Search

Study Design: Basic science (finite element analysis). Objectives: Pedicle subtraction osteotomy (PSO) at L5 is an effective treatment for sagittal imbalance, especially in select cases of patients showing kyphosis with the apex at L4-L5 but has been scarcely investigated. The aim of this study was to simulate various "high-demand" instrumentation approaches, including varying numbers of rods and sacropelvic implants, for the stabilization of a PSO at L5. Methods: A finite element model of T10-pelvis was modified to simulate posterior fixation with pedicle screws and rods from T10 to S1, alone or in combination with an L5 PSO. Five additional configurations were then created by employing rods and novel porous fusion/fixation implants across the sacroiliac joints, in varying numbers. All models were loaded using pure moments of 7.5 Nm in flexion-extension, lateral bending, and axial rotation. Results: The osteotomy resulted in a general increase in motion and stresses in posterior rods and S1 pedicle screws. When the number of rods was varied, three- and four-rod configurations were effective in limiting the maximal rod stresses; values approached those of posterior fixation with no osteotomy. Maximum stresses in the accessory rods were similar to or less than those observed in the primary rods. Multiple sacropelvic implants were effective in reducing range of motion, particularly of the SIJ. Conclusions: Multi-rod constructs and sacropelvic fixation generally reduced maximal implant stresses and motion in comparison with standard posterior fixation, suggesting a reduced risk of rod breakage and increased joint stability, respectively, when a high-demand construct is utilized for the correction of sagittal imbalance. [ABSTRACT FROM AUTHOR]

Study Design: Retrospective study. Objectives: Huge amounts of images and medical reports are being generated in radiology departments. While these datasets can potentially be employed to train artificial intelligence tools to detect findings on radiological images, the unstructured nature of the reports limits the accessibility of information. In this study, we tested if natural language processing (NLP) can be useful to generate training data for deep learning models analyzing planar radiographs of the lumbar spine. Methods: NLP classifiers based on the Bidirectional Encoder Representations from Transformers (BERT) model able to extract structured information from radiological reports were developed and used to generate annotations for a large set of radiographic images of the lumbar spine (N = 10 287). Deep learning (ResNet-18) models aimed at detecting radiological findings directly from the images were then trained and tested on a set of 204 human-annotated images. Results: The NLP models had accuracies between 0.88 and 0.98 and specificities between 0.84 and 0.99; 7 out of 12 radiological findings had sensitivity >0.90. The ResNet-18 models showed performances dependent on the specific radiological findings with sensitivities and specificities between 0.53 and 0.93. Conclusions: NLP generates valuable data to train deep learning models able to detect radiological findings in spine images. Despite the noisy nature of reports and NLP predictions, this approach effectively mitigates the difficulties associated with the manual annotation of large quantities of data and opens the way to the era of big data for artificial intelligence in musculoskeletal radiology. [ABSTRACT FROM AUTHOR]

In order to analyze the information carried by arterial blood pressure (ABP) variability, a multivariate parametric model of interactions involving systolic ABP (SAP), diastolic ABP (DAP), pulse pressure (PP), heart period (HP), and respiration is proposed. The model defines SAP as sum of the preceding DAP and PP values; DAP model accounts for arterial baroreflex, diastolic runoff; PP reflects changes in stroke volume related to respiration and HP, afterload; equation residuals reveal other vascular and cardiac output modulations. The model was applied to data from nine young volunteers (aged 29 [+ or -] 6 years) during supine cycling at 10%, 20%, and 30% of their maximum effort. Significant basal values and changes across the epochs of the experiment were found in all hemodynamic parameters describing fast, beat-by-beat responses; in SAP and PP total power, DAP low- and high-frequency power (LF, HF), PP very low frequency (VLF), and LF and HF power. A primary role of vascular control through DAP and PP was emphasized by the considered feedbacks and the model residuals. The model proved to be able to assess beat-by-beat cardiovascular interactions and offer a comprehensive view of arterial tree control. Index Terms--Arterial pressure variability, cardiovascular control, closed loop, exercise, multivariate modeling.