Search

This review examines the latest advancements in compositional and quantitative cartilage MRI techniques, addressing both their potential and challenges. The integration of these advancements promises to improve disease detection, treatment monitoring, and overall patient care. We want to highlight the pivotal task of translating these techniques into widespread clinical use, the transition of cartilage MRI from technical validation to clinical application, emphasizing its critical role in identifying early signs of degenerative and inflammatory joint diseases. Recognizing these changes early may enable informed treatment decisions, thereby facilitating personalized medicine approaches. The evolving landscape of cartilage MRI underscores its increasing importance in clinical practice, offering valuable insights for patient management and therapeutic interventions. This review aims to discuss the old evidence and new insights about the evaluation of articular cartilage through MRI, with an update on the most recent literature published on novel quantitative sequences.

Abstract Total hip arthroplasty (THA) is the best surgical approach for treating advanced hip degeneration, providing pain relief, and improved function in most cases. In the past, MR imaging quality has been highly compromised by in-plane distortions, inadequate fat saturation, and other artifacts due to metal components of THA. Technological advancements have made pathologic conditions, which were previously hidden by periprosthetic artifacts, outstanding features due to the optimization of several sequences. To date, several short and long-term complications involving bony and soft-tissue structures may be detected through magnetic resonance imaging (MRI). The use of MRI with adapted sequences and protocols may drastically reduce artifacts thereby providing essential pre-operative elements for planning revision surgery of failed THA. This review has the purpose of conveying new insights to musculoskeletal radiologists about the techniques to suppress metal-related artifacts and the hallmark MRI findings of painful THA. Critical relevance statement Advancements in metal-suppression have given radiologists the opportunity to play an emerging role in THA management. This article provides technical and imaging insights into challenges that can be encountered in cases of THA, which may present complications and characteristic imaging findings. Key Points Imaging total hip arthroplasty requires adapted MRI protocol and awareness of the common complications. We have reported the available metal-suppression sequences for evaluating total hip arthroplasty. Many structures and conditions should be considered when dealing with painful aseptic or septic arthroplasty. Graphical Abstract

Abstract Artificial intelligence (AI) has demonstrated great potential in a wide variety of applications in interventional radiology (IR). Support for decision-making and outcome prediction, new functions and improvements in fluoroscopy, ultrasound, computed tomography, and magnetic resonance imaging, specifically in the field of IR, have all been investigated. Furthermore, AI represents a significant boost for fusion imaging and simulated reality, robotics, touchless software interactions, and virtual biopsy. The procedural nature, heterogeneity, and lack of standardisation slow down the process of adoption of AI in IR. Research in AI is in its early stages as current literature is based on pilot or proof of concept studies. The full range of possibilities is yet to be explored. Relevance statement Exploring AI’s transformative potential, this article assesses its current applications and challenges in IR, offering insights into decision support and outcome prediction, imaging enhancements, robotics, and touchless interactions, shaping the future of patient care. Key points • AI adoption in IR is more complex compared to diagnostic radiology. • Current literature about AI in IR is in its early stages. • AI has the potential to revolutionise every aspect of IR. Graphical Abstract

Computed tomography (CT) arthrography is a quickly available imaging modality to investigate elbow disorders. Its excellent spatial resolution enables the detection of subtle pathologic changes of intra-articular structures, which makes this technique extremely valuable in a joint with very tiny chondral layers and complex anatomy of articular capsule and ligaments. Radiation exposure has been widely decreased with the novel CT scanners, thereby increasing the indications of this examination. The main applications of CT arthrography of the elbow are the evaluation of capsule, ligaments, and osteochondral lesions in both the settings of acute trauma, degenerative changes, and chronic injury due to repeated microtrauma and overuse. In this review, we discuss the normal anatomic findings, technical tips for injection and image acquisition, and pathologic findings that can be encountered in CT arthrography of the elbow, shedding light on its role in the diagnosis and management of different orthopedic conditions. We aspire to offer a roadmap for the integration of elbow CT arthrography into routine clinical practice, fostering improved patient outcomes and a deeper understanding of elbow pathologies.

Abstract This narrative review focuses on clinical applications of artificial intelligence (AI) in musculoskeletal imaging. A range of musculoskeletal disorders are discussed using a clinical-based approach, including trauma, bone age estimation, osteoarthritis, bone and soft-tissue tumors, and orthopedic implant-related pathology. Several AI algorithms have been applied to fracture detection and classification, which are potentially helpful tools for radiologists and clinicians. In bone age assessment, AI methods have been applied to assist radiologists by automatizing workflow, thus reducing workload and inter-observer variability. AI may potentially aid radiologists in identifying and grading abnormal findings of osteoarthritis as well as predicting the onset or progression of this disease. Either alone or combined with radiomics, AI algorithms may potentially improve diagnosis and outcome prediction of bone and soft-tissue tumors. Finally, information regarding appropriate positioning of orthopedic implants and related complications may be obtained using AI algorithms. In conclusion, rather than replacing radiologists, the use of AI should instead help them to optimize workflow, augment diagnostic performance, and keep up with ever-increasing workload. Relevance statement This narrative review provides an overview of AI applications in musculoskeletal imaging. As the number of AI technologies continues to increase, it will be crucial for radiologists to play a role in their selection and application as well as to fully understand their potential value in clinical practice. Key points • AI may potentially assist musculoskeletal radiologists in several interpretative tasks. • AI applications to trauma, age estimation, osteoarthritis, tumors, and orthopedic implants are discussed. • AI should help radiologists to optimize workflow and augment diagnostic performance. Graphical Abstract

Abstract Objective To systematically review radiomic feature reproducibility and model validation strategies in recent studies dealing with CT and MRI radiomics of bone and soft-tissue sarcomas, thus updating a previous version of this review which included studies published up to 2020. Methods A literature search was conducted on EMBASE and PubMed databases for papers published between January 2021 and March 2023. Data regarding radiomic feature reproducibility and model validation strategies were extracted and analyzed. Results Out of 201 identified papers, 55 were included. They dealt with radiomics of bone (n = 23) or soft-tissue (n = 32) tumors. Thirty-two (out of 54 employing manual or semiautomatic segmentation, 59%) studies included a feature reproducibility analysis. Reproducibility was assessed based on intra/interobserver segmentation variability in 30 (55%) and geometrical transformations of the region of interest in 2 (4%) studies. At least one machine learning validation technique was used for model development in 34 (62%) papers, and K-fold cross-validation was employed most frequently. A clinical validation of the model was reported in 38 (69%) papers. It was performed using a separate dataset from the primary institution (internal test) in 22 (40%), an independent dataset from another institution (external test) in 14 (25%) and both in 2 (4%) studies. Conclusions Compared to papers published up to 2020, a clear improvement was noted with almost double publications reporting methodological aspects related to reproducibility and validation. Larger multicenter investigations including external clinical validation and the publication of databases in open-access repositories could further improve methodology and bring radiomics from a research area to the clinical stage. Critical relevance statement An improvement in feature reproducibility and model validation strategies has been shown in this updated systematic review on radiomics of bone and soft-tissue sarcomas, highlighting efforts to enhance methodology and bring radiomics from a research area to the clinical stage. Key points • 2021–2023 radiomic studies on CT and MRI of musculoskeletal sarcomas were reviewed. • Feature reproducibility was assessed in more than half (59%) of the studies. • Model clinical validation was performed in 69% of the studies. • Internal (44%) and/or external (29%) test datasets were employed for clinical validation. Graphical Abstract

Abstract Background The aim of this systematic review is to evaluate the diagnostic performance of Artificial Intelligence (AI) models designed for the detection of caries lesion (CL). Materials and methods An electronic literature search was conducted on PubMed, Web of Science, SCOPUS, LILACS and Embase databases for retrospective, prospective and cross-sectional studies published until January 2023, using the following keywords: artificial intelligence (AI), machine learning (ML), deep learning (DL), artificial neural networks (ANN), convolutional neural networks (CNN), deep convolutional neural networks (DCNN), radiology, detection, diagnosis and dental caries (DC). The quality assessment was performed using the guidelines of QUADAS-2. Results Twenty articles that met the selection criteria were evaluated. Five studies were performed on periapical radiographs, nine on bitewings, and six on orthopantomography. The number of imaging examinations included ranged from 15 to 2900. Four studies investigated ANN models, fifteen CNN models, and two DCNN models. Twelve were retrospective studies, six cross-sectional and two prospective. The following diagnostic performance was achieved in detecting CL: sensitivity from 0.44 to 0.86, specificity from 0.85 to 0.98, precision from 0.50 to 0.94, PPV (Positive Predictive Value) 0.86, NPV (Negative Predictive Value) 0.95, accuracy from 0.73 to 0.98, area under the curve (AUC) from 0.84 to 0.98, intersection over union of 0.3–0.4 and 0.78, Dice coefficient 0.66 and 0.88, F1-score from 0.64 to 0.92. According to the QUADAS-2 evaluation, most studies exhibited a low risk of bias. Conclusion AI-based models have demonstrated good diagnostic performance, potentially being an important aid in CL detection. Some limitations of these studies are related to the size and heterogeneity of the datasets. Future studies need to rely on comparable, large, and clinically meaningful datasets. Protocol PROSPERO identifier: CRD42023470708

In patients with total hip arthroplasty (THA) with recurrent pain, symptoms may be caused by several conditions involving not just the joint, but also the surrounding soft tissues including tendons, muscles, bursae, and peripheral nerves. US and US-guided interventional procedures are important tools in the diagnostic work-up of patients with painful THA given that it is possible to reach a prompt diagnosis both directly identifying the pathological changes of periprosthetic structures and indirectly evaluating the response and pain relief to local injection of anesthetics under US monitoring. Then, US guidance can be used for the aspiration of fluid from the joint or periarticular collections, or alternatively to follow the biopsy needle to collect samples for culture analysis in the suspicion of prosthetic joint infection. Furthermore, US-guided percutaneous interventions may be used to treat several conditions with well-established minimally invasive procedures that involve injections of corticosteroid, local anesthetics, and platelet-rich plasma or other autologous products. In this review, we will discuss the clinical and technical applications of US-guided percutaneous interventional procedures in painful THA that can be used in routine daily practice for diagnostic and therapeutic purposes.

Summary: Background: Atypical cartilaginous tumour (ACT) and high-grade chondrosarcoma (CS) of long bones are respectively managed with active surveillance or curettage and wide resection. Our aim was to determine diagnostic performance of X-rays radiomics-based machine learning for classification of ACT and high-grade CS of long bones. Methods: This retrospective, IRB-approved study included 150 patients with surgically treated and histology-proven lesions at two tertiary bone sarcoma centres. At centre 1, the dataset was split into training (n = 71 ACT, n = 24 high-grade CS) and internal test (n = 19 ACT, n = 6 high-grade CS) cohorts, respectively, based on the date of surgery. At centre 2, the dataset constituted the external test cohort (n = 12 ACT, n = 18 high-grade CS). Manual segmentation was performed on frontal view X-rays, using MRI or CT for preliminary identification of lesion margins. After image pre-processing, radiomic features were extracted. Dimensionality reduction included stability, coefficient of variation, and mutual information analyses. In the training cohort, after class balancing, a machine learning classifier (Support Vector Machine) was automatically tuned using nested 10-fold cross-validation. Then, it was tested on both the test cohorts and compared to two musculoskeletal radiologists' performance using McNemar's test. Findings: Five radiomic features (3 morphology, 2 texture) passed dimensionality reduction. After tuning on the training cohort (AUC = 0.75), the classifier had 80%, 83%, 79% and 80%, 89%, 67% accuracy, sensitivity, and specificity in the internal (temporally independent) and external (geographically independent) test cohorts, respectively, with no difference compared to the radiologists (p ≥ 0.617). Interpretation: X-rays radiomics-based machine learning accurately differentiates between ACT and high-grade CS of long bones. Funding: AIRC Investigator Grant.

Abstract Background Humans should sleep for about a third of their lifetime and the choice of the mattress is very important from a quality-of-life perspective. Therefore, the primary aim of this study was to assess the changes of lumbar angles, evaluated in a supine position using magnetic resonance imaging (MRI), on a mattress versus a rigid surface. Methods Twenty healthy subjects (10 females, 10 males), aged 32.3 ± 6.5 (mean ± standard deviation), with body mass index 22.4 ± 2.9, completed three evaluations: (i) spine MRI in supine position on a mattress (MAT); (ii) spine MRI in supine position on rigid surface (CON); and (iii) biplanar radiographic imaging in standing position. The following indexes were calculated for both MAT and CON: lumbar lordosis angles L1–L5, L1–S1, L5–S1, and the sacral slope (SS). Further, pelvic incidence (PI) was calculated from the biplanar radiographic images. Results Main findings were (i) L1–L5 and SS were greater in MAT than CON (L1:L5: +2.9°; SS: +2.0°); (ii) L5–S1 was lower in MAT than CON (−1.6°); (iii) L1–S1 was greater in MAT than CON only for male subjects (+2.0°); (iv) significant and positive correlations between PI and L1–L5, L1–S1 and SS were observed in both CON and MAT. Conclusions The use of a mattress determined small but statistically significant changes in lumbar angles. Relevance statement The use of a mattress determines small but statistically significant changes in radiological angles describing the sagittal alignment of the lumbar spine when lying in the supine position. Key points • Lordosis angle L1–L5 was greater in MAT than in CON condition (+2.9°). • Sacral slope was greater in MAT than in CON condition (+2.0°). • Lordosis angle L5–S1 was lower in MAT than in CON condition (−1.6°). Graphical Abstract

Abstract Purpose To test the technical feasibility of an augmented reality (AR) navigation system to guide bone biopsies. Methods We enrolled patients subjected to percutaneous computed tomography (CT)-guided bone biopsy using a novel AR navigation system. Data from prospectively enrolled patients (AR group) were compared with data obtained retrospectively from previous standard CT-guided bone biopsies (control group). We evaluated the following: procedure duration, number of CT passes, patient’s radiation dose (dose-length product), complications, and specimen adequacy. Technical success was defined as the ability to complete the procedure as planned, reaching the target center. Technical efficacy was assessed evaluating specimen adequacy. Results Eight patients (4 males) aged 58 ± 24 years (mean ± standard deviation) were enrolled in the AR group and compared with 8 controls (4 males) aged 60 ± 15 years. No complications were observed. Procedure duration, number of CT passes, and radiation dose were 22 ± 5 min, 4 (median) [4, 6 interquartile range] and 1,034 ± 672 mGy*cm for the AR group and 23 ± 5 min, 9 [7.75, 11.25], and 1,954 ± 993 mGy*cm for controls, respectively. No significant differences were observed for procedure duration (p = 0.878). Conversely, number of CT passes and radiation doses were significantly lower for the AR group (p

Bone Strain Index (BSI), based on dual-energy X-ray absorptiometry (DXA), is a densitometric index of bone strength of the femur and lumbar spine. Higher BSI values indicate a higher strain applied to bone, predisposing to higher fracture risk. This retrospective, multicentric study on Italian women reports the BSI normative age-specific reference curves. A cohort of Caucasian Italian women aged 20 to 90 years was selected from three different clinical centres. Bone mineral density (BMD) and BSI measurements were obtained for the lumbar spine vertebrae (L1–L4) and for the femur (neck, trochanter and intertrochanter) using Hologic densitometers scans. The data were compared with BMD normative values provided by the densitometer manufacturer. Then, the age-specific BSI curve for the femur and lumbar spine was generated. No significant difference was found between the BMD of the subjects in this study and BMD reference data provided by Hologic (p = 0.68 for femur and p = 0.90 for lumbar spine). Spine BSI values (L1–L4) increase by 84% between 20 and 90 years of age. The mean BSI of the total femur increases about 38% in the same age range. The BSI age-specific reference curve could help clinicians improve osteoporosis patient management, allowing an appropriate patient classification according to the bone resistance to the applied loads and fragility fracture risk assessment.

Abstract Bizarre parosteal osteochondromatous proliferation (BPOP) is a surface-based bone lesion belonging to the group of benign chondrogenic tumors. The aim of this review is to familiarize the readers with imaging features and differential diagnosis of BPOP, also addressing pathological presentation and treatment options. The peak of incidence of BPOP is in the third and fourth decades of life, although it can occur at any age. Hands are the most common location of BPOP (55%), followed by feet (15%) and long bones (25%). On imaging, BPOP appears as a well-marginated mass of heterotopic mineralization arising from the periosteal aspect of the bone. Typical features of BPOP are contiguity with the underlying bone and lack of cortico-medullary continuity, although cortical interruption and medullary involvement have been rarely reported. Histologically, BPOP is a benign bone surface lesion characterized by osteocartilaginous proliferation with disorganized admixture of cartilage with bizarre features, bone and spindle cells. Differential diagnosis includes both benign—such as florid reactive periostitis, osteochondroma, subungual exostosis, periosteal chondroma and myositis ossificans—and malignant lesions—such as periosteal chondrosarcoma and surface-based osteosarcoma. Treatment consists of surgical resection. Local recurrences are common and treated with re-excision. Critical relevance statement Bizarre parosteal osteochondromatous proliferation is a benign mineralized mass arising from the periosteal aspect of bone cortex. Multi-modality imaging characteristics, pathology features and differential diagnosis are here highlighted to familiarize the readers with this entity and offer optimal patient care.

Key points 1. Diffusion tensor imaging shows excellent inter-observer reliability. 2. Diffusion tensor imaging shows excellent inter-vendor reliability. 3. Quantitative results in musculoskeletal application of diffusion tensor imaging are reproducible in all the three magnetic resonance scanners. 4. DTI, as a reproducible magnetic resonance sequence, can be used for a quantitative evaluation of muscle microstructures during daily practice.

Abstract Background To compare electrical impedance myography (EIM) and MRI in assessing lumbar skeletal muscle composition. Methods One hundred forty-one patients (78 females, mean age 57 ± 19 years) were prospectively enrolled and underwent lumbar spine MRI, EIM with Skulpt®, and clinical evaluation including the questionnaire SARC-F. MRIs were reviewed to assess the Goutallier score of paravertebral muscles at L3 level and to calculate the cross sectional area (CSA) of both psoas, quadratus lumborum, erector spinae, and multifidus muscles on a single axial slice at L3 level, in order to calculate the skeletal muscle index (SMI=CSA/height2). We tested the correlation between EIM-derived parameters [body fat percentage (BF%) and muscle quality] and body mass index (BMI), Goutallier score (1–4), SMI, and SARC-F scores (0–10) using the Pearson correlation coefficient. The strength of association was considered large (0.5 to 1.0), medium (0.3 to 0.5), small (0.1 to 0.3). Results Pearson’s correlation coefficient showed small (0.26) but significant (p

IntroductionA novel classification scheme for endplate lesions, based on T2-weighted images from magnetic resonance imaging (MRI) scan, has been recently introduced and validated. The scheme categorizes intervertebral spaces as “normal,” “wavy/irregular,” “notched,” and “Schmorl's node.” These lesions have been associated with spinal pathologies, including disc degeneration and low back pain. The exploitation of an automatic tool for the detection of the lesions would facilitate clinical practice by reducing the workload and the diagnosis time. The present work exploits a deep learning application based on convolutional neural networks to automatically classify the type of lesion.MethodsT2-weighted MRI scans of the sagittal lumbosacral spine of consecutive patients were retrospectively collected. The middle slice of each scan was manually processed to identify the intervertebral spaces from L1L2 to L5S1, and the corresponding lesion type was labeled. A total of 1,559 gradable discs were obtained, with the following types of distribution: “normal” (567 discs), “wavy/irregular” (485), “notched” (362), and “Schmorl's node” (145). The dataset was divided randomly into a training set and a validation set while preserving the original distribution of lesion types in each set. A pretrained network for image classification was utilized, and fine-tuning was performed using the training set. The retrained net was then applied to the validation set to evaluate the overall accuracy and accuracy for each specific lesion type.ResultsThe overall rate of accuracy was found equal to 88%. The accuracy for the specific lesion type was found as follows: 91% (normal), 82% (wavy/irregular), 93% (notched), and 83% (Schmorl's node).DiscussionThe results indicate that the deep learning approach achieved high accuracy for both overall classification and individual lesion types. In clinical applications, this implementation could be employed as part of an automatic detection tool for pathological conditions characterized by the presence of endplate lesions, such as spinal osteochondrosis.

Background: Our aim was to understand how musculoskeletal training is structured in Italian residency programmes and the needs of young trainees. Methods: We sent out an online questionnaire (17 questions) to Italian Society of Radiology residents and board-certified radiologists aged up to 39 years. Results: A total of 1144 out of 4210 (27.2%) members participated in the survey; 64.7% were residents and 35.3% were board-certified radiologists. Just 26.6% of participants had dedicated rotations for musculoskeletal training during their residency, although this percentage substantially increased in replies from northern Italy. One-fourth of residents had a scheduled period of musculoskeletal ultrasound. Most participants (76.3%) had

The methylation of the O6-methylguanine-DNA methyltransferase (MGMT) promoter is a molecular marker associated with a better response to chemotherapy in patients with glioblastoma (GB). Standard pre-operative magnetic resonance imaging (MRI) analysis is not adequate to detect MGMT promoter methylation. This study aims to evaluate whether the radiomic features extracted from multiple tumor subregions using multiparametric MRI can predict MGMT promoter methylation status in GB patients. This retrospective single-institution study included a cohort of 277 GB patients whose 3D post-contrast T1-weighted images and 3D fluid-attenuated inversion recovery (FLAIR) images were acquired using two MRI scanners. Three separate regions of interest (ROIs) showing tumor enhancement, necrosis, and FLAIR hyperintensities were manually segmented for each patient. Two machine learning algorithms (support vector machine (SVM) and random forest) were built for MGMT promoter methylation prediction from a training cohort (196 patients) and tested on a separate validation cohort (81 patients), based on a set of automatically selected radiomic features, with and without demographic variables (i.e., patients’ age and sex). In the training set, SVM based on the selected radiomic features of the three separate ROIs achieved the best performances, with an average of 83.0% (standard deviation: 5.7%) for accuracy and 0.894 (0.056) for the area under the curve (AUC) computed through cross-validation. In the test set, all classification performances dropped: the best was obtained by SVM based on the selected features extracted from the whole tumor lesion constructed by merging the three ROIs, with 64.2% (95% confidence interval: 52.8–74.6%) accuracy and 0.572 (0.439–0.705) for AUC. The performances did not change when the patients’ age and sex were included with the radiomic features into the models. Our study confirms the presence of a subtle association between imaging characteristics and MGMT promoter methylation status. However, further verification of the strength of this association is needed, as the low diagnostic performance obtained in this validation cohort is not sufficiently robust to allow clinically meaningful predictions.

Abstract Background To evaluate the distribution of lingual foramina (LF) and their correlation with demographic characteristics and mandible width, shape, and bone thickness in Caucasian Italian patients subjected to cone-beam CT (CBCT). Methods CBCTs were reviewed to assess the number of all LF, midline and lateral LF. We also assessed the relationship of the number of lateral LF with gender and mandibular width, shape, and bone thickness using the Chi Square test. A p value .05). Conclusions We have shown the high variability of number and anatomic distribution of LF in an Italian group of patients subjected to CBCT without reporting any association with gender and mandible width, shape, and bone thickness.

Abstract Background We applied an artificial intelligence-based model to predict fragility fractures in postmenopausal women, using different dual-energy x-ray absorptiometry (DXA) parameters. Methods One hundred seventy-four postmenopausal women without vertebral fractures (VFs) at baseline (mean age 66.3 ± 9.8) were retrospectively evaluated. Data has been collected from September 2010 to August 2018. All subjects performed a spine x-ray to assess VFs, together with lumbar and femoral DXA for bone mineral density (BMD) and the bone strain index (BSI) evaluation. Follow-up exams were performed after 3.34 ± 1.91 years. Considering the occurrence of new VFs at follow-up, two groups were created: fractured versus not-fractured. We applied an artificial neural network (ANN) analysis with a predictive tool (TWIST system) to select relevant input data from a list of 13 variables including BMD and BSI. A semantic connectivity map was built to analyse the connections among variables within the groups. For group comparisons, an independent-samples t-test was used; variables were expressed as mean ± standard deviation. Results For each patient, we evaluated a total of n = 6 exams. At follow-up, n = 69 (39.6%) women developed a VF. ANNs reached a predictive accuracy of 79.56% within the training testing procedure, with a sensitivity of 80.93% and a specificity of 78.18%. The semantic connectivity map showed that a low BSI at the total femur is connected to the absence of VFs. Conclusion We found a high performance of ANN analysis in predicting the occurrence of VFs. Femoral BSI appears as a useful DXA index to identify patients at lower risk for lumbar VFs.

ObjectiveThe extent of response to neoadjuvant chemotherapy predicts survival in Ewing sarcoma. This study focuses on MRI radiomics of skeletal Ewing sarcoma and aims to investigate feature reproducibility and machine learning prediction of response to neoadjuvant chemotherapy.Materials and methodsThis retrospective study included thirty patients with biopsy-proven skeletal Ewing sarcoma, who were treated with neoadjuvant chemotherapy before surgery at two tertiary sarcoma centres. 7 patients were poor responders and 23 were good responders based on pathological assessment of the surgical specimen. On pre-treatment T1-weighted and T2-weighted MRI, 2D and 3D tumour segmentations were manually performed. Features were extracted from original and wavelet-transformed images. Feature reproducibility was assessed through small geometrical transformations of the regions of interest mimicking multiple manual delineations, and intraclass correlation coefficient >0.75 defined feature reproducibility. Feature selection also consisted of collinearity and significance analysis. After class balancing in the training cohort, three machine learning classifiers were trained and tested on unseen data using hold-out cross-validation.Results1303 (77%) 3D and 620 (65%) 2D radiomic features were reproducible. 4 3D and 4 2D features passed feature selection. Logistic regression built upon 3D features achieved the best performance with 85% accuracy (AUC=0.9) in predicting response to neoadjuvant chemotherapy.ConclusionCompared to 2D approach, 3D MRI radiomics of Ewing sarcoma had superior reproducibility and higher accuracy in predicting response to neoadjuvant chemotherapy, particularly when using logistic regression classifier.

We developed and used a deep learning tool to process biplanar radiographs of 9,832 non-surgical patients suffering from spinal deformities, with the aim of reporting the statistical distribution of radiological parameters describing the spinal shape and the correlations and interdependencies between them. An existing tool able to automatically perform a three-dimensional reconstruction of the thoracolumbar spine has been improved and used to analyze a large set of biplanar radiographs of the trunk. For all patients, the following parameters were calculated: spinopelvic parameters; lumbar lordosis; mismatch between pelvic incidence and lumbar lordosis; thoracic kyphosis; maximal coronal Cobb angle; sagittal vertical axis; T1-pelvic angle; maximal vertebral rotation in the transverse plane. The radiological parameters describing the sagittal alignment were found to be highly interrelated with each other, as well as dependent on age, while sex had relatively minor but statistically significant importance. Lumbar lordosis was associated with thoracic kyphosis, pelvic incidence and sagittal vertical axis. The pelvic incidence-lumbar lordosis mismatch was found to be dependent on the pelvic incidence and on age. Scoliosis had a distinct association with the sagittal alignment in adolescent and adult subjects. The deep learning-based tool allowed for the analysis of a large imaging database which would not be reasonably feasible if performed by human operators. The large set of results will be valuable to trigger new research questions in the field of spinal deformities, as well as to challenge the current knowledge.

To determine the short-term intra-operator precision and inter-operator repeatability of radiofrequency echographic multi-spectrometry (REMS) at the lumbar spine (LS) and proximal femur (FEM). All patients underwent an ultrasound scan of the LS and FEM. Both precision and repeatability, expressed as root-mean-square coefficient of variation (RMS-CV) and least significant change (LSC) were obtained using data from two consecutive REMS acquisitions by the same operator or two different operators, respectively. The precision was also assessed in the cohort stratified according to BMI classification. The mean (±SD) age of our subjects was 48.9 ± 6.8 for LS and 48.3 ± 6.1 for FEM. Precision was assessed on 42 subjects at LS and 37 subjects on FEM. Mean (±SD) BMI was 24.71 ± 4.2 for LS and 25.0 ± 4.84 for FEM. Respectively, the intra-operator precision error (RMS-CV) and LSC resulted in 0.47% and 1.29% at the spine and 0.32% and 0.89% at the proximal femur evaluation. The inter-operator variability investigated at the LS yielded an RMS-CV error of 0.55% and LSC of 1.52%, whereas for the FEM, the RMS-CV was 0.51% and the LSC was 1.40%. Similar values were found when subjects were divided into BMI subgroups. REMS technique provides a precise estimation of the US-BMD independent of subjects’ BMI differences.

Abstract Background Feature reproducibility and model validation are two main challenges of radiomics. This study aims to systematically review radiomic feature reproducibility and predictive model validation strategies in studies dealing with CT and MRI radiomics of bone and soft-tissue sarcomas. The ultimate goal is to promote achieving a consensus on these aspects in radiomic workflows and facilitate clinical transferability. Results Out of 278 identified papers, forty-nine papers published between 2008 and 2020 were included. They dealt with radiomics of bone (n = 12) or soft-tissue (n = 37) tumors. Eighteen (37%) studies included a feature reproducibility analysis. Inter-/intra-reader segmentation variability was the theme of reproducibility analysis in 16 (33%) investigations, outnumbering the analyses focused on image acquisition or post-processing (n = 2, 4%). The intraclass correlation coefficient was the most commonly used statistical method to assess reproducibility, which ranged from 0.6 and 0.9. At least one machine learning validation technique was used for model development in 25 (51%) papers, and K-fold cross-validation was the most commonly employed. A clinical validation of the model was reported in 19 (39%) papers. It was performed using a separate dataset from the primary institution (i.e., internal validation) in 14 (29%) studies and an independent dataset related to different scanners or from another institution (i.e., independent validation) in 5 (10%) studies. Conclusions The issues of radiomic feature reproducibility and model validation varied largely among the studies dealing with musculoskeletal sarcomas and should be addressed in future investigations to bring the field of radiomics from a preclinical research area to the clinical stage.

We aimed to demonstrate the applicability of Snyder’s arthroscopic classification of rotator cuff tears (RCT) in shoulder ultrasound (US) and to compare it with MR arthrography (MRA). Forty-six patients (34 males; mean age:34 ± 14 years) underwent shoulder US and MRA. Two radiologists (R1 = 25 years of experience; R2 = 2 years of experience) assigned A1–4, B1–4, or C1–4 values depending on the extent of RCT in both US and MRA. Inter-reader intra-modality and intra-reader inter-modality agreement were calculated using Cohen’s kappa coefficient. US sensitivity and specificity of both readers were calculated using MRA as the gold standard. Patients were divided into intact cuff vs. tears, mild (A1/B1) vs. moderate (A2–3/B2–3) tears, mild-moderate (A2/B2) vs. high-moderate (A3/B3) cuff tears, moderate (A2–3/B2–3) vs. advanced (A4/B4) and full-thickness (C) tears. The highest agreement values in inter-reader US evaluation were observed for mild-moderate vs. high-moderate RCT (K = 0.745), in inter-reader MRA evaluation for mild vs. moderate RCT (K = 0.821), in R1 inter-modality (US-MRA) for mild-moderate vs. high-moderate and moderate vs. advanced/full-thickness RCT (K = 1.000), in R2 inter-modality (US-MRA) for moderate vs. advanced/full-thickness RCT (K = 1.000). US sensitivity ranged from 88.89%(R1)–84.62%(R2) to 100% (both readers), while specificity from 77.78%(R1)–90.00%(R2) to 100% (both readers). Snyder’s classification can be used in US to ensure the correct detection and characterization of RCT.

Summary: Background: Atypical cartilaginous tumour (ACT) and grade II chondrosarcoma (CS2) of long bones are respectively managed with watchful waiting or curettage and wide resection. Preoperatively, imaging diagnosis can be challenging due to interobserver variability and biopsy suffers from sample errors. The aim of this study is to determine diagnostic performance of MRI radiomics-based machine learning in differentiating ACT from CS2 of long bones. Methods: One-hundred-fifty-eight patients with surgically treated and histology-proven cartilaginous bone tumours were retrospectively included at two tertiary bone tumour centres. The training cohort consisted of 93 MRI scans from centre 1 (n=74 ACT; n=19 CS2). The external test cohort consisted of 65 MRI scans from centre 2 (n=45 ACT; n=20 CS2). Bidimensional segmentation was performed on T1-weighted MRI. Radiomic features were extracted. After dimensionality reduction and class balancing in centre 1, a machine-learning classifier (Extra Trees Classifier) was tuned on the training cohort using 10-fold cross-validation and tested on the external test cohort. In centre 2, its performance was compared with an experienced musculoskeletal oncology radiologist using McNemar's test. Findings: After tuning on the training cohort (AUC=0.88), the machine-learning classifier had 92% accuracy (60/65, AUC=0.94) in identifying the lesions in the external test cohort. Its accuracies in correctly classifying ACT and CS2 were 98% (44/45) and 80% (16/20), respectively. The radiologist had 98% accuracy (64/65) with no difference compared to the classifier (p=0.134). Interpretation: Machine learning showed high accuracy in classifying ACT and CS2 of long bones based on MRI radiomic features. Funding: ESSR Young Researchers Grant.

Abstract Background We focus on the importance of interpreting the quality of the labeling used as the input of predictive models to understand the reliability of their output in support of human decision-making, especially in critical domains, such as medicine. Methods Accordingly, we propose a framework distinguishing the reference labeling (or Gold Standard) from the set of annotations from which it is usually derived (the Diamond Standard). We define a set of quality dimensions and related metrics: representativeness (are the available data representative of its reference population?); reliability (do the raters agree with each other in their ratings?); and accuracy (are the raters’ annotations a true representation?). The metrics for these dimensions are, respectively, the degree of correspondence, Ψ, the degree of weighted concordance ϱ, and the degree of fineness, Φ. We apply and evaluate these metrics in a diagnostic user study involving 13 radiologists. Results We evaluate Ψ against hypothesis-testing techniques, highlighting that our metrics can better evaluate distribution similarity in high-dimensional spaces. We discuss how Ψ could be used to assess the reliability of new predictions or for train-test selection. We report the value of ϱ for our case study and compare it with traditional reliability metrics, highlighting both their theoretical properties and the reasons that they differ. Then, we report the degree of fineness as an estimate of the accuracy of the collected annotations and discuss the relationship between this latter degree and the degree of weighted concordance, which we find to be moderately but significantly correlated. Finally, we discuss the implications of the proposed dimensions and metrics with respect to the context of Explainable Artificial Intelligence (XAI). Conclusion We propose different dimensions and related metrics to assess the quality of the datasets used to build predictive models and Medical Artificial Intelligence (MAI). We argue that the proposed metrics are feasible for application in real-world settings for the continuous development of trustable and interpretable MAI systems.

Abstract Background The aim is to describe a new arthroscopically assisted Latarjet technique. Methods We evaluated the clinical and radiological findings of 60 patients with chronic recurrent anterior gleno-humeral instability who underwent, between September 2013 and November 2014, an arthroscopically-assisted Latarjet procedure with double round endobutton fixation. Inclusion criteria were: chronic anterior recurrent instability, Instability Severity Index Score (ISIS) greater than three points, a glenoid bone defect > 15% or a Hill Sachs lesion with concomitant glenoid bone defect > 10%. During surgery the joint capsule and the anterior glenoid labrum were detached. Two drill tunnels perpendicular to the neck of the glenoid were made through a guide. An accessible pilot hole through the glenoid was created to allows the passage of guidewires for coracoid guidance and final fixation onto the anterior glenoid. Through a restricted deltopectoral access a coracoid osteotomy was made. Finally, the graft was prepared, inserted and secured using half-stitches. Results The mean follow-up was 32.5 months (range 24–32 months). At a mean follow-up, 56 of the 60 subjects claimed a stable shoulder without postoperative complaints, two (3.3%) had an anterior dislocation after new traumatic injury, and two (3.3%) complained of subjective instability. At the latest follow-up, four subjects complained of painful recurrent anterior instability during abduction-external rotation with apprehension. At 1 year, the graft had migrated in one patient (1.7%) and judged not healed and high positioned in another patient (1.7%). Moreover, a glenoid bony gain of 26.3% was recorded. At the latest follow-up, three patients had grade 1 according to Samilson and Prieto classification asymptomatic degenerative changes. Nerve injuries and infections were not detected. None of the 60 patients underwent revision surgery. Healing rate of the graft was 96.7%. Conclusions This technique of arthroscopically assisted Latarjet combines mini-open and arthroscopic approach for improving the precision of the bony tunnels in the glenoid and coracoid placement, minimizing any potential risk of neurologic complications. It can be an option in subjects with anterior gleno-humeral instability and glenoid bone defect. Further studies should be performed to confirm our preliminary results. Trial registration Trial registration number 61/int/2017 Name of registry: ORS Date of registration 11.5.2017 Date of enrolment of the first participant to the trial: September 2013 ‘retrospectively registered’ Level of evidence IV

Abstract Background Magnetic resonance arthrography (MRA) requires intra-articular injection of gadolinium-based diluted paramagnetic contrast material. To our knowledge, gadobenate dimeglumine (Gd-BOPTA) has never been used for intra-articular applications. Our aim was to test in vitro different concentrations of Gd-BOPTA to be potentially used to perform MRA. Methods Gd-BOPTA was diluted in saline (NaCl 0.9%) to achieve different concentrations (4 mmol/l; 2 mmol/l; 1 mmol/l; 0.67 mmol/l; 0.5 mmol/l). Six sets of five sterile pipes were prepared with 5 ml of each solution, five sets added with 0.5 ml of fresh synovial fluid. Two separate pipes were prepared with 5 ml of gadopentetate dimeglumine (Gd-DTPA) at 2 mmol/l, one pipe added with 0.5 ml of synovial fluid. Pipes were imaged using a T1-weighted sequence at 1.5 T. For each pipe, signal intensity (SI) in arbitrary units (au) was measured. Results SI reproducibility range was 86–99%. Mean Gd-BOPTA SI in pipes containing synovial fluid increased from 1236 ± 8au (0.5 mmol/l) up to 1610 ± 44au (1 mmol/l) and down to 1405 ± 33au (4 mmol/l). Mean Gd-BOPTA SI in pipes without synovial fluid increased from 1184 ± 29au (0.5 mmol/l) up to 1530 ± 38au (1 mmol/l), and down to 1347 ± 39au (4 mmol/l). SI of pipes without synovial fluid was lower than that of pipes with synovial fluid for both Gd-BOPTA and Gd-DTPA (P ≤ 0.002). Regarding pipes with synovial fluid, mean Gd-DTPA SI at 2 mmol/l was 1246 ± 27au. Compared with Gd-BOPTA, SI was not different at 0.5 mmol/l (− 0.2%, P = 0.587) while it was higher (P

Italy was the first European country to face the SARS-CoV-2 virus (COVID-19) pandemic in 2020. The country quickly implemented strategies to contain contagions and re-organize medical resources. We evaluated the COVID-19 effects on the activity of a tertiary-level orthopedic emergency department (ED) during the first and second pandemic waves. We retrospectively collected and compared clinical radiological data of ED admissions during four periods: period A, first pandemic wave; period B, second pandemic wave; period C, three months before the COVID-19 outbreak; period D, same timeframe of the first wave but in 2019. During period A, we found a reduction in ED admissions (−68.2% and −59.9% compared with periods D and C) and a decrease in white codes (non-urgent) (−7.5%) compared with pre-pandemic periods, with a slight increase for all other codes: +6.3% green (urgent, not critical), +0.8% yellow (moderately critical) and +0.3% red (highly urgent, risk of death). We observed an increased rate of fracture diagnosis in period A: +14.9% and +13.3% compared with periods D and C. Our study shows that the COVID-19 pandemic caused a drastic change in the ED patient flow and clinical radiological activity, with a marked reduction in admissions and an increased rate of more severe triage codes and diagnosed fractures.

We evaluated the correlation of electrical impedance myography (EIM) measurements of knee muscles composition using Skulpt ChiselTM with MRI data retrieved from muscles segmentation. A total of 140 patients (71 females, 52 ± 21 years) underwent knee MRI, EIM with Skulpt®, and clinical evaluation (SARC-F questionnaire). MRIs were reviewed to assess the cross-sectional area (CSA) and skeletal muscle index (SMI = CSA/height2) of vastus medialis, vastus lateralis, biceps, semimembranosus, and sartorius. We tested the correlations of EIM-derived parameters [body fat-percentage (BF%) and muscle quality] with total CSA, CSA of each muscle, SMI, and SARC-F scores (0–10) using Pearson correlation coefficient. We found medium negative correlation of BF% with SMI (r = −0.430, p < 0.001) and total CSA (r = −0.445, p < 0.001), particularly with biceps (r = −0.479, p < 0.001), sartorius (r = −0.440, p < 0.001), and semimembranosus (r = −0.357, p < 0.001). EIM-derived muscle quality showed small-to-medium positive correlation with MRI measurements, ranging from r = 0.234 of biceps (p = 0.006) to r = 0.302 of total CSA (p < 0.001), except for vastus lateralis (r = 0.014, p = 0.873). SARC-F scores showed small correlations with EIM and MRI data, ranging from r = −0.132 (p = 0.121) with EIM muscle quality to r = −0.288 (p = 0.001) with CSA of vastus medialis. Hence, we observed small-to-medium correlations of muscle parameters derived from Skulpt ChiselTM with SARC-F scores and MRI parameters. We recommend using Skulpt ChiselTM with caution for assessing knee skeletal muscles composition.

We aimed to evaluate the diagnostic performance of shoulder MR arthrography (MRA) acquired in the neutral (N), internal rotation (IR), and external rotation (ER) positions of the shoulder to detect SLAP lesions. Three observers evaluated 130 MRAs to detect SLAP lesions and to calculate labral diastasis in this triple-blinded study. Sensitivity was much higher in the ER (92.5–97.5%) than in the N (60–72.5%) and IR (42.5–52.5%) positions, and the specificity of all the reviewers was 100% in all the positions. The diagnostic accuracy was higher in the ER too (97.7–99.2%). The diastasis length was significantly higher in the ER (median = 2.5–2.8 mm) than in the N (1 mm) and IR (0 mm) positions and was also significantly higher in those patients requiring surgery (p = 0.001). The highest inter-rater agreement values were observed in the ER both in SLAP detection (k = 0.982) and the diastasis length evaluation (ICC = 0.962). The diastasis length threshold in the ER that best separated the patients who did and did not require surgery was 3.1 mm (AUC = 0.833). In 14.6% of the cases, ER enabled the detection of SLAP lesions not identified in the N position. MRA with the ER improves the diagnosis of SLAP lesions and, together with the IR position, provides additional dynamic information about the diastasis of the lesions. It is recommended to perform additional ER and IR scans in the shoulder MRA protocol.

Background: Clinical management ranges from surveillance or curettage to wide resection for atypical to higher-grade cartilaginous tumours, respectively. Our aim was to investigate the performance of computed tomography (CT) radiomics-based machine learning for classification of atypical cartilaginous tumours and higher-grade chondrosarcomas of long bones. Methods: One-hundred-twenty patients with histology-proven lesions were retrospectively included. The training cohort consisted of 84 CT scans from centre 1 (n=55 G1 or atypical cartilaginous tumours; n=29 G2-G4 chondrosarcomas). The external test cohort consisted of the CT component of 36 positron emission tomography-CT scans from centre 2 (n=16 G1 or atypical cartilaginous tumours; n=20 G2-G4 chondrosarcomas). Bidimensional segmentation was performed on preoperative CT. Radiomic features were extracted. After dimensionality reduction and class balancing in centre 1, the performance of a machine-learning classifier (LogitBoost) was assessed on the training cohort using 10-fold cross-validation and on the external test cohort. In centre 2, its performance was compared with preoperative biopsy and an experienced radiologist using McNemar's test. Findings: The classifier had 81% (AUC=0.89) and 75% (AUC=0.78) accuracy in identifying the lesions in the training and external test cohorts, respectively. Specifically, its accuracy in classifying atypical cartilaginous tumours and higher-grade chondrosarcomas was 84% and 78% in the training cohort, and 81% and 70% in the external test cohort, respectively. Preoperative biopsy had 64% (AUC=0.66) accuracy (p=0.29). The radiologist had 81% accuracy (p=0.75). Interpretation: Machine learning showed good accuracy in classifying atypical and higher-grade cartilaginous tumours of long bones based on preoperative CT radiomic features. Funding: ESSR Young Researchers Grant.

Abstract Background Osteoid osteoma is a benign bone-forming tumour, which very unfrequently has multifocal or multicentric presentation. We report the first known case of a multicentric, multifocal and recurrent osteoid osteoma treated using radiofrequency ablation. Case presentation A 39-year-old man with two-year history of left hip pain was admitted at our Institution. The pain was more intense during the night and partially relieved by salicylates. Pelvis CT demonstrated two lytic lesions (8 and 7 mm, respectively) with surrounding sclerotic reactive bone, both with a central focal area of high attenuation, located in the femoral neck and along the anterior portion of the acetabulum, respectively. Both lesions had clinical and imaging findings consistent with multicentric osteoid osteoma. Thus, the two lesions were biopsied – with pathologic confirmation of osteoid osteoma – and treated using radiofrequency ablation. Hip pain decreased but did not disappear, actually increasing a few months after treatment. CT and MRI were performed showing a smaller lesion (5 mm) with the same imaging features, surrounded by marrow oedema, along the posterior column of the acetabulum. The lesion was considered suspicious for osteoid osteoma, overlooked on previous examinations. Therefore, a diagnosis of multicentric and multifocal osteoid osteoma was established. The new lesion was again treated with radiofrequency ablation with symptom disappearance. However, hip pain relapsed after 18 months, and CT and MRI showed an osteoid osteoma recurrence on the posterior column of the acetabulum, which was biopsied and successfully treated using radiofrequency ablation. Conclusions To our knowledge, this is the first reported case of multicentric, multifocal, recurrent osteoid osteoma. Our case report highlights the importance of considering a diagnosis of multifocal osteoid osteoma when dealing with multifocal lytic lesions of the bone and with pain persistence after treatment. It also emphasises the combined role of CT and MRI in this setting.

Abstract The widespread use of imaging examinations has increased the detection of incidental adrenal lesions, which are mostly benign and non-functioning adenomas. The differentiation of a benign from a malignant adrenal mass can be crucial especially in oncology patients since it would greatly affect treatment and prognosis. In this setting, imaging plays a key role in the detection and characterization of adrenal lesions, with several imaging tools which can be employed by radiologists. A thorough knowledge of the imaging features of adrenal masses is essential to better characterize these lesions, avoiding a misinterpretation of imaging findings, which frequently overlap between benign and malignant conditions, thus helping clinicians and surgeons in the management of patients. The purpose of this paper is to provide an overview of the main imaging features of adrenal masses and tumor-like conditions recalling the strengths and weaknesses of imaging modalities commonly used in adrenal imaging.

Minimally invasive image-guided thermal ablation is becoming increasingly common as an alternative to surgery for the treatment of benign thyroid nodules. Among the various techniques for thermal ablation, laser ablation (LA) is the least invasive, using the smallest applicators available on the market and enabling extremely precise energy deposition. However, in some cases, multiple laser fibers must be used simultaneously for the treatment of large nodules. In this review, the LA technique is described, and its main clinical applications and results are discussed and illustrated.

Benign thyroid nodules are a common clinical occurrence and usually do not require treatment unless symptomatic. During the last years, ultrasound-guided minimally invasive treatments (MIT) gained an increasing role in the management of nodules causing local symptoms. In February 2018, the Italian MIT Thyroid Group was founded to create a permanent cooperation between Italian and international physicians dedicated to clinical research and assistance on MIT for thyroid nodules. The group drafted this list of statements based on literature review and consensus opinion of interdisciplinary experts to facilitate the diffusion and the appropriate use of MIT of thyroid nodules in clinical practice. (#1) Predominantly cystic/cystic symptomatic nodules should first undergo US-guided aspiration; ethanol injection should be performed if relapsing (level of evidence [LoE]: ethanol is superior to simple aspiration = 2); (#2) In symptomatic cystic nodules, thermal ablation is an option when symptoms persist after ethanol ablation (LoE = 4); (#3) Double cytological benignity confirmation is needed before thermal ablation (LoE = 2); (#4) Single cytological sample is adequate in ultrasound low risk (EU-TIRADS ≤3) and in autonomously functioning nodules (LoE = 2); (#5) Thermal ablation may be proposed as first-line treatment for solid, symptomatic, nonfunctioning, benign nodules (LoE = 2); (#6) Thermal ablation may be used for dominant lesions in nonfunctioning multinodular goiter in patients refusing/not eligible for surgery (LoE = 5); (#7) Clinical and ultrasound follow-up is appropriate after thermal ablation (LoE = 2); (#8) Nodule re-treatment can be considered when symptoms relapse or partially resolve (LoE = 2); (#9) In case of nodule regrowth, a new cytological assessment is suggested before second ablation (LoE = 5); (#10) Thermal ablation is an option for autonomously functioning nodules in patients refusing/not eligible for radioiodine or surgery (LoE = 2); (#11) Small autonomously functioning nodules can be treated with thermal ablation when thyroid tissue sparing is a priority and ≥80% nodule volume ablation is expected (LoE = 3).

Abstract Background Deep learning is a ground-breaking technology that is revolutionising many research and industrial fields. Generative models are recently gaining interest. Here, we investigate their potential, namely conditional generative adversarial networks, in the field of magnetic resonance imaging (MRI) of the spine, by performing clinically relevant benchmark cases. Methods First, the enhancement of the resolution of T2-weighted (T2W) images (super-resolution) was tested. Then, automated image-to-image translation was tested in the following tasks: (1) from T1-weighted to T2W images of the lumbar spine and (2) vice versa; (3) from T2W to short time inversion-recovery (STIR) images; (4) from T2W to turbo inversion recovery magnitude (TIRM) images; (5) from sagittal standing x-ray projections to T2W images. Clinical and quantitative assessments of the outputs by means of image quality metrics were performed. The training of the models was performed on MRI and x-ray images from 989 patients. Results The performance of the models was generally positive and promising, but with several limitations. The number of disc protrusions or herniations showed good concordance (κ = 0.691) between native and super-resolution images. Moderate-to-excellent concordance was found when translating T2W to STIR and TIRM images (κ ≥ 0.842 regarding disc degeneration), while the agreement was poor when translating x-ray to T2W images. Conclusions Conditional generative adversarial networks are able to generate perceptually convincing synthetic images of the spine in super-resolution and image-to-image translation tasks. Taking into account the limitations of the study, deep learning-based generative methods showed the potential to be an upcoming innovation in musculoskeletal radiology.

Abstract Background Image-guided elastoplasty is an innovative method for percutaneous vertebral augmentation with a silicone elastomeric material. Our aim was to evaluate its technical success, safety and efficacy as well as the rate of secondary fractures. Methods Nineteen patients (13 women and 6 men, age 72 ± 10 years, mean ± standard deviation) underwent elastoplasty between 2010 and 2016. A total of 33 vertebrae were treated. A total of 2–6 mL of silicone-based elastomeric polymer material (VK100) was used. Visual analogue scale (VAS) and Oswestry disability index (ODI) pain scores were used. Results In all cases, it was possible to complete the procedure (technical success 100%). No major complications occurred. In 6/19 (31.5%) patients, asymptomatic leakage of the material was observed during the procedure. Full pain recovery was obtained in 18/19 (94%) patients. One patient with a painful angioma did not experience any change in symptoms. VAS and ODI were significantly reduced after the procedure, from 7.9 ± 1.1 to 0.7 ± 1.4 and from 79.6 ± 12% to 9.9 ± 14% respectively (p

BackgroundOsteoporosis is an asymptomatic disease of high prevalence and incidence, leading to bone fractures burdened by high mortality and disability, mainly when several subsequent fractures occur. A fragility fracture predictive model, Artificial Intelligence-based, to identify dual X-ray absorptiometry (DXA) variables able to characterise those patients who are prone to further fractures called Bone Strain Index, was evaluated in this study.MethodsIn a prospective, longitudinal, multicentric study 172 female outpatients with at least one vertebral fracture at the first observation were enrolled. They performed a spine X-ray to calculate spine deformity index (SDI) and a lumbar and femoral DXA scan to assess bone mineral density (BMD) and bone strain index (BSI) at baseline and after a follow-up period of 3 years in average. At the end of the follow-up, 93 women developed a further vertebral fracture. The further vertebral fracture was considered as one unit increase of SDI. We assessed the predictive capacity of supervised Artificial Neural Networks (ANNs) to distinguish women who developed a further fracture from those without it, and to detect those variables providing the maximal amount of relevant information to discriminate the two groups. ANNs choose appropriate input data automatically (TWIST-system, Training With Input Selection and Testing). Moreover, we built a semantic connectivity map usingthe Auto Contractive Map to provide further insights about the convoluted connections between the osteoporotic variables under consideration and the two scenarios (further fracture vs no further fracture).ResultsTWIST system selected 5 out of 13 available variables: age, menopause age, BMI, FTot BMC, FTot BSI. With training testing procedure, ANNs reached predictive accuracy of 79.36%, with a sensitivity of 75% and a specificity of 83.72%. The semantic connectivity map highlighted the role of BSI in predicting the risk of a further fracture.ConclusionsArtificial Intelligence is a useful method to analyse a complex system like that regarding osteoporosis, able to identify patients prone to a further fragility fracture. BSI appears to be a useful DXA index in identifying those patients who are at risk of further vertebral fractures.

Bone and soft-tissue primary malignant tumors or sarcomas are a large, diverse group of mesenchymal-derived malignancies. They represent a model for intra- and intertumoral heterogeneities, making them particularly suitable for radiomics analyses. Radiomic features offer information on cancer phenotype as well as the tumor microenvironment which, combined with other pertinent data such as genomics and proteomics and correlated with outcomes data, can produce accurate, robust, evidence-based, clinical-decision support systems. Our purpose in this narrative review is to offer an overview of radiomics studies dealing with Magnetic Resonance Imaging (MRI)-based radiomics models of bone and soft-tissue sarcomas that could help distinguish different histotypes, low-grade from high-grade sarcomas, predict response to multimodality therapy, and thus better tailor patients’ treatments and finally improve their survivals. Although showing promising results, interobserver segmentation variability, feature reproducibility, and model validation are three main challenges of radiomics that need to be addressed in order to translate radiomics studies to clinical applications. These efforts, together with a better knowledge and application of the “Radiomics Quality Score” and Image Biomarker Standardization Initiative reporting guidelines, could improve the quality of sarcoma radiomics studies and facilitate radiomics towards clinical translation.

Teriparatide is a bone-forming therapy for osteoporosis that increases bone quantity and texture, with uncertain action on bone geometry. No data are available regarding its influence on bone strain. To investigate teriparatide action on parameters of bone quantity and quality and on Bone Strain Index (BSI), also derived from DXA lumbar scan, based on the mathematical model finite element method. Forty osteoporotic patients with fractures were studied before and after two years of daily subcutaneous 20 mcg of teriparatide with dual X-ray photon absorptiometry to assess bone mineral density (BMD), hip structural analysis (HSA), trabecular bone score (TBS), BSI. Spine deformity index (SDI) was calculated from spine X-ray. Shapiro-Wilks, Wilcoxon and Student's t test were used for classical statistical analysis. Auto Contractive Map was used for Artificial Neural Network Analysis (ANNs). In the entire population, the ameliorations after therapy regarded BSI (-13.9%), TBS (5.08%), BMD (8.36%). HSA parameters of femoral shaft showed a worsening. Dividing patients into responders (BMD increase >10%) and non-responders, the first presented TBS and BSI ameliorations (11.87% and -25.46%, respectively). Non-responders presented an amelioration of BSI only, but less than in the other subgroup (-6.57%). ANNs maps reflect the mentioned bone quality improvements. Teriparatide appears to ameliorate not only BMD and TBS, but also BSI, suggesting an increase of bone strength that may explain the known reduction in fracture risk, not simply justified by BMD increase. BSI appears to be a sensitive index of TPD effect. ANNs appears to be a valid tool to investigate complex clinical systems.

The study aimed to evaluate masseter muscle stiffness in adult healthy volunteers referred to a massage treatment and also to investigate whether shear-wave elastography can be used to monitor the effect of massage on the masseter muscle. The study included 21 healthy volunteers, who were subjected to a 30-minute massage of the masseter muscle. Muscle stiffness was measured by shear-wave elastography before and directly after the massage. Pain during the massage was assessed using the visual analogue scale (VAS). The data of 20 patients (one excluded due to severe pain) with a median age of 34.5 years were analysed. The stiffness values were 11.46 ± 1.55 kPa before and 8.97 ± 0.96 kPa after the massage (p

Background: The purpose of this study was to evaluate plantar fasciae of healthy subjects and patients with plantar fasciitis by shear wave velocity (SWV) and stiffness with correlation to B-Mode and color Doppler ultrasonography (CDUS) and to establish cut-off values. Methods: This observational study was conducted with the approval of the Institutional Review Boards (IRBs) of our institution. 108 unilateral plantar fasciae were evaluated by including 87 consecutive patients (mean age 51.7; range: 29–82) (66 women and 21 men) with plantar fasciitis and 21 asymptomatic age matched healthy volunteers (mean age 47.3; range: 32–58) (15 women and 6 men). All patients were prospectively imaged between July 2018 and March 2019. B-mode US was used to measure thickness and CDUS to grade vascularity. SWE measurements were repeated 3 times and mean was used for statistical analysis. Results: Mean SWV value in healthy subjects was 6.94 m/s and in patients 4.98 m/s with a mean stiffness value of 152.88 kPa and 93.54 kPa respectively (p < 0.001). For SWV a cut-off value of 6.16 m/s had a specificity of 80.95% and sensitivity of 79.31%. For stiffness a cut-off value of 125.57 kPa had a specificity of 80.95% and sensitivity of 80.46%. No correlation to CDUS was found. The mean thickness of healthy fascias was 3.3 mm (range 2.4–3.9) compared to 6.1 mm (range 2.0–22.0) in plantar fasciitis (p < 0.001) with no correlation to SWV or to stiffness (r² = 0.02, p = 0.06). Conclusion: SWE allows quantitative assessment of plantar fascia stiffness, which decreases in patients with plantar fasciitis. No correlation to the thickness of the plantar fascia was found, therefore it represents an independent factor for the diagnosis of plantar fasciitis and could be helpful in addition to thickness measurement in unclear cases.

Abstract Background The purpose of our study was to assess the clinical and imaging outcome of autologous matrix-induced chondrogenesis (AMIC) technique consisting of microfractures followed by the filling of osteochondral lesions of the talus (OLTs) with a cell-free biphasic collagen-hydroxyapatite osteochondral scaffold (MaioRegen). Methods Sixteen patients (eight males, age: 42.6 ± 18.4, range 14–74) with OLT repaired using AMIC technique, with implantation of MaioRegen, were clinically evaluated through the American Orthopedic Foot and Ankle Society Score (AOFAS) and a 10-point Visual Analogue Scale (VAS) pain score after a mean follow-up of 30 ± 16.9 months. The MRI examinations were performed 12 and 24 months after surgery. A paired t-test was applied to compare pre- and post-operative clinical findings (VAS and AOFAS) and Magnetic resonance observation of cartilage repair tissue (MOCART) score changes in the follow-up. To assess the correlation between variation of AOFAS and MOCART scores, the Pearson’s correlation coefficient was calculated. Results No complications after surgery were encountered. From pre-operative to post-operative values, there was a significant (P

Abstract Objectives Our aim was to evaluate the quality of published guidelines on musculoskeletal ultrasound (MSK-US) for adults. Methods Between June and July 2016, we conducted an online search for MSK-US guidelines, which were evaluated by four independent readers blinded to each other using the AGREE II tool. A fifth independent reviewer calculated scores per each domain and agreement between reviewers’ scores using the intraclass correlation coefficient (ICC). Results Five guidelines were included in this appraisal. They were published between 2001 and 2014. Our appraisal showed intermediate results, with four out of five guidelines scoring “average” as overall quality. Domain 1 (scope and purpose) achieved the highest result (total score = 71.1% ± 18.7%). Domain 6 (editorial independence) had the lowest score (total score = 26.3% ± 19.3%). Interobserver agreement was very good for all the evaluated guidelines (ICC ranged between 0.932 and 0.956). Conclusions Overall, quality of MSK-US guidelines ranges from low to average when evaluated using the AGREE II tool. The editorial independence domain was the most critical, thus deserving more attention when developing future guidelines. Main messages • Four of five guidelines on MSK-US had an average quality level. • Scope/purpose had the highest score (71.1% ± 18.7%). • Editorial independence had the lowest score (26.3% ± 19.3%). • Interobserver agreement was very good (ranges: 0.932–0.956).

Abstract Objectives Dual energy X-ray absorptiometry (DXA) is the most widely used technique to measure bone mineral density (BMD). Appropriate and accurate use of DXA is of great importance, and several guidelines have been developed in the last years. Our aim was to evaluate the quality of published guidelines on DXA for adults. Methods Between June and July 2016 we conducted an online search for DXA guidelines, which were evaluated by four independent readers blinded to each other using the AGREE II instrument. A fifth independent reviewer calculated scores per each domain and agreement between reviewers’ scores. Results Four out of 59 guidelines met inclusion criteria and were included. They were published between 2005 and 2014. Three out of four guidelines reached a high level of quality, having at least five domain scores higher than 60%. Domain 1 (Scope and Purpose) achieved the highest result (total score = 86.8 ± 3.7%). Domain 6 (Editorial Independence) had the lowest score (total score = 54.7 ± 12.5%). Interobserver agreement ranged from fair (0.230) to good (0.702). Conclusions Overall, the quality of DXA guidelines is satisfactory when evaluated using the AGREE II instrument. The Editorial Independence domain was the most critical, thus deserving more attention when developing future guidelines. Main messages • Three of four guidelines on DXA had a high quality level (>60%). • Scope/purpose had the highest score (86.8 ± 3.7%). • Editorial Independence had the lowest score (54.7 ± 12.5%). • Interobserver agreement ranged from fair (0.230) to good (0.702).

Background: Radiation doses and capability of EOS, conventional radiography (CR), and computed tomography (CT) to detect and measure enchondromas in a dedicated five-year-old anthropomorphic phantom were compared. Methods: To simulate enchondromas, minced pieces of chicken bone and cartilage were packed in conventional kitchen plastic foil to create ovoidal/rounded masses and randomly hung on the phantom. The phantom was imaged five times with CR, CT, and EOS, each time changing the number and position of inserts. All images were reviewed by a senior radiologist and a radiology resident. Results: EOS and CR detected all inserts in 4/5 cases (80%), while in one case 1/17 inserts was not seen. Excellent agreement of EOS with CR (88% reproducibility; bias = 14 mm; repeatability coefficient (CoR) 2.9; 95% CI from −2.8 to 3.1 mm; p = 0.5) and CT (81% reproducibility; bias = 15 mm; CoR 5.2; 95% CI from −5.5 to 5.2 mm; p = 0.7) was found. EOS showed 71% interobserver reproducibility (CoR 7.2; bias = 0.6 mm; 95% CI from −6.6 to 7.8 mm; p = 0.25). The EOS-Fast radiation dose was also significantly lower than the median radiation dose of CR (644.7 (599.4–651.97) mGy•cm2, p = 0.004). Conclusions: Low-dose EOS has the same capability as CR to detect and measure enchondroma-like inserts on a phantom and may be considered to monitor patients with multiple enchondromas.

Diagnosing a peri-prosthetic joint infection (PJI) remains challenging despite the availability of a variety of clinical signs, serum and synovial markers, imaging techniques, microbiological and histological findings. Moreover, the one and only true definition of PJI does not exist, which is reflected by the existence of at least six different definitions by independent societies. These definitions are composed of major and minor criteria for defining a PJI, but most of them do not include imaging techniques. This paper highlights the pros and cons of available imaging techniques—X-ray, ultrasound, computed tomography (CT), Magnetic Resonance Imaging (MRI), bone scintigraphy, white blood cell scintigraphy (WBC), anti-granulocyte scintigraphy, and fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT), discusses the added value of hybrid camera systems—single photon emission tomography/computed tomography (SPECT/CT), PET/CT and PET/MRI and reports consensus answers on important clinical questions that were discussed during the Third European Congress on Inflammation/Infection Imaging in Rome, December 2019.

Background: We prospectively tested technical feasibility and clinical outcome of percutaneous ultrasound-guided tenotomy of long head of biceps tendon (LHBT). Methods: We included 11 patients (6 women; age: 73 ± 8.6 years) with symptomatic full-thickness rotator cuff tear and intact LHBT, in whom surgical repair was not possible/refused. After ultrasound-guided injection of local anesthetic, the LHBT was cut with a scalpel under continuous ultrasound monitoring until it became no longer visible. Pain was recorded before and at least six months after procedure. An eight-item questionnaire was administered to patients at follow-up. Results: A median of 4 tendon cuts were needed to ensure complete tenotomy. Mean procedure duration was 65 ± 5.7 s. Mean length of skin incision was 5.8 ± 0.6 mm. Pre-tenotomy VAS score was 8.2 ± 0.7, post-tenotomy VAS was 2.8 ± 0.6 (p < 0.001). At follow-up, 5/11 patients were very satisfied, 5/11 satisfied and 1/11 neutral. One patient experienced cramping and very minimal pain in the biceps. Six patients had still moderate shoulder pain, 1/11 minimal pain, 2/11 very minimal pain, while 2/11 had no pain. No patients had weakness in elbow flexion nor limits of daily activities due to LHBT. One patient showed Popeye deformity. All patients would undergo ultrasound-guided tenotomy again. Conclusion: ultrasound-guided percutaneous LHBT tenotomy is technically feasible and effective.