Demographics and additional haematologic cancers of patients with histiocytic/dendritic cell neoplasms.

AIMS: The discovery of somatic genetic alterations established many histiocytic disorders as haematologic neoplasms. We aimed to investigate the demographic characteristics and additional haematologic cancers of patients diagnosed with histiocytic disorders in The Netherlands. METHODS AND RESULTS: We retrieved data on histiocytosis patients from the Dutch Nationwide Pathology Databank (Palga). During 1993 to 2022, more than 4000 patients with a pathologist-assigned diagnosis of a histiocytic disorder were registered in Palga. Xanthogranulomas were the most common subtype, challenging the prevailing assumption that Langerhans cell histiocytosis (LCH) is the most common histiocytic disorder. LCH and juvenile xanthogranuloma (JXG) had a peak incidence in the first years of life; males were overrepresented among all histiocytosis subgroups. 118 patients had a histiocytic disorder and an additional haematologic malignancy, including 107 (91%) adults at the time of histiocytosis diagnosis. In 16/118 patients, both entities had been analysed for the same genetic alteration(s). In 11 of these 16 patients, identical genetic alterations had been detected in both haematologic neoplasms. This included two patients with PAX5 p.P80R mutated B cell acute lymphoblastic leukaemia and secondary histiocytic sarcoma, further supporting that PAX5 alterations may predispose (precursor) B cells to differentiate into the myeloid lineage. All 4/11 patients with myeloid neoplasms as their additional haematologic malignancy had shared N/KRAS mutations. CONCLUSIONS: This population-based study highlights the frequency of xanthogranulomas. Furthermore, our data add to the growing evidence supporting clonal relationships between histiocytic/dendritic cell neoplasms and additional myeloid or lymphoid malignancies. Particularly adult histiocytosis patients should be carefully evaluated for the development of these associated haematologic cancers.

Robust IMPT and follow-up toxicity in skull base chordoma and chondrosarcoma-a single-institution clinical experience.

BACKGROUND: Chordomas and chondrosarcomas of the skull base are rare, slowly growing malignant bone neoplasms. Despite their radioresistant properties, proton therapy has been successfully used as an adjunct to resection or as a definitive treatment. Herewith, we present our experience with robustly optimized intensity-modulated proton therapy (IMPT) and related toxicities in skull base chordoma and chondrosarcoma patients treated at HollandPTC, Delft, the Netherlands. METHODS: Clinical data, treatment plans, and acute toxicities of patients treated between July 2019 and August 2021 were reviewed. CT and 3.0T MRI scans for treatment planning were performed in supine position in a thermoplastic mold. In total, 21 dose optimization and 28 dose evaluation scenarios were simulated. Acute toxicity was scored weekly before and during the treatment according to the CTCAE v4.0. Median follow-up was 35 months (range 12-36 months). RESULTS: Overall, 9 chordoma and 3 chondrosarcoma patients with 1-3 resections prior to IMPT were included; 4 patients had titanium implants. Brainstem core and surface and spinal cord core and surface were used for nominal plan robust optimization in 11, 10, 8, and 7 patients, respectively. Middle ear inflammation, dry mouth, radiation dermatitis, taste disorder, and/or alopecia of grades 1-3 were noted at the end of treatment among 6 patients without similar complaints at inclusion; symptoms disappeared 3 months following the treatment. CONCLUSION: Robustly optimized IMPT is clinically feasible as a postoperative treatment for skull base chordoma and chondrosarcoma patients. We observed acceptable early toxicities (grade 1-3) that disappeared within the first 3 months after irradiation.

Recommender-based bone tumour classification with radiographs-a link to the past.

OBJECTIVES: To develop an algorithm to link undiagnosed patients to previous patient histories based on radiographs, and simultaneous classification of multiple bone tumours to enable early and specific diagnosis. MATERIALS AND METHODS: For this retrospective study, data from 2000 to 2021 were curated from our database by two orthopaedic surgeons, a radiologist and a data scientist. Patients with complete clinical and pre-therapy radiographic data were eligible. To ensure feasibility, the ten most frequent primary tumour entities, confirmed histologically or by tumour board decision, were included. We implemented a ResNet and transformer model to establish baseline results. Our method extracts image features using deep learning and then clusters the k most similar images to the target image using a hash-based nearest-neighbour recommender approach that performs simultaneous classification by majority voting. The results were evaluated with precision-at-k, accuracy, precision and recall. Discrete parameters were described by incidence and percentage ratios. For continuous parameters, based on a normality test, respective statistical measures were calculated. RESULTS: Included were data from 809 patients (1792 radiographs; mean age 33.73 ± 18.65, range 3-89 years; 443 men), with Osteochondroma (28.31%) and Ewing sarcoma (1.11%) as the most and least common entities, respectively. The dataset was split into training (80%) and test subsets (20%). For k = 3, our model achieved the highest mean accuracy, precision and recall (92.86%, 92.86% and 34.08%), significantly outperforming state-of-the-art models (54.10%, 55.57%, 19.85% and 62.80%, 61.33%, 23.05%). CONCLUSION: Our novel approach surpasses current models in tumour classification and links to past patient data, leveraging expert insights. CLINICAL RELEVANCE STATEMENT: The proposed algorithm could serve as a vital support tool for clinicians and general practitioners with limited experience in bone tumour classification by identifying similar cases and classifying bone tumour entities. KEY POINTS: • Addressed accurate bone tumour classification using radiographic features. • Model achieved 92.86%, 92.86% and 34.08% mean accuracy, precision and recall, respectively, significantly surpassing state-of-the-art models. • Enhanced diagnosis by integrating prior expert patient assessments.

Automatic detection, segmentation, and classification of primary bone tumors and bone infections using an ensemble multi-task deep learning framework on multi-parametric MRIs: a multi-center study.

OBJECTIVES: To develop an ensemble multi-task deep learning (DL) framework for automatic and simultaneous detection, segmentation, and classification of primary bone tumors (PBTs) and bone infections based on multi-parametric MRI from multi-center. METHODS: This retrospective study divided 749 patients with PBTs or bone infections from two hospitals into a training set (N = 557), an internal validation set (N = 139), and an external validation set (N = 53). The ensemble framework was constructed using T1-weighted image (T1WI), T2-weighted image (T2WI), and clinical characteristics for binary (PBTs/bone infections) and three-category (benign/intermediate/malignant PBTs) classification. The detection and segmentation performances were evaluated using Intersection over Union (IoU) and Dice score. The classification performance was evaluated using the receiver operating characteristic (ROC) curve and compared with radiologist interpretations. RESULT: On the external validation set, the single T1WI-based and T2WI-based multi-task models obtained IoUs of 0.71 ± 0.25/0.65 ± 0.30 for detection and Dice scores of 0.75 ± 0.26/0.70 ± 0.33 for segmentation. The framework achieved AUCs of 0.959 (95%CI, 0.955-1.000)/0.900 (95%CI, 0.773-0.100) and accuracies of 90.6% (95%CI, 79.7-95.9%)/78.3% (95%CI, 58.1-90.3%) for the binary/three-category classification. Meanwhile, for the three-category classification, the performance of the framework was superior to that of three junior radiologists (accuracy: 65.2%, 69.6%, and 69.6%, respectively) and comparable to that of two senior radiologists (accuracy: 78.3% and 78.3%). CONCLUSION: The MRI-based ensemble multi-task framework shows promising performance in automatically and simultaneously detecting, segmenting, and classifying PBTs and bone infections, which was preferable to junior radiologists. CLINICAL RELEVANCE STATEMENT: Compared with junior radiologists, the ensemble multi-task deep learning framework effectively improves differential diagnosis for patients with primary bone tumors or bone infections. This finding may help physicians make treatment decisions and enable timely treatment of patients. KEY POINTS: • The ensemble framework fusing multi-parametric MRI and clinical characteristics effectively improves the classification ability of single-modality models. • The ensemble multi-task deep learning framework performed well in detecting, segmenting, and classifying primary bone tumors and bone infections. • The ensemble framework achieves an optimal classification performance superior to junior radiologists' interpretations, assisting the clinical differential diagnosis of primary bone tumors and bone infections.

Analgesic efficacy and safety of percutaneous thermal ablation plus cementoplasty for painful bone metastases: a systematic review and meta-analysis.

PURPOSE: To conduct a systematic review and meta-analysis of publications to evaluate the analgesic efficacy and safety of percutaneous thermal ablation (PTA) plus percutaneous cementoplasty (PCP) (PTA + PCP) for painful bone metastases. METHODS: We searched PubMed, Cochrane Library and Embase for articles published up to October 2022. Outcomes were a 10-point pain scale, morphine equivalents daily dose (MEDD) and complications. A subgroup confined to spinal bone metastases was analyzed. RESULTS: Twenty-one articles were selected for the analysis. The 21 selected articles involved a total of 661 cases. The pooled pain scales at pre-PTA + PCP, 1 day, 1 week and 1-, 3-, and 6 months post-PTA + PCP were 7.60 (95% confidence interval [CI], 7.26-7.95, I2 = 89%), 3.30 (95% CI, 2.25-4.82, I2 = 98%), 2.58 (95% CI, 1.99-3.35, I2 = 94%), 2.02 (95% CI, 1.50-2.71, I2 = 93%), 1.78 (95% CI, 1.26-2.53, I2 = 95%), and 1.62 (95% CI, 1.14-2.31, I2 = 88%), and in the subgroup, 7.97 (95% CI, 7.45-8.52, I2 = 86%), 3.01 (95% CI, 1.43-6.33, I2 = 98%), 2.95 (95% CI, 1.93-4.51, I2 = 95%), 2.34 (95% CI, 1.82-3.01, I2 = 68%), 2.18 (95% CI, 1.57-3.03, I2 = 78%), and 2.01 (95% CI, 1.16-3.48, I2 = 86%). Mean MEDD decreased up to 3 months post-PTA + PCP in 4 articles. The overall pooled major complication rate was 4% (95% CI, 2-6%, I2 = 2%). CONCLUSIONS: The updated systematic review and meta-analysis indicates that PTA + PCP for painful bone metastases is safe, and can lead to rapid and sustained pain reduction.

Spectrum of imaging findings of primary bone lymphoma in pediatric patients.

Primary bone lymphoma, a rare oncologic entity, may initially present with minimal symptoms. Presenting symptoms range from local pain and mild systemic symptoms to large palpable masses and pathologic fractures. The term "primary bone lymphoma" indicates the finding of bone involvement without other organ sites for at least 6 months. Although some radiological features may raise suspicion about this tumor form, there are no pathognomonic imaging findings, and the diagnosis will likely be delayed for a long time. The most critical radiological feature is soft tissue involvement associated with a preserved cortical layer, much more than expected for an infiltrating lesion. Anyway, very different radiological findings may be displayed in patients with primary bone lymphoma. Although these radiological features of primary bone lymphoma have been discussed in the literature by various authors, there is little data concerning imaging in pediatric patients. This paper aims to depict the possible spectrum of imaging features of primary bone lymphoma in the pediatric age, providing an exemplification pictorial essay extracted from a single institution experience in the year range period 2006-2022.

Combining morphological and functional imaging parameters to diagnose primary bone neoplasms in the skull base, spine and sacrum.

PURPOSE: Morphological magnetic resonance (MR) and computed tomography (CT) features are used in combination with histology for diagnosis and treatment selection of primary bone neoplasms. Isolated functional MRI parameters have shown potential in diagnosis. Our goal is to facilitate diagnosis of primary bone neoplasms of the skull base, mobile spine and sacrum, by a comprehensive approach, combining morphological and functional imaging parameters. MATERIALS AND METHODS: Pre-treatment MR of 80 patients with histologically proven diagnosis of a primary bone neoplasm of the skull base, mobile spine and sacrum were retrospectively analyzed for morphological and functional MRI parameters. Functional parameters were measured in 4 circular regions of interest per tumor placed on non-adjacent scan slices. Differences in values of functional parameters between different histologies were analyzed with Dunn's test. RESULTS: Chordomas were the predominant histology (60.0%). Most neoplasms (80.0%) originated in the midline and had geographical (78.2%) bone destruction. Amorphous-type calcification (pre-existing bone) was seen only in chordomas. Homogeneous contrast enhancement pattern was seen only in chondrosarcoma and plasmacytoma. Ktrans and Kep were significantly lower in both chordoma, and chondrosarcoma compared to giant cell tumor of the bone (p = 0.006 - 0.011), and plasmacytoma (p = 0.004 - 0.014). Highest diffusion-weighted MRI apparent diffusion coefficient (ADC) values corresponded to chondrosarcoma and were significantly higher to those of chordoma (p = 0.008). CONCLUSION: We identified the most discriminating morphological parameters and added functional MR parameters based on histopathological features that are useful in making a confident diagnosis of primary bone neoplasms in the skull base, mobile spine and sacrum.

A comprehensive exploration of artificial intelligence in orthopaedics within lower-middle-income countries: a narrative review.

Integrating Artificial Intelligence (AI) in orthopaedic within lower-middle-income countries (LMICs) promises landmark improvement in patient care. Delving into specific use cases-fracture detection, spine imaging, bone tumour classification, and joint surgery optimisation-the review illuminates the areas where AI can significantly enhance orthopaedic practices. AI could play a pivotal role in improving diagnoses, enabling early detection, and ultimately enhancing patient outcomes- crucial in regions with constrained healthcare services. Challenges to the integration of AI include financial constraints, shortage of skilled professionals, data limitations, and cultural and ethical considerations. Emphasising AI's collaborative role, it can act as a complementary tool working in tandem with physicians, aiming to address gaps in healthcare access and education. We need continued research and a conscientious approach, envisioning AI as a catalyst for equitable, efficient, and accessible orthopaedic healthcare for patients in LMICs.

Primary bone tumor detection and classification in full-field bone radiographs via YOLO deep learning model.

OBJECTIVES: Automatic bone lesions detection and classifications present a critical challenge and are essential to support radiologists in making an accurate diagnosis of bone lesions. In this paper, we aimed to develop a novel deep learning model called You Only Look Once (YOLO) to handle detecting and classifying bone lesions on full-field radiographs with limited manual intervention. METHODS: In this retrospective study, we used 1085 bone tumor radiographs and 345 normal bone radiographs from two centers between January 2009 and December 2020 to train and test our YOLO deep learning (DL) model. The trained model detected bone lesions and then classified these radiographs into normal, benign, intermediate, or malignant types. The intersection over union (IoU) was used to assess the model's performance in the detection task. Confusion matrices and Cohen's kappa scores were used for evaluating classification performance. Two radiologists compared diagnostic performance with the trained model using the external validation set. RESULTS: In the detection task, the model achieved accuracies of 86.36% and 85.37% in the internal and external validation sets, respectively. In the DL model, radiologist 1 and radiologist 2 achieved Cohen's kappa scores of 0.8187, 0.7927, and 0.9077 for four-way classification in the external validation set, respectively. The YOLO DL model illustrated a significantly higher accuracy for intermediate bone tumor classification than radiologist 1 (95.73% vs 88.08%, p = 0.004). CONCLUSIONS: The developed YOLO DL model could be used to assist radiologists at all stages of bone lesion detection and classification in full-field bone radiographs. KEY POINTS: • YOLO DL model can automatically detect bone neoplasms from full-field radiographs in one shot and then simultaneously classify radiographs into normal, benign, intermediate, or malignant. • The dataset used in this retrospective study includes normal bone radiographs. • YOLO can detect even some challenging cases with small volumes.

Deep learning-based diagnosis of osteoblastic bone metastases and bone islands in computed tomograph images: a multicenter diagnostic study.

OBJECTIVE: To develop and validate a deep learning (DL) model based on CT for differentiating bone islands and osteoblastic bone metastases. MATERIALS AND METHODS: The patients with sclerosing bone lesions (SBLs) were retrospectively included in three hospitals. The images from site 1 were randomly assigned to the training (70%) and intrinsic verification (10%) datasets for developing the two-dimensional (2D) DL model (single-slice input) and "2.5-dimensional" (2.5D) DL model (three-slice input) and to the internal validation dataset (20%) for evaluating the performance of both models. The diagnostic performance was evaluated using the internal validation set from site 1 and additional external validation datasets from site 2 and site 3. And statistically analyze the performance of 2D and 2.5D DL models. RESULTS: In total, 1918 SBLs in 728 patients in site 1, 122 SBLs in 71 patients in site 2, and 71 SBLs in 47 patients in site 3 were used to develop and test the 2D and 2.5D DL models. The best performance was obtained using the 2.5D DL model, which achieved an AUC of 0.996 (95% confidence interval [CI], 0.995-0.996), 0.958 (95% CI, 0.958-0.960), and 0.952 (95% CI, 0.951-0.953) and accuracies of 0.950, 0.902, and 0.863 for the internal validation set, the external validation set from site 2 and site 3, respectively. CONCLUSION: A DL model based on a three-slice CT image input (2.5D DL model) can improve the prediction of osteoblastic bone metastases, which can facilitate clinical decision-making. KEY POINTS: • This study investigated the value of deep learning models in identifying bone islands and osteoblastic bone metastases. • Three-slice CT image input (2.5D DL model) outweighed the 2D model in the classification of sclerosing bone lesions. • The 2.5D deep learning model showed excellent performance using the internal (AUC, 0.996) and two external (AUC, 0.958; AUC, 0.952) validation sets.

Osteoid osteomas of the hip: a well-recognized entity with a proclivity for misdiagnosis.

OBJECTIVES: The diagnosis of osteoid osteomas (OO) about the hip can be challenging as presenting symptoms can mimic other, more common, periarticular pathologies. Our aims were to identify the most common misdiagnoses and treatments, mean delay in diagnosis, characteristic imaging features and provide tips for avoiding diagnostic imaging pitfalls for patients with OO of the hip. METHODS: We identified 33 patients (34 tumors) with OO about the hip who were referred for radiofrequency ablation between 1998 and 2020. Imaging studies reviewed included radiographs (n = 29), CT (n = 34), and MRI (n = 26). RESULTS: The most common initial diagnoses were femoral neck stress fracture (n = 8), femoroacetabular impingement (FAI) (n = 7), and malignant tumor or infection (n = 4). The mean time from symptom onset to diagnosis of OO was 15 months (range, 0.4-84). The mean time from initial incorrect diagnosis to OO diagnosis was 9 months (range, 0-46). CONCLUSIONS: The diagnosis of OO of the hip is challenging, with up to 70% of cases initially misdiagnosed as a femoral neck stress fracture, FAI, bone tumor, or other joint pathology in our series. Consideration of OO in the differential diagnosis of hip pain in adolescent patients and awareness of the characteristic imaging findings are critical for making an accurate diagnosis. KEY POINTS: • The diagnosis of osteoid osteoma of the hip can be challenging, as demonstrated by long delays in time to initial diagnosis and high rates of misdiagnoses which can lead to inappropriate interventions. • Familiarity with the spectrum of imaging features of OO, especially on MRI, is imperative given the increase in the utilization of this modality for the evaluation of young patients with hip pain and FAI. • Consideration of OO in the differential diagnosis of hip pain in adolescent patients and awareness of the characteristic imaging findings, including bone marrow edema and the utility of CT, are critical for making a timely and accurate diagnosis.

What are the indications and survivorship of tumor endoprosthetic reconstructions for patients with extremity metastatic bone disease?

BACKGROUND AND OBJECTIVES: Given advances in therapies, endoprosthetic reconstruction (EPR) in metastatic bone disease (MBD) may be increasingly indicated. The objectives were to review the indications, and implant and patient survivorship in patients undergoing EPR for MBD. METHODS: A review of patients undergoing EPR for extremity MBD between 1992 and 2022 at two centers was performed. Surgical data, implant survival, patient survival, and implant failure modes were examined. RESULTS: One hundred fifteen patients were included with a median follow-up of 14.9 months (95% confidence interval [CI]: 9.2-19.3) and survival of 19.4 months (95% CI: 13.6-26.1). The most common diagnosis was renal cell carcinoma (34/115, 29.6%) and the most common location was proximal femur (43/115, 37.4%). Indications included: actualized fracture (58/115, 50.4%), impending fracture (30/115, 26.1%), and failed fixation (27/115, 23.5%). Implant failure was uncommon (10/115, 8.7%). Patients undergoing EPR for failed fixation were more likely to have renal or lung cancer (p = 0.006). CONCLUSIONS: EPRs were performed most frequently for renal cell carcinoma and in patients with a relatively favorable survival. EPR was indicated for failed previous fixation in 23.5% of cases, emphasizing the importance of predictive survival modeling. EPR can be a reliable and durable surgical option for patients with MBD.

Quality of life issues in patients with bone metastases: A systematic review.

INTRODUCTION: Bones are frequent sites of metastatic disease, observed in 30-75% of advanced cancer patients. Quality of life (QoL) is an important endpoint in studies evaluating the treatments of bone metastases (BM), and many patient-reported outcome tools are available. The primary objective of this systematic review was to compile a list of QoL issues relevant to BM and its interventions. The secondary objective was to identify common tools used to assess QoL in patients with BM, and the QoL issues they fail to address. METHODS: A search was conducted on Ovid MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials databases between 1946 and 27 January 2023 with the keywords "bone metastases", "quality of life", and "patient reported outcomes". Specific QoL issues in original research studies and the QoL tools used were extracted. RESULTS: The review identified the QoL issues most prevalent to BM in the literature. Physical and functional issues observed in patients included pain, interference with ambulation and daily activities, and fatigue. Psychological symptoms, such as helplessness, depression, and anxiety were also common. These issues interfered with patients' relationships and social activities. Items not mentioned in existing QoL tools were related to newer treatments of BM, such as pain flare, flu-like symptoms, and jaw pain due to osteonecrosis. CONCLUSIONS: This systematic review highlights that QoL issues for patients with BM have expanded over time due to advances in BM-directed treatments. If they are relevant, additional treatment-related QoL issues identified need to be validated prospectively by patients and added to current assessment tools.

Therapeutic validity and effectiveness of exercise interventions after lower limb-salvage surgery for sarcoma: a systematic review.

BACKGROUND: An increasing number of patients are surviving sarcoma after lower limb-salvage surgery (LSS) and are left with functional limitations. This systematic review aimed to determine the therapeutic validity and effectiveness of exercise interventions after lower limb-salvage surgery (LSS) for sarcoma. METHODS: A systematic review was conducted using formal narrative synthesis of intervention studies (with and without control group) identified through PubMed, Embase, Cochrane Library, CINAHL, and PEDro databases. Studies were included if participants were treated with LSS for unilateral lower limb sarcoma and followed an exercise intervention using active exercise, physical training, or rehabilitation before and/or after surgery. This review's outcome measures were interventions' therapeutic validity, assessed using the CONTENT scale (0 to 9); methodological quality, identified using the Downs & Black checklist (0 to 28); interventions' effectiveness, assessed based on differences in outcome measures between intervention and control groups; and certainty of evidence, classified according to the GRADE approach. RESULTS: Seven studies involving 214 participants were included. None of the included interventions were therapeutically valid (median 5, range 1-5). All but one study were of at least fair methodological quality (median 18, range 14-21). There was very low-quality evidence that exercise interventions resulted in increased knee range of motion (MD 10-15°) or compliance (MD 30%), and reduced functionality scores (MD -5%) compared to usual care. CONCLUSIONS: We found overall low therapeutic validity of interventions, performed in overall low-quality studies. Combined with the very low certainty of evidence, the results prevent drawing valid conclusions on the interventions' effectiveness. Future studies should aim for uniformity among their methodological approaches and outcome measures, using the CONTENT scale as a template to avert insufficient reporting. TRIAL REGISTRATION: PROSPERO CRD42021244635.

Aneurysmal bone cyst-like changes developed in melorheostosis with epiphyseal osteopoikilosis.

Aneurysmal bone cyst (ABC) is a rare and usually painful condition, representing about 1% of all bone tumors. A geographical lytic, expansile, and septated radiological pattern, with fluid-fluid levels on MRI, is classically displayed. ABC can be a primary bone lesion (70% of patients) or can arise in an underlying condition and is subsequently named "ABC-like changes" (30%). ABC-like changes are more frequently encountered in skeletal segments affected by chondroblastoma, fibrous dysplasia, giant cell tumor, osteoblastoma, non-ossifying fibroma, and osteosarcoma. In this article, we describe the first case of ABC-like changes developed in association with an ultra-rare sclerosing bone disease: melorheostosis. Melorheostosis is characterized by recognizable patterns on radiological studies with a pathological increased bone density and a cortical thickening within the periosteal or endosteal space, usually with a "dripping candle wax" appearance. More rarely, other different radiological patterns can be observed, such as "osteopatia striata-like," "osteoma-like," "myositis ossificans-like," and mixed patterns. Pain and limb hypotrophy are the most common clinical manifestations. We report the case of a Caucasian male with a clinic-radiological diagnosis of melorheostosis (with epiphyseal osteopoikilosis) since the age of twelve. At the age of nineteen, he suffered from increased pain in the proximal right thigh, and the radiological control revealed an expansive septated lesion at the right proximal femoral bone. The diagnosis of ABC-like changes developed in melorheostosis was obtained after CT-guided bone biopsy and confirmed by open-incisional biopsy.

Musculoskeletal oncology and thermal ablation: the current and emerging role of interventional radiology.

The role of interventional radiology (IR) is expanding. With new techniques being developed and tested, this radiology subspecialty is taking a step forward in different clinical scenarios, especially in oncology. Musculoskeletal tumoral diseases would definitely benefit from a low-invasive approach that could reduce mortality and morbidity in particular. Thermal ablation through IR has already become important in the palliation and consolidation of bone metastases, oligometastatic disease, local recurrences, and treating specific benign tumors, with a more tailored approach, considering the characteristics of every patient. As image-guided ablation techniques lower their invasiveness and increase their efficacy while the collateral effects and complications decrease, they become more relevant and need to be considered in patient care pathways and clinical management, to improve outcomes. We present a literature review of the different percutaneous and non-invasive image-guided thermal ablation methods that are currently available and that could in the future become relevant to manage musculoskeletal oncologic diseases.

Initial Characterization of Focal Bone Lesions with Conventional Radiographs or Computed Tomography: Diagnostic Performance and Interobserver Agreement Assessment.

Objectives: To ascertain the role of CT and conventional radiographs for the initial characterization of focal bone lesions.Methods: Images from 184 patients with confirmed bone tumors included in an ethics committee-approved study were retrospectively evaluated. The reference for benign-malignant distribution was based on histological analysis and long-term follow-up. Radiographs and CT features were analyzed by 2 independent musculoskeletal radiologists blinded to the final diagnosis. Lesion margins, periosteal reaction, cortical lysis, endosteal scalloping, presence of pathologic fracture, and lesion mineralization were evaluated. Results: The benign-malignant distribution in the study population was 68.5-31.5% (126 benign and 58 malignant). In the lesions that could be seen in both radiographs and CT, the performance of these methods for the benign-malignant differentiation was similar (accuracy varying from 72.8% to 76.5%). The interobserver agreement for the overall evaluation of lesion aggressiveness was considerably increased on CT compared to radiographs (Kappa of .63 vs .22). With conventional radiographs, 18 (9.7%) and 20 (10.8%) of the lesions evaluated were not seen respectively by readers 1 and 2. Among these unseen lesions, 50%-61.1% were located in the axial skeleton. Compared to radiographs, the number of lesions with cortical lysis and endosteal scalloping was 26-34% higher with CT. Conclusion: Although radiographs remain the primary imaging tool for lesions in the peripheral skeleton, CT should be performed for axial lesions. CT imaging can assess the extent of perilesional bone lysis more precisely than radiographs with a better evaluation of lesion fracture risk.

A deep learning-machine learning fusion approach for the classification of benign, malignant, and intermediate bone tumors.

OBJECTIVES: To build and validate deep learning and machine learning fusion models to classify benign, malignant, and intermediate bone tumors based on patient clinical characteristics and conventional radiographs of the lesion. METHODS: In this retrospective study, data were collected with pathologically confirmed diagnoses of bone tumors between 2012 and 2019. Deep learning and machine learning fusion models were built to classify tumors as benign, malignant, or intermediate using conventional radiographs of the lesion and potentially relevant clinical data. Five radiologists compared diagnostic performance with and without the model. Diagnostic performance was evaluated using the area under the curve (AUC). RESULTS: A total of 643 patients' (median age, 21 years; interquartile range, 12-38 years; 244 women) 982 radiographs were included. In the test set, the binary category classification task, the radiological model of classification for benign/not benign, malignant/nonmalignant, and intermediate/not intermediate had AUCs of 0.846, 0.827, and 0.820, respectively; the fusion models had an AUC of 0.898, 0.894, and 0.865, respectively. In the three-category classification task, the radiological model achieved a macro average AUC of 0.813, and the fusion model had a macro average AUC of 0.872. In the observation test, the mean macro average AUC of all radiologists was 0.819. With the three-category classification fusion model support, the macro AUC improved by 0.026. CONCLUSION: We built, validated, and tested deep learning and machine learning models that classified bone tumors at a level comparable with that of senior radiologists. Model assistance may somewhat help radiologists' differential diagnoses of bone tumors. KEY POINTS: • The deep learning model can be used to classify benign, malignant, and intermediate bone tumors. • The machine learning model fusing information from radiographs and clinical characteristics can improve the classification capacity for bone tumors. • The diagnostic performance of the fusion model is comparable with that of senior radiologists and is potentially useful as a complement to radiologists in a bone tumor differential diagnosis.

Development and evaluation of machine learning models based on X-ray radiomics for the classification and differentiation of malignant and benign bone tumors.

OBJECTIVES: To develop and validate machine learning models to distinguish between benign and malignant bone lesions and compare the performance to radiologists. METHODS: In 880 patients (age 33.1 ± 19.4 years, 395 women) diagnosed with malignant (n = 213, 24.2%) or benign (n = 667, 75.8%) primary bone tumors, preoperative radiographs were obtained, and the diagnosis was established using histopathology. Data was split 70%/15%/15% for training, validation, and internal testing. Additionally, 96 patients from another institution were obtained for external testing. Machine learning models were developed and validated using radiomic features and demographic information. The performance of each model was evaluated on the test sets for accuracy, area under the curve (AUC) from receiver operating characteristics, sensitivity, and specificity. For comparison, the external test set was evaluated by two radiology residents and two radiologists who specialized in musculoskeletal tumor imaging. RESULTS: The best machine learning model was based on an artificial neural network (ANN) combining both radiomic and demographic information achieving 80% and 75% accuracy at 75% and 90% sensitivity with 0.79 and 0.90 AUC on the internal and external test set, respectively. In comparison, the radiology residents achieved 71% and 65% accuracy at 61% and 35% sensitivity while the radiologists specialized in musculoskeletal tumor imaging achieved an 84% and 83% accuracy at 90% and 81% sensitivity, respectively. CONCLUSIONS: An ANN combining radiomic features and demographic information showed the best performance in distinguishing between benign and malignant bone lesions. The model showed lower accuracy compared to specialized radiologists, while accuracy was higher or similar compared to residents. KEY POINTS: • The developed machine learning model could differentiate benign from malignant bone tumors using radiography with an AUC of 0.90 on the external test set. • Machine learning models that used radiomic features or demographic information alone performed worse than those that used both radiomic features and demographic information as input, highlighting the importance of building comprehensive machine learning models. • An artificial neural network that combined both radiomic and demographic information achieved the best performance and its performance was compared to radiology readers on an external test set.

Diagnosis of joint invasion in patients with malignant bone tumors: value and reproducibility of direct and indirect signs on MR imaging.

OBJECTIVES: To evaluate the performance and reproducibility of MR imaging features in the diagnosis of joint invasion (JI) by malignant bone tumors. METHODS: MR images of patients with and without JI (n = 24 each), who underwent surgical resection at our institution, were read by three radiologists. Direct (intrasynovial tumor tissue (ITT), intraarticular destruction of cartilage/bone, invasion of capsular/ligamentous insertions) and indirect (tumor size, signal alterations of epiphyseal/transarticular bone (bone marrow replacement/edema-like), synovial contrast enhancement, joint effusion) signs of JI were assessed. Odds ratios, sensitivity, specificity, PPV, NPV, and reproducibilities (Cohen's and Fleiss' κ) were calculated for each feature. Moreover, the diagnostic performance of combinations of direct features was assessed. RESULTS: Forty-eight patients (28.7 ± 21.4 years, 26 men) were evaluated. All readers reliably assessed the presence of JI (sensitivity = 92-100 %; specificity = 88-100%, respectively). Best predictors for JI were direct visualization of ITT (OR = 186-229, p < 0.001) and destruction of intraarticular bone (69-324, p < 0.001). Direct visualization of ITT was also highly reliable in assessing JI (sensitivity, specificity, PPV, NPV = 92-100 %), with excellent reproducibility (κ = 0.83). Epiphyseal bone marrow replacement and synovial contrast enhancement were the most sensitive indirect signs, but lacked specificity (29-54%). By combining direct signs with high specificity, sensitivity was increased (96 %) and specificity (100 %) was maintained. CONCLUSION: JI by malignant bone tumors can reliably be assessed on preoperative MR images with high sensitivity, specificity, and reproducibility. Particularly direct visualization of ITT, destruction of intraarticular bone, and a combination of highly specific direct signs were valuable, while indirect signs were less predictive and specific. KEY POINTS: • Direct visualization of intrasynovial tumor was the single most sensitive and specific (92-100%) MR imaging sign of joint invasion. • Indirect signs of joint invasion, such as joint effusion or synovial enhancement, were less sensitive and specific compared to direct signs. • A combination of the most specific direct signs of joint invasion showed best results with perfect specificity and PPV (both 100%) and excellent sensitivity and NPV (both 96 %).