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.

To construct a nomogram based on clinical factors and paraspinal muscle features to predict vertebral fractures occurring after acute osteoporotic vertebral compression fracture (OVCF). We retrospectively enrolled 307 patients with acute OVCF between January 2013 and August 2022, and performed magnetic resonance imaging of the L3/4 and L4/5 intervertebral discs (IVDs) to estimate the cross-sectional area (CSA) and degree of fatty infiltration (FI) of the paraspinal muscles. We also collected clinical and radiographic data. We used univariable and multivariable Cox proportional hazards models to identify factors that should be included in the predictive nomogram. Post-OVCF vertebral fracture occurred within 3, 12, and 24 months in 33, 69, and 98 out of the 307 patients (10.8%, 22.5%, and 31.9%, respectively). Multivariate analysis revealed that this event was associated with percutaneous vertebroplasty treatment, higher FI at the L3/4 IVD levels of the psoas muscle, and lower relative CSA of functional muscle at the L4/5 IVD levels of the multifidus muscle. Area under the curve values for subsequent vertebral fracture at 3, 12, and 24 months were 0.711, 0.724, and 0.737, respectively, indicating remarkable accuracy of the nomogram. We developed a model for predicting post-OVCF vertebral fracture from diagnostic information about prescribed treatment, FI at the L3/4 IVD levels of the psoas muscle, and relative CSA of functional muscle at the L4/5 IVD levels of the multifidus muscle. This model could facilitate personalized predictions and preventive strategies.

Objectives: Osteoporotic vertebral fractures (OVFs) are a critical public health concern requiring urgent attention, and severe OVFs impose substantial health and economic burdens on patients and society. Analysis of the risk factors for severe OVF is imperative to actively prevent the occurrence of this degenerative disorder. This study aimed to investigate the risk factors associated with the severity of OVF, with a specific focus on changes in the paraspinal muscles.Methods: A total of 281 patients with a first‐time single‐level acute OVF between January 2016 and January 2023 were enrolled in the study. Clinical and radiological data were collected and analyzed. The cross‐sectional area (CSA) and degree of fatty infiltration (FI) of the paraspinal muscles, including the multifidus muscles (MFMs), erector spinae muscles (ESMs), and psoas major muscles (PSMs), were measured by magnetic resonance imaging (MRI) of the L4/5 intervertebral discs. According to the classification system of osteoporotic fractures (OF classification) and recommended treatment plan, OVFs were divided into a low‐grade OF group and a high‐grade OF group. Univariate and multivariate logistic regression analyse s were performed to identify risk factors associated with the severity of OVF.Results: Ninety‐eight patients were included in the low‐grade OF group, and 183 patients were included in the high‐grade OF group. Univariate analysis revealed a significantly higher incidence of a high degree of FI of MFMs (OR = 1.71, p = 0.002) and ESMs (OR = 1.56, p = 0.021) in the high‐grade OF group. Further multivariate logistic regression analysis demonstrated that a high degree of FI of the MFMs (OR = 1.71, p = 0.002) is an independent risk factor for the severity of OVF.Conclusion: A high degree of FI of the MFMs was identified as an independent risk factor for the severity of OVF. Decreasing the degree of FI in the MFMs might lower the incidence of the severity of OVF, potentially reducing the necessity for surgical intervention in OVF patients.

Primary biliary non-Hodgkin's lymphoma (PBNHL) is an extremely rare biliary malignancy often misdiagnosed as cholangiocarcinoma due to nonspecific imaging. Delayed recognition can result in unnecessary surgery and treatment delays, as chemotherapy is the primary therapy. We report an elderly male presenting with upper abdominal pain, initially suspected of cholangiocarcinoma, but diagnosed with B-cell non-Hodgkin's lymphoma (B-NHL) on histopathology. Due to poor baseline performance and multiple acute complications, he received only two days of R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) chemotherapy and subsequently died from sepsis. We report a case of PBNHL, emphasizing its characteristic cholangiographic features and the potential contribution of immunosuppressive conditions, including Epstein-Barr virus (EBV) infection.