Deep Learning on Multiphysical Features and Hemodynamic Modeling for Abdominal Aortic Aneurysm Growth Prediction.

Prediction of abdominal aortic aneurysm (AAA) growth is of essential importance for the early treatment and surgical intervention of AAA. Capturing key features of vascular growth, such as blood flow and intraluminal thrombus (ILT) accumulation play a crucial role in uncovering the intricated mechanism of vascular adaptation, which can ultimately enhance AAA growth prediction capabilities. However, local correlations between hemodynamic metrics, biological and morphological characteristics, and AAA growth rates present high inter-patient variability that results in that the temporal-spatial biochemical and mechanical processes are still not fully understood. Hence, this study aims to integrate the physics-based knowledge with deep learning with a patch-based convolutional neural network (CNN) approach by incorporating important multiphysical features relating to its pathogenesis for validating its impact on AAA growth prediction. For this task, we observe that the unstructured multiphysical features cannot be directly employed in the kernel-based CNN. To tackle this issue, we propose a parameterization of features to leverage the spatio-temporal relations between multiphysical features. The proposed architecture was tested on different combinations of four features including radius, intraluminal thrombus thickness, time-average wall shear stress, and growth rate from 54 patients with 5-fold cross-validation with two metrics, a root mean squared error (RMSE) and relative error (RE). We conduct extensive experiments on AAA patients, the results show the effect of leveraging multiphysical features and demonstrate the superiority of the presented architecture to previous state-of-the-art methods in AAA growth prediction.

The analgesic potency dose of remifentanil to minimize stress response induced by intubation and measurement uncertainty of Surgical Pleth Index.

BACKGROUND: The aim of this study was to evaluate the analgesic potency dose of remifentanil to maintain Surgical Pleth Index (SPI) values at less than 50 after intubation in patients undergoing general anesthesia with target-controlled infusion of propofol and remifentanil. METHODS: We randomly allocated 120 patients to receive one of three remifentanil target effect-site concentrations (5, 7, or 9 ng×mL-1) during intubation. The target effect-site concentrations of propofol were adjusted within a range of 2.5-3 µg×mL-1 to maintain bispectral index values at less than 60 during anesthesia induction. A reusable SPI sensor was placed on the index finger of the arm, and the SPI values were continuously recorded. The predicted probability for maintaining the SPI values at less than 50 after intubation against the cumulative amount of remifentanil was analyzed using logistic regression. The measurands were the baseline SPI value in patients without pain scheduled for surgery, and the maximal SPI value after intubation in patients receiving remifentanil with a target effect-site concentration of 7 ng×mL-1. RESULTS: The estimated cumulative amount of remifentanil associated with a 50% and 95% probability of maintaining the SPI values at less than 50 after intubation were 135.0 µg and 330.4 µg, respectively. The estimated expanded uncertainty for the baseline and maximal SPI values after intubation in patients scheduled for surgery were 54.9±44.4 and 54.1±37.9, respectively, which corresponded to a confidence level of approximately 95%. CONCLUSIONS: The analgesic potency dose of remifentanil to maintain SPI values at less than 50 after intubation was 135.0 µg.

Cranial growth after distraction osteogenesis of the craniosynostosis.

The authors describe the continuance of the growth of the distracted cranium after the reshaping of the cranium by distraction osteogenesis (DO) in children with simple symmetric and asymmetric craniosynostosis. From 2000 until 2002, 9 children with simple craniosynostosis underwent cranial reshaping by gradual distraction using an external distraction device. Four patients have symmetric deformities caused by bicoronal and sagittal craniosynostosis, and 5 patients have asymmetric deformities caused by unicoronal and unilambdoidal craniosynostosis. The distraction device was developed and applied by the author. Preoperative simulation surgery was done on the three-dimensional rapid prototyped model and on the three-dimensional computerized tomography scan to determine the favorable osteotomy line. The distraction rate was from 1 to 1.5 mm/d, and the latency period was from 1 to 5 days. The extent of distraction was determined on the basis of the results of simulation surgery and the change of external appearance. Evaluation of the growth of reshaped cranium was processed from the data of the reconstructed three-dimensional computerized tomography scans before operation, immediate end of distraction, and the last follow-up time. The anteroposterior length and bitemporal width were measured in symmetric synostosis cases, and the distance from supratrochlear notch to occiput was measured in asymmetric synostosis cases. The results showed that the immediate morphologic changes of cranium after DO were maintained in both symmetric and asymmetric synostosis up to the last follow-up without evidence of relapse. Cases of asymmetric deformity also showed that the affected side and the unaffected side had grown with the maintenance of the symmetry that was corrected at the immediate end of the distraction. The cranium modified by the DO was well maintained with the children's growth without any signs of recurrent restricted growth of the original disease. The corrected symmetry of asymmetric deformity was well maintained during a long-term follow-up period as well.