Random Forest and Feature Importance Measures for Discriminating the Most Influential Environmental Factors in Predicting Cardiovascular and Respiratory Diseases.

BACKGROUND: Several studies suggest that environmental and climatic factors are linked to the risk of mortality due to cardiovascular and respiratory diseases; however, it is still unclear which are the most influential ones. This study sheds light on the potentiality of a data-driven statistical approach by providing a case study analysis. METHODS: Daily admissions to the emergency room for cardiovascular and respiratory diseases are jointly analyzed with daily environmental and climatic parameter values (temperature, atmospheric pressure, relative humidity, carbon monoxide, ozone, particulate matter, and nitrogen dioxide). The Random Forest (RF) model and feature importance measure (FMI) techniques (permutation feature importance (PFI), Shapley Additive exPlanations (SHAP) feature importance, and the derivative-based importance measure (κALE)) are applied for discriminating the role of each environmental and climatic parameter. Data are pre-processed to remove trend and seasonal behavior using the Seasonal Trend Decomposition (STL) method and preliminary analyzed to avoid redundancy of information. RESULTS: The RF performance is encouraging, being able to predict cardiovascular and respiratory disease admissions with a mean absolute relative error of 0.04 and 0.05 cases per day, respectively. Feature importance measures discriminate parameter behaviors providing importance rankings. Indeed, only three parameters (temperature, atmospheric pressure, and carbon monoxide) were responsible for most of the total prediction accuracy. CONCLUSIONS: Data-driven and statistical tools, like the feature importance measure, are promising for discriminating the role of environmental and climatic factors in predicting the risk related to cardiovascular and respiratory diseases. Our results reveal the potential of employing these tools in public health policy applications for the development of early warning systems that address health risks associated with climate change, and improving disease prevention strategies.

Blue-green roofs as nature-based solutions for urban areas: hydrological performance and climatic index analyses.

The water-regulating capacity of nature-based solutions (NBSs) plays a crucial role in providing a full range of ecosystem services and enhancing the resilience of urban systems. This work focuses on the hydrological performance of a particular NBS, the so-called blue-green roof (BGR). The BGR is designed to collect infiltrated rainfall in a water storage layer beneath the soil to support vegetation maintenance, enhance evapotranspiration and cooling, and minimize runoff and drainage system load. The study aims to evaluate the hydrological performance of the BGR at global and event scale and, for the first time, to model climatic factors (easy to measure using common sensors) that affect its stormwater retention capacity. The data collected over 2 years and 2 months at a 5-min resolution from a pilot study in Central Italy were analysed. Additionally, a new climatic index called AWWP-x (Antecedent Wet Weather Period index) was introduced and calculated. Results show that the BGR has an overall relevant retention rate (67.1%), although the value depends on the rainfall of the observed period. Approximately 50% of the rainfall events did not produce any runoff, and during the dry season, all events were totally absorbed by the BGR. Four climatic variables were identified as significant factors for predicting BGR retention performance (R2 = 0.50). The results suggest that AWWP-130 (number of days to reach 130 mm cumulative precipitation) could be a possible proxy for the BGR stormwater retention rate. In general, this study demonstrates the potential for evaluating, planning, and designing NBSs by considering the annual and interannual climatic variability of the investigated specific location.

Integrating spatially-and temporally-heterogeneous data on river network dynamics using graph theory.

The study of non-perennial streams requires extensive experimental data on the temporal evolution of surface flow presence across different nodes of channel networks. However, the consistency and homogeneity of available datasets is threatened by the empirical burden required to map stream network expansions and contractions. Here, we developed a data-driven, graph-theory framework aimed at representing the hierarchical structuring of channel network dynamics (i.e., the order of node activation/deactivation during network expansion/retraction) through a directed acyclic graph. The method enables the estimation of the configuration of the active portion of the network based on a limited number of observed nodes, and can be utilized to combine datasets with different temporal resolutions and spatial coverage. A proof-of-concept application to a seasonally-dry catchment in central Italy demonstrated the ability of the approach to reduce the empirical effort required for monitoring network dynamics and efficiently extrapolate experimental observations in space and time.

Global-scale massive feature extraction from monthly hydroclimatic time series: Statistical characterizations, spatial patterns and hydrological similarity.

Hydroclimatic time series analysis focuses on a few feature types (e.g., autocorrelations, trends, extremes), which describe a small portion of the entire information content of the observations. Aiming to exploit a larger part of the available information and, thus, to deliver more reliable results (e.g., in hydroclimatic time series clustering contexts), here we approach hydroclimatic time series analysis differently, i.e., by performing massive feature extraction. In this respect, we develop a big data framework for hydroclimatic variable behaviour characterization. This framework relies on approximately 60 diverse features and is completely automatic (in the sense that it does not depend on the hydroclimatic process at hand). We apply the new framework to characterize mean monthly temperature, total monthly precipitation and mean monthly river flow. The applications are conducted at the global scale by exploiting 40-year-long time series originating from over 13 000 stations. We extract interpretable knowledge on seasonality, trends, autocorrelation, long-range dependence and entropy, and on feature types that are met less frequently. We further compare the examined hydroclimatic variable types in terms of this knowledge and, identify patterns related to the spatial variability of the features. For this latter purpose, we also propose and exploit a hydroclimatic time series clustering methodology. This new methodology is based on Breiman's random forests. The descriptive and exploratory insights gained by the global-scale applications prove the usefulness of the adopted feature compilation in hydroclimatic contexts. Moreover, the spatially coherent patterns characterizing the clusters delivered by the new methodology build confidence in its future exploitation. Given this spatial coherence and the scale-independent nature of the delivered feature values (which makes them particularly useful in forecasting and simulation contexts), we believe that this methodology could also be beneficial within regionalization frameworks, in which knowledge on hydrological similarity is exploited in technical and operative terms.

Investigating runoff formation dynamics: field observations at Cape Fear experimental plot.

In this paper, we explore the dynamics of surface runoff formation in an outdoor experimental plot, Cape Fear, by reporting the relationships among rainfall, runoff, and soil moisture for 101 rainfall-runoff events observed in the time span of more than five years (January 2014-March 2019). Cape Fear is a recently developed 7 × 7 m2 experimental plot that combines features from both small scale facilities and catchment-scale experimental hillslopes, thus leveraging observation of major hydrological variables at high temporal and spatial resolution. Despite the small dimension and simplicity of the plot, the relations among hydrological variables are unexpectedly quite spread. Experimental results seem to suggest that Cape Fear runoff response presents an increasing and non-linear relationship with rainfall, with a surface runoff coefficient increasing for higher rainfall. Direct runoff apparently increases with soil moisture, while initial abstraction seems not to be influenced by rainfall and is found to decrease with increasing soil moisture. Observations suggest that complex interactions between soil moisture conditions and rainfall pattern properties modulate the plot response.

Unraveling flow patterns through nonlinear manifold learning.

From climatology to biofluidics, the characterization of complex flows relies on computationally expensive kinematic and kinetic measurements. In addition, such big data are difficult to handle in real time, thereby hampering advancements in the area of flow control and distributed sensing. Here, we propose a novel framework for unsupervised characterization of flow patterns through nonlinear manifold learning. Specifically, we apply the isometric feature mapping (Isomap) to experimental video data of the wake past a circular cylinder from steady to turbulent flows. Without direct velocity measurements, we show that manifold topology is intrinsically related to flow regime and that Isomap global coordinates can unravel salient flow features.

Cardiac complications and diabetes in thalassaemia major: a large historical multicentre study.

The relationship between diabetes mellitus (DM) and cardiac complications has never been systematically studied in thalassaemia major (TM). We evaluated a large retrospective historical cohort of TM to determine whether DM is associated with a higher risk of heart complications. We compared 86 TM patients affected by DM with 709 TM patients without DM consecutively included in the Myocardial Iron Overload in Thalassaemia database where clinical/instrumental data are recorded from birth to the first cardiovascular magnetic resonance (CMR) exam. All of the cardiac events considered were developed after the DM diagnosis. In DM patients versus non-DM patients we found a significantly higher frequency of cardiac complications (46.5% vs. 16.9%, P < 0.0001), heart failure (HF) (30.2% vs. 11.7%, P < 0.0001), hyperkinetic arrhythmias (18.6% vs. 5.5%, P < 0.0001) and myocardial fibrosis assessed by late gadolinium enhancement (29.9% vs. 18.4%, P = 0.008). TM patients with DM had a significantly higher risk of cardiac complications [odds ratio (OR) 2.84, P < 0.0001], HF (OR 2.32, P = 0.003), hyperkinetic arrhythmias (OR 2.21, P = 0.023) and myocardial fibrosis (OR 1.91, P = 0.021), also adjusting for the absence of myocardial iron overload assessed by T2* CMR and for the covariates (age and/or endocrine co-morbidity). In conclusion, DM significantly increases the risk for cardiac complications, HF, hyperkinetic arrhythmias and myocardial fibrosis in TM patients.

Assessment of fluorescent particles for surface flow analysis.

In this paper, a systematic performance assessment of the measurement system for surface flow analysis developed by our group in (Tauro et al., Sensors, 2010) is presented. The system is based on the detection of buoyant fluorescent microspheres through a low-cost apparatus, which incorporates light sources to elicit fluorescence response and a digital camera to identify the particles' transit. Experiments are conducted using green fluorescent particles and further tests are executed to evaluate the system performance for red and orange particles varying in emission wavelength, degree of biocompatibility, and cost. The influence of the following parameters on surface flow sensing using fluorescent beads is investigated: (i) distance of the light sources from the water surface, (ii) presence of an ad-hoc filter tuned at the particle emission wavelength, (iii) camera resolution and frame rate, (iv) flow regime, and (v) ambient light. Experimental results are used to inform implementation guidelines for surface flow analysis in natural environments.

A new anatomic approach for robot-assisted laparoscopic prostatectomy: a feasibility study for completely intrafascial surgery.

Robot-assisted laparoscopic prostatectomy (RALP) has been disseminated widely, changing the knowledge of surgical anatomy of the prostate. The aim of our study is to demonstrate the feasibility of a new, purely intrafascial approach. The Bocciardi approach for RALP passes through the Douglas space, following a completely intrafascial plane without any dissection of the anterior compartment, which contains neurovascular bundles, Aphrodite's veil, endopelvic fascia, the Santorini plexus, pubourethral ligaments, and all of the structures thought to play a role in maintenance of continence and potency. In this case series, we present our first five patients undergoing the Bocciardi approach for RALP. We report the results of our technique in three patients following two unsuccessful attempts. No perioperative major complication was recorded. Pathologic stage was pT2c in two patients and pT2a in one patient, with no positive surgical margin. The day after removing the catheter, two of the three patients reported use of a single, small safety pad, and one patient was discharged without any pad. One patient reported an erection the day after removing the catheter. The anatomic rationale for better results compared with traditional RALP is strong, but well-designed studies are needed to evaluate the advantages of our technique.

Characterization of buoyant fluorescent particles for field observations of water flows.

In this paper, the feasibility of off-the-shelf buoyant fluorescent microspheres as particle tracers in turbid water flows is investigated. Microspheres' fluorescence intensity is experimentally measured and detected in placid aqueous suspensions of increasing concentrations of clay to simulate typical conditions occurring in natural drainage networks. Experiments are conducted in a broad range of clay concentrations and particle immersion depths by using photoconductive cells and image-based sensing technologies. Results obtained with both methodologies exhibit comparable trends and show that the considered particles are fairly detectable in critically turbid water flows. Further information on performance and integration of the studied microspheres in low-cost measurement instrumentation for field observations is obtained through experiments conducted in a custom built miniature water channel. This experimental characterization provides a first assessment of the feasibility of commercially available buoyant fluorescent beads in the analysis of high turbidity surface water flows. The proposed technology may serve as a minimally invasive sensing system for hazardous events, such as pollutant diffusion in natural streams and flash flooding due to extreme rainfall.