Radiographic outcomes and non-union factor analysis in fragmentary segmental femoral shaft fractures (AO/OTA 32C3) treated with reamed antegrade nailing.

This study retrospectively assessed radiographic outcomes and risk factors associated with non-union in femoral shaft fragmentary segmental fractures (AO/OTA 32C3) treated with reamed antegrade intra-medullary nailing. Radiological outcomes, including union and alignment, were evaluated. The risk factors for non-union were investigated, including demographics and treatment-related characteristics, such as the number of interlocking screws, segmentation length, main third fragment length, distance of the main third fragment, width ratio and exposed nail length in one cortex from immediate post-operative radiographs. Multivariate logistic regression was used for statistical analysis. Among 2295 femoral shaft fracture patients from three level-1 trauma centers, 51 met the inclusion criteria. The radiological union was achieved in 37 patients (73%) with a mean union time of 10.7 ± 4.8 months. The acceptable axial alignment was observed in 30 patients (59%). Multiple logistic regression analysis identified only exposed nail length as a significant risk factor for non-union (odds ratio: 1.599, p = 0.003) and the cut-off value was 19.1 mm (sensitivity, 0.786; specificity, 0.811). The study revealed high rates of non-union (27%) and malalignment (41%). Therefore, patients who underwent intramedullary nailing with an exposed nail length greater than 19.1 mm or about twice the nail diameter should be cautioned of the potential non-union.

Deep learning model integrating radiologic and clinical data to predict mortality after ischemic stroke.

Objective: Most prognostic indexes for ischemic stroke mortality lack radiologic information. We aimed to create and validate a deep learning-based mortality prediction model using brain diffusion weighted imaging (DWI), apparent diffusion coefficient (ADC), and clinical factors. Methods: Data from patients with ischemic stroke who admitted to tertiary hospital during acute periods from 2013 to 2019 were collected and split into training (n = 1109), validation (n = 437), and internal test (n = 654). Data from patients from secondary cardiovascular center was used for external test set (n = 507). The algorithm for predicting mortality, based on DWI and ADC (DLP_DWI), was initially trained. Subsequently, important clinical factors were integrated into this model to create the integrated model (DLP_INTG). The performance of DLP_DWI and DLP_INTG was evaluated by using time-dependent area under the receiver operating characteristic curves (TD AUCs) and Harrell concordance index (C-index) at one-year mortality. Results: The TD AUC of DLP_DWI was 0.643 in internal test set, and 0.785 in the external dataset. DLP_INTG had a higher performance at predicting one-year mortality than premise score in internal dataset (TD- AUC: 0.859 vs. 0.746; p = 0.046), and in external dataset (TD- AUC: 0.876 vs. 0.808; p = 0.007). DLP_DWI and DLP_INTG exhibited strong discrimination for the high-risk group for one-year mortality. Interpretation: A deep learning model using brain DWI, ADC and the clinical factors was capable of predicting mortality in patients with ischemic stroke.