Search

The two-dimensional hydrodynamic flood models are computationally expensive, especially in large-scale watersheds, which limits their application when several simulation runs are required or in large-scale catchments. More recently, the literature presented a significant increase in the use of Deep Learning as an alternative to hydrodynamic models because of their computational efficiency. However, studies show that Artificial Neural Networks (ANN) applications for flood prediction lack the ability to predict floods outside the training datasets satisfactorily. In this study, we used a conditional generative adversarial network (cGAN) aiming to generalize pluvial flood predictions (cGAN-Flood) in catchments not included in the training dataset. Our approach uses two generators to solve a rain-on-grid problem by first identifying wet cells and then calculating the water depths. cGAN-Flood distributes a target flood volume (vt) in a given catchment, which can be calculated via water balance from any hydrological simulations. Our approach was trained on ten and tested on five urban catchments. The testing areas have distinct characteristics and were compared to HEC-RAS outputs for different rainfall magnitudes and surface roughness. The averages of the root square mean error (RSME) and critical success index (CSI) were 0.12m and 70.1%, respectively. The successful water depths predictions in the testing areas show that cGAN-Flood can generalize. cGAN-Flood, combined with an SWMM model to compute vt, was 250 times faster than HEC-RAS. Due to its computational efficiency and accuracy, cGAN-Flood can be applied when fast simulations are necessary, and it can be a viable for flood forecasts in large-scale watersheds., The 28th IUGG General Assembly (IUGG2023) (Berlin 2023)

Using data from the PRIAMO study, we investigated non-motor symptoms (NMS) versus frontal lobe dysfunction in patients with idiopathic Parkinson disease (PD); 808 patients with PD and 118 with atypical parkinsonisms (AP) were consecutively enrolled at 55 Centers in Italy. Twelve categories of NMS were investigated. Cognitive impairment was defined as a Mini-Mental Status Evaluation score ≤ 23.8 and frontal lobe dysfunction as a Frontal Assessment Battery (FAB) score ≤ 3.48. Multivariable logistic regression was used to identify predictor of frontal lobe dysfunction in 524 PD patients, and a generalized linear model was used for each of the six FAB items. Not only the total FAB scores but also the single FAB items were lower in AP versus PD (p ≤ 0.005). Age (OR = 1.05), cognitive impairment (OR = 9.54), lack of cardiovascular symptoms (OR = 3.25), attention or memory problems (OR = 0.59) and treatment with L: -DOPA (OR = 5.58) were predictors of frontal lobe dysfunction. MMSE was negatively associated with all FAB items (β ≤ -0.16) and age with all FAB items but prehension behavior (β ≤ -0.01). Previous use of L: -DOPA was negatively associated with verbal fluency (β = -0.32) possibly acting as surrogate marker of disease duration. Cognitive impairment is a predictor of frontal lobe dysfunction. Among NMS, lack of attention or memory problems were negatively associated with frontal impairment. Further studies are nonetheless needed to better identify the predictors of frontal impairment in PD patients.