Search

Purpose: Subchondral insufficiency fractures (SIF) and advanced osteoarthritis (OA) of the knee are usually seen in conjunction with bone marrow lesions (BMLs) and their differentiation may pose a significant diagnostic challenge. We aimed to develop a convolutional neural network (CNN) ensemble which could successfully differentiate between these two entities., Materials and Methods: A total of 212 knees with SIF and 102 knees with advanced OA with BMLs were retrospectively included. Coronal fat suppressed PD-w images were augmented, resized and normalized, reaching a total of 1174 images. Data was used to fine-tune three ImageNet-pretrained CNNs (VGG-16, InceptionV3 and Inception-ResNet-V2). Agreement of at least two networks was recorded as the decision of the network ensemble. Ensemble performance was compared to that of two MSK radiologists on the validation set. Receiver operating characteristics (ROC) curves and the respective areas under the curve (AUC) were used to evaluate human and machine performance., Results: InceptionV3 achieved the highest AUC (93.68%) and VGG-16 the lowest AUC (82.18%) among individual CNNs. CNN ensemble achieved the highest overall performance with an AUC of 95.97%. The first of the two MSK radiologists achieved a performance similar to the ensemble, reaching an AUC of 91.95%. The second radiologist achieved lower AUC of 82.76% which was lower than both the other specialist and the ensemble (P < 0.001)., Conclusion: A CNN ensemble was highly accurate in differentiating between SIF and OA, achieving a higher or equal performance to MSK radiologists., Competing Interests: Declarations of Competing Interest None, (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Over the past decade rapid advancements in molecular imaging (MI) and artificial intelligence (AI) have revolutionized traditional musculoskeletal radiology. Molecular imaging refers to the ability of various methods to in vivo characterize and quantify biological processes, at a molecular level. The extracted information provides the tools to understand the pathophysiology of diseases and thus to early detect, to accurately evaluate the extend and to apply and evaluate targeted treatments. At present, molecular imaging mainly involves CT, MRI, radionuclide, US, and optical imaging and has been reported in many clinical and preclinical studies. Although originally MI techniques targeted at central nervous system disorders, later on their value on musculoskeletal disorders was also studied in depth. Meaningful exploitation of the large volume of imaging data generated by molecular and conventional imaging techniques, requires state-of-the-art computational methods that enable rapid handling of large volumes of information. AI allows end-to-end training of computer algorithms to perform tasks encountered in everyday clinical practice including diagnosis, disease severity classification and image optimization. Notably, the development of deep learning algorithms has offered novel methods that enable intelligent processing of large imaging datasets in an attempt to automate decision-making in a wide variety of settings related to musculoskeletal trauma. Current applications of AI include the diagnosis of bone and soft tissue injuries, monitoring of the healing process and prediction of injuries in the professional sports setting. This review presents the current applications of novel MI techniques and methods and the emerging role of AI regarding the diagnosis and evaluation of musculoskeletal trauma., Competing Interests: Declaration of Competing Interest No conflict of interest to disclose for all authors, (Copyright © 2020 Elsevier Ltd. All rights reserved.)

MRI has been established as an essential tool for accurate diagnosis in patients with musculoskeletal trauma. Its major advantages include excellent soft tissue contrast, high spatial resolution and lack of ionizing radiation. Although plain radiographs remain the basic tool for diagnosis and treatment planning in bone fractures assisted by CT in pelvic, spine and large joints injuries, there are specific circumstances that require MRI. For instance, tendinous, ligamentous, intraarticular structures such as the cartilage and menisci, and intramedullary injury are seen mostly with MRI. Volumetric 3D techniques are now commercially available and provide higher spatial resolution which improves anatomic detail, allows multiplanar reformations and reduces the acquisition time. Newer applications on quantitative rather than morphologic imaging, such as relaxometry and diffusion tensor imaging, may be of paramount importance in treatment planning in the near future. Software improvements reduce metal induced artefacts, allowing thus imaging of the postoperative patient with metallic implants. A tendency towards a structured reporting pattern and standardised medical communication needs to be further explored for the benefit of orthopaedic surgeons, radiologists and patients., (Copyright © 2014 Elsevier Ltd. All rights reserved.)

We present the case of a 62-year-old patient suffering from the presence of a cervical spinal cord intramedullary teratoma and treated operatively. The purpose of this case report is to describe the highly unusual localization of the intramedullary teratoma associated with other vertebral malformations. A review of the literature is also presented.

Brucellosis is a zoonosis of worldwide distribution presenting with a wide clinical spectrum. Brucellosis can involve any organ or system. The axial skeleton is the most common site of involvement with a frequency ranging from 2% to 53%. Multiple-level spinal involvements are rare. This report describes the first case of noncontiguous synchronous multifocal involvement of all cervical, thoracic, and lumbar regions in a patient with brucellar spondylodiscitis.

Retroclival epidural hematomas (RCEDH) make part of posterior fossa epidural hematomas (PFEDH) and represent an extremely rare entity not always easily diagnosed with computerized tomography (CT) due to beam hardening artifacts. The authors present a case of a child-to our knowledge-the first reported until now-featured the rare combination of a longitudinal clivus fracture associated with concomitant epidural hematoma treated conservatively with favorable outcome of the patient.

The purpose of this study was to compare the relative conspicuity of the pancreaticobiliary tree on respiratory-triggered three-dimensional turbo spin echo (3D TSE RT) and breath hold single-shot turbo spin echo (SSTSE BH) acquisitions respectively in MRCP imaging. Both techniques were applied to 61 patients with clinically suspected pancreaticobiliary disease using a 1.0 T MR system. All images were reviewed blindly. Qualitative comparison was made by grading subjectively the conspicuity of extrahepatic, intrahepatic, and main pancreatic ducts. Quantitative comparison included calculations of signal-to-noise ratio of the common bile duct, main pancreatic duct, gallbladder, liver, and contrast-to-noise ratio, relative contrast between common bile duct, gallbladder, and liver. 3D TSE RT provided significantly higher signal-to-noise ratio of the common bile duct (mean value 163.19) and main pancreatic duct (mean value 95.37) compared to SSTSE BH (mean values 76.24 and 26.22, respectively). 3D TSE RT was inferior to SSTSE BH for the depiction of intrahepatic ducts and pancreatic duct (head portion). 3D TSE RT and SSTSE BH sequences provide complimentary information in the visualization of the biliary and pancreatic ducts. Further comparative clinical studies are needed to redefine the sensitivity, specificity, and accuracy of MRCP using both sequences.

The purpose of this study was to evaluate the ability of three dimensional T1-weighted multi-shot Echo Planar Imaging (3D T1w EPI) MR pulse sequence to provide comparable to T1w Spin Echo (SE) results in various diseases of the brain, during shorter acquisition times. Thirty-six patients (aged 30-74 years) with various indications were included in the study. All examinations were performed with a 1T MR scanner with a maximum gradient strength of 15 mT/m. The SE sequence lasted 3 min 50s and the 3D T1w EPI 59s. The quantitative analysis included number of enhancing lesions, signal-to-noise ratio of the enhancing lesions and contrast-to-noise ratio (CNR) between enhancing lesions and white matter in both sequences before and after i.v. administration of 0.1 mmol/kg gadopentetate dimeglumine. In addition, the percentage increase of enhancement was measured in each lesion of each sequence. The qualitative analysis included a) conspicuity of the lesions and b) presence of artifacts. The T1w SE sequence was significantly better compared to 3D T1w EPI in all quantitative measurements with the exception of CNR of enhancing lesions before contrast administration and the percentage enhancement. The conspicuity of the lesions did not differ between the two sequences. The EPI sequence presented with significantly more artifacts. We conclude that the 3D T1w EPI sequence could not be used instead of the conventional T1w SE, in routine imaging of the brain. Its overall diagnostic capability, could be useful only in uncooperative patients.