BACKGROUND: Pathogenic concepts of right ventricular (RV) failure in pulmonary arterial hypertension focus on a critical loss of microvasculature. However, the methods underpinning prior studies did not take into account the 3-dimensional (3D) aspects of cardiac tissue, making accurate quantification difficult. We applied deep-tissue imaging to the pressure-overloaded RV to uncover the 3D properties of the microvascular network and determine whether deficient microvascular adaptation contributes to RV failure. METHODS: Heart sections measuring 250-µm-thick were obtained from mice after pulmonary artery banding (PAB) or debanding PAB surgery and properties of the RV microvascular network were assessed using 3D imaging and quantification. Human heart tissues harvested at the time of transplantation from pulmonary arterial hypertension cases were compared with tissues from control cases with normal RV function. RESULTS: Longitudinal 3D assessment of PAB mouse hearts uncovered complex microvascular remodeling characterized by tortuous, shorter, thicker, highly branched vessels, and overall preserved microvascular density. This remodeling process was reversible in debanding PAB mice in which the RV function recovers over time. The remodeled microvasculature tightly wrapped around the hypertrophied cardiomyocytes to maintain a stable contact surface to cardiomyocytes as an adaptation to RV pressure overload, even in end-stage RV failure. However, microvasculature-cardiomyocyte contact was impaired in areas with interstitial fibrosis where cardiomyocytes displayed signs of hypoxia. Similar to PAB animals, microvascular density in the RV was preserved in patients with end-stage pulmonary arterial hypertension, and microvascular architectural changes appeared to vary by etiology, with patients with pulmonary veno-occlusive disease displaying a lack of microvascular complexity with uniformly short segments. CONCLUSIONS: 3D deep tissue imaging of the failing RV in PAB mice, pulmonary hypertension rats, and patients with pulmonary arterial hypertension reveals complex microvascular changes to preserve the microvascular density and maintain a stable microvascular-cardiomyocyte contact. Our studies provide a novel framework to understand microvascular adaptation in the pressure-overloaded RV that focuses on cell-cell interaction and goes beyond the concept of capillary rarefaction.

Purchasing air tickets by the lowest price is a challenging task for consumers since the prices might fluctuate over time influenced by several factors. In order to support users’ decision, some price prediction techniques have been developed. Considering that this problem could be solved by multi-target approaches from Machine Learning, this work proposes a novel method looking forward to obtaining an improvement in air ticket prices prediction. The method, called Deep Regressor Stacking (DRS), applies a naive deep learning methodology to reach more accurate predictions. To evaluate the contribution of the DRS, it was compared with the competence of the single-target regression and two state-of-the-art multi-target regressions (Stacked Single Target and Ensemble of Regressor Chains). All four approaches were performed based on Random Forest and Support Vector Machine algorithms over two real-life airfares datasets. After results, it was concluded DRS outperformed the other three methods, being the most indicated (most predictive) to assist air passengers in the prediction of flight ticket price.

With data collected by Internet of Things sensors, deep learning (DL) models can forecast the generation capacity of photovoltaic (PV) power plants. This functionality is especially relevant for PV power operators and users as PV plants exhibit irregular behavior related to environmental conditions. However, DL models are vulnerable to adversarial examples, which may lead to increased predictive error and wrong operational decisions. This work proposes a new scheme to detect adversarial examples and mitigate their impact on DL forecasting models. This approach is based on one-class classifiers and features extracted from the data inputted to the forecasting models. Tests were performed using data collected from a real-world PV power plant along with adversarial samples generated by the Fast Gradient Sign Method under multiple attack patterns and magnitudes. One-class Support Vector Machine and Local Outlier Factor were evaluated as detectors of attacks to Long-Short Term Memory and Temporal Convolutional Network forecasting models. According to the results, the proposed scheme showed a high capability of detecting adversarial samples with an average F1-score close to 90%. Moreover, the detection and mitigation approach strongly reduced the prediction error increase caused by adversarial samples.