Aspirin improves transplant-free survival after TIPS implantation in patients with refractory ascites: a retrospective multicentre cohort study

BACKGROUND AND AIMS

Transjugular intrahepatic portosystemic shunt (TIPS) implantation is an established procedure to treat portal hypertension. Impact of administration of aspirin on transplant-free survival after TIPS remains unknown.

METHODS

A multicenter retrospective analysis including patients with TIPS implantation between 2011 and 2018 at three tertiary German Liver Centers was performed. N = 583 patients were included. Survival analysis was performed in a matched cohort after propensity score matching. Patients were grouped according to whether aspirin was (PSM-aspirin-cohort) or was not (PSM-no-aspirin-cohort) administered after TIPS. Primary endpoint of the study was transplant-free survival at 12 months after TIPS.

RESULTS

Aspirin improved transplant-free survival 12 months after TIPS with 90.7% transplant-free survival compared to 80.0% (p = 0.001) after PSM. Separated by TIPS indication, aspirin did improve transplant-free survival in patients with refractory ascites significantly (89.6% vs. 70.6% transplant-free survival, p < 0.001), while no significant effect was observed in patients with refractory variceal bleeding (91.1% vs. 92.2% transplant-free survival, p = 0.797).

CONCLUSION

This retrospective multicenter study provides first data indicating a beneficial effect of aspirin on transplant-free survival after TIPS implantation in patients with refractory ascites.

Predicting Tree Species From 3D Laser Scanning Point Clouds Using Deep Learning

Automated species classification from 3D point clouds is still a challenge. It is, however, an important task for laser scanning-based forest inventory, ecosystem models, and to support forest management. Here, we tested the performance of an image classification approach based on convolutional neural networks (CNNs) with the aim to classify 3D point clouds of seven tree species based on 2D representation in a computationally efficient way. We were particularly interested in how the approach would perform with artificially increased training data size based on image augmentation techniques. Our approach yielded a high classification accuracy (86%) and the confusion matrix revealed that despite rather small sample sizes of the training data for some tree species, classification accuracy was high. We could partly relate this to the successful application of the image augmentation technique, improving our result by 6% in total and 13, 14, and 24% for ash, oak and pine, respectively. The introduced approach is hence not only applicable to small-sized datasets, it is also computationally effective since it relies on 2D instead of 3D data to be processed in the CNN. Our approach was faster and more accurate when compared to the point cloud-based "PointNet" approach.