RESUMO
Background: Although chest tube-omitted video-assisted thoracoscopic surgery (VATS) has been proven to be safe and efficacious, its universal application is precluded by a varying morbidity rate due to a lack of standardization. Since digital chest drainage has already shown improved accuracy and consistency in the management of postoperative air leak, we incorporated it in the strategy of intraoperative chest tube withdrawal, aiming to achieve better results. Methods: We collected the clinical data of 114 consecutive patients who underwent elective uniportal VATS pulmonary wedge resection at the Shanghai Pulmonary Hospital from May 2021 to February 2022. Their chest tubes were withdrawn intraoperatively after an air-tightness test facilitated by digital drainage: the end flow rate had to be kept ≤30 mL/min for >15 s at the setting of -8 cmH2O suctioning. The recordings and patterns of the air suctioning process were documented and analyzed as potential standards of chest tube withdrawal. Results: The mean age of the patients was 49.7±11.7 years. The mean size of the nodules was 1.0±0.2 cm. The location of the nodules encompassed all lobes, and 90 (78.9%) patients received preoperative localization. The postoperative morbidity and mortality rates were 7.0% and 0%, respectively. Six patients had clinically overt pneumothorax and two patients had postoperative bleeding that required intervention. All of the patients recovered on conservative treatment except for one case of pneumothorax that required additional tube thoracostomy. The median length of postoperative stay was 2 days; and the median time of suctioning, peak flow rate, and end flow rate were 126 s, 210 mL/min, and 0 mL/min, respectively. The median numeric rating scale for pain was 1 on postoperative day (POD) 1 and 0 on the day of discharge. Conclusions: Chest tube-free VATS assisted by digital drainage is feasible with minimal morbidity. Its strength of quantitative air leak monitoring produces important measurements for the prediction of postoperative pneumothorax and future standardization of the procedure.
RESUMO
Composite-database micro-expression recognition is attracting increasing attention as it is more practical for real-world applications. Though the composite database provides more sample diversity for learning good representation models, the important subtle dynamics are prone to disappearing in the domain shift such that the models greatly degrade their performance, especially for deep models. In this paper, we analyze the influence of learning complexity, including input complexity and model complexity, and discover that the lower-resolution input data and shallower-architecture model are helpful to ease the degradation of deep models in composite-database task. Based on this, we propose a recurrent convolutional network (RCN) to explore the shallower-architecture and lower-resolution input data, shrinking model and input complexities simultaneously. Furthermore, we develop three parameter-free modules (i.e., wide expansion, shortcut connection and attention unit) to integrate with RCN without increasing any learnable parameters. These three modules can enhance the representation ability in various perspectives while preserving not-very-deep architecture for lower-resolution data. Besides, three modules can further be combined by an automatic strategy (a neural architecture search strategy) and the searched architecture becomes more robust. Extensive experiments on the MEGC2019 dataset (composited of existing SMIC, CASME II and SAMM datasets) have verified the influence of learning complexity and shown that RCNs with three modules and the searched combination outperform the state-of-the-art approaches.
