proliferation potential of ACL-MSCs is greater than that of BMSCs, and the former is apt to differentiate into chondrocytes. ACL-MSCs are promising cells to promote tendon-bone healing.