Alberto De Santis | Università degli Studi "La Sapienza" di Roma (original) (raw)

Papers by Alberto De Santis

Frontiers in Nutrition

This study aims to assess the carbon footprint associated with vegan, vegetarian, and omnivorous ... more This study aims to assess the carbon footprint associated with vegan, vegetarian, and omnivorous menus for primary school lunches in Italy. For this purpose, healthy and acceptable menus with minimal greenhouse gas emissions have been designed by a binary linear programming model. The results show that the adoption of a specific diet may help in reducing the carbon footprint of menus, but it is the optimal selection of dishes that ultimately makes the difference. Interestingly enough, the optimal choice of dishes and the restriction of meat consumption in omnivorous menus can lead up to a 40% emission reduction compared to the current school lunch menu of the municipality of Rome. Moreover, the optimal choice of dishes in vegan menus provides the menu with the lowest carbon footprint among all kinds of diets.

Nutrients, 2022

The non-professional sport environment is a grey zone not as widely assessed as that of elite ath... more The non-professional sport environment is a grey zone not as widely assessed as that of elite athletes. The purpose of this research was to investigate the dietary supplementation habits and the nutrition knowledge on sport (NKS) in a sample of gym users. The level of adequacy of NKS was set at ≥60% of correct answers. Almost half (46.4%) of respondents stated they used food supplements, in particular multivitamins (31.0%), amino acid pills (29.5%), minerals (29.1%), and protein powders (28.7%). Supplements were used to increase muscle mass (36.9%) and to repair muscle (35.1%). Gym trainers were the preferred source of information on the use of supplements, especially in males (84%). The NKS correct response rate was 57.1% and the proportion of respondents with a sufficient level of NKS was 47.3%. The prevalence of correct answers was highest in males (61.5%) and for respondents with the highest educational attainment levels (44.5% and 53%). This study demonstrated that non-professi...

An efficient eye tracking procedure is presented providing a non-invasive method for real time de... more An efficient eye tracking procedure is presented providing a non-invasive method for real time detection of a subject pupils in a sequence of frames captured by low cost equipment. The procedure can be easily adapted to any application relying on eye tracking. The eye pupil identification is performed by a hierarchical optimal segmentation procedure: a contextual picture zoning yielding the eyes position, and a further binarization extracting the pupils coordinates. No eye movement model is required to predict the future eyes position to restrict the image searching area, since the procedure first step is fast enough to obtain a frame to frame eyes position update.

The pupil morphological characteristics are of great interest for non invasive early diagnosis of... more The pupil morphological characteristics are of great interest for non invasive early diagnosis of the central nervous system response to environmental stimuli of different nature. Their evaluation in subjects suffering some typical diseases such as diabetes, Alzheimer disease, schizophrenia, drug and alcohol addiction is of concern. In this paper geometrical pupil features such as area, centroid coordinates, eccentricity, major and minor axes lengths are estimated by a procedure based on an image segmentation algorithm. It exploits the level set formulation of the related variational problem. A discrete set up of this problem is proposed: an arbitrary initial curve is evolved towards the unique optimal segmentation boundary by a difference equation. Numerical tests are performed on real pupillometry data taken in different illumination conditions showing a high degree of robustness of the shape parameters estimates

IET Systems Biology, 2007

Gene expression is to a large extent controlled at the level of mRNA accumulation. Genes whose pr... more Gene expression is to a large extent controlled at the level of mRNA accumulation. Genes whose products function together are likely under a common regulatory system (e.g. signal transduction pathways, sets of regulatory proteins) such that they are expressed in a coordinated manner. This property has been frequently used in the analysis of genome-wide expression data, as the experimental observation that a group of genes is co-expressed frequently implies that the genes share a common regulatory mechanism. The authors have investigated the situation in which dissimilarity in gene-expression time profiles may still result from the presence of the same regulatory signal, as in the case of common transcription factors. To this aim, a dynamic model that takes into account the effect of specific mRNA degradation on the shape of gene-expression time series has been developed, and the concept of 'dynamically co-regulated' genes has accordingly been introduced as the goodness-of-fit to such a model (called dynamic R 2). The statistical analysis of dynamic R 2 over a number of different experimental data sets and organisms shows that the presence of dynamically co-regulated genes is by far more significant than that expected from the randomised data. Furthermore, as an example of the usefulness of the proposed method, genome-wide yeast measurements such as cell-cycle time series and transcription factors targets data, were used to prove that dynamic co-regulation is statistically related to the presence of common transcription factor(s). This latter property is very useful when trying to infer computational indications of co-regulation for not-yet annotated genes that do not display a co-expression pattern.

Image Analysis and Processing - ICIAP 2017, 2017

Light Microscopy (LM) represents the method by which pathologists study histological sections; th... more Light Microscopy (LM) represents the method by which pathologists study histological sections; the observations by LM can be considered the gold standard for making diagnosis and for its diagnostic accuracy. The classes that can be defined through the observation of LM images of the liver are: normal, steatosis, fibrosis, cirrhosis and hepatocarcinoma (HCC). Normally, a pathologist has to examine by LM many histological sections to perform a complete and accurate diagnosis. For this reason, an automatic system for the analysis of LM images of the liver would be particularly useful. Goal of this paper is to propose an automatic multi-stage procedure to classify the normal tissue, and the pathologic ones from human liver microphotographs. Due to the articulated nature of the examined images, the analysis will first assess if steatosis is present, by using objects analysis, and then determine whether the image belongs to a normal tissue or to one of the other pathologic ones, by using a machine learning based technique. To this aim some texture features are calculated, and the Principal Component Analysis is applied to derive the best representation of the data. Four binary Support Vector Machines classifiers are trained, one for each kind the four classes of liver conditions to be identified. Experimental results show the classification capability of the proposed system, with promising theoretical and experimental basis for developing a fully automatic decision support system.

Journal of Optimization Theory and Applications, 2022

In this paper, we focus on the linear functionals defining an approximate version of the gradient... more In this paper, we focus on the linear functionals defining an approximate version of the gradient of a function. These functionals are often used when dealing with optimization problems where the computation of the gradient of the objective function is costly or the objective function values are affected by some noise. These functionals have been recently considered to estimate the gradient of the objective function by the expected value of the function variations in the space of directions. The expected value is then approximated by a sample average over a proper (random) choice of sample directions in the domain of integration. In this way, the approximation error is characterized by statistical properties of the sample average estimate, typically its variance. Therefore, while useful and attractive bounds for the error variance can be expressed in terms of the number of function evaluations, nothing can be said on the error of a single experiment that could be quite large. This w...

Annals of Operations Research, 2022

Accurate modeling of the patient flow within an Emergency Department (ED) is required by all stud... more Accurate modeling of the patient flow within an Emergency Department (ED) is required by all studies dealing with the increasing and well-known problem of overcrowding. Since Discrete Event Simulation (DES) models are often adopted with the aim of assessing solutions for reducing the impact of this worldwide phenomenon, an accurate estimation of the service time of the ED processes is necessary to guarantee the reliability of the results. However, simulation models concerning EDs are frequently affected by data quality problems, thus requiring a proper estimation of the missing parameters. In this paper, a simulation-based optimization approach is used to estimate the incomplete data in the patient flow within an ED by adopting a model calibration procedure. The objective function of the resulting minimization problem represents the deviation between simulation output and real data, while the constraints ensure that the response of the simulation is sufficiently accurate according to the precision required. Data from a real case study related to a big ED in Italy is used to test the effectiveness of the proposed approach. The experimental results show that the model calibration allows recovering the missing parameters, thus leading to an accurate DES model.

Training a large multilayer neural network can present many difficulties due to the large number ... more Training a large multilayer neural network can present many difficulties due to the large number of useless stationary points. These points usually attract the minimization algorithm used during the training phase, which therefore results inefficient. Extending some results proposed in literature for shallow networks, we propose the mathematical characterization of a class of such stationary points that arise in deep neural networks training. Availing such a description, we are able to define an incremental training algorithm that avoids getting stuck in the region of attraction of these undesirable stationary points.

Nutrients, 2021

School is considered a privileged environment for health education and school feeding represents ... more School is considered a privileged environment for health education and school feeding represents an opportunity for promoting sustainable foods to young generations. The objective of this paper is to demonstrate that is possible to select, from existing school menus, recipes that combine healthy foods with low environmental impact. A national sample of Italian school menus was collected and a total number of 194 recipes were included on a database containing 70 first courses, 83 s courses, 39 side dishes, 1 portion of fruit, and 1 portion of bread. A mathematical model was conceived to combine nutritional adequacy and acceptability criteria while minimizing GHGs emissions. The result is a four-week menu characterized by large vegetable components that were used not only as side dishes but also as ingredients in the first and second courses. Legumes and pasta are often included, and white meat is selected instead of red meat. The findings presented in this paper demonstrated that it ...

Flexible Services and Manufacturing Journal, 2021

Modeling the arrival process to an Emergency Department (ED) is the first step of all studies dea... more Modeling the arrival process to an Emergency Department (ED) is the first step of all studies dealing with the patient flow within the ED. Many of them focus on the increasing phenomenon of ED overcrowding, which is afflicting hospitals all over the world. Since Discrete Event Simulation models are often adopted to assess solutions for reducing the impact of this problem, proper nonstationary processes are taken into account to reproduce time–dependent arrivals. Accordingly, an accurate estimation of the unknown arrival rate is required to guarantee the reliability of results. In this work, an integer nonlinear black–box optimization problem is solved to determine the best piecewise constant approximation of the time-varying arrival rate function, by finding the optimal partition of the 24 h into a suitable number of not equally spaced intervals. The black-box constraints of the optimization problem make the feasible solutions satisfy proper statistical hypotheses; these ensure the ...

Frontiers in Nutrition, 2020

Background: Food consumption is one of the most important drivers of the relation between human w... more Background: Food consumption is one of the most important drivers of the relation between human well-being and Earth's ecosystems. The current production level is difficult to sustain without compromising environmental integrity or public health. This calls for a decisive change in food consumption patterns in order to improve nutrition quality while respecting biodiversity and ecosystems. This change will produce some effect only if it is also culturally acceptable, accessible, economically fair and affordable. The design of food plans is traditionally carried out using mathematical optimization models, such as linear programming. This method has proved to be successful in providing nutritionally adequate diets while minimizing their economic and environmental impact. Nevertheless, cultural habits as well as attractiveness and variety of meals is very difficult to deal with, and no fully satisfactory way to include these issues in linear programming has been found. Objective: The aim of this paper is to move from traditional linear programming to a new programming methodology in order to cope also with acceptability in the design of meal plans. Method: Binary integer linear programming is the new modeling paradigm. In the proposed model, meal plans consist of providing the sequence and composition of daily meals over a given period of time and each meal can be composed using dishes from a given set. Therefore, instead of defining just a level of consumption of food groups or food items, the proposed model provides a realistic menu. To cope with sustainability, the energy and nutritional content of each dish is calculated together with its price and environmental impact. Furthermore, acceptability can be explicitly taken into account in a very natural way, that is bounding the daily, weekly, or total repetitions of single dishes and of dishes in the same food groups. Results: The paper reviews three successful studies with increasing complexity considering lunch plans for schools and full-board menus for nursing homes. The case studies show a great reduction of the environmental impact of the meal plans while ensuring an adequate nutritional intake, affordable prices and most importantly the plans are varied and culturally acceptable.

Optimization Letters, 2020

A trading strategy simply consists in a procedure which defines conditions for buying or selling ... more A trading strategy simply consists in a procedure which defines conditions for buying or selling a security on a financial market. These decisions rely on the values of some indicators that, in turn, affect the tuning of the strategy parameters. The choice of these parameters significantly affects the performance of the trading strategy. In this work, an optimization procedure is proposed to find the best parameter values of a chosen trading strategy by using the security price values over a given time period; these parameter values are then applied to trade on the next incoming security price sequence. The idea is that the market is sufficiently stable so that a trading strategy that is optimally tuned in a given period still performs well in the successive period. The proposed optimization approach tries to determine the parameter values which maximize the profit in a trading session, therefore the objective function is not defined in closed form but through a procedure that computes the profit obtained in a sequence of transactions. For this reason the proposed optimization procedures are based on a black-box optimization approach. Namely they do not require the assumption that the objective function is continuously differentiable and do not use any first order information. Numerical results obtained in a real case seem to be encouraging.

Journal of Cleaner Production, 2019

Sustainability of food consumption concerns both environmental and economic issues. In fact, the ... more Sustainability of food consumption concerns both environmental and economic issues. In fact, the United Nations Food and Agricultural Organization defines as sustainable diets those that are protective and respectful of ecosystems, culturally acceptable, economically affordable, besides ensuring an adequate and healthy nutrition. In this paper, a systematic methodology to plan menus complying with nutritional and health issues, close to current eating habits, affordable and with low greenhouse gas emissions is presented. The methodology relies on a multi-objective optimization model with binary variables. The objectives, that is the greenhouse gas emissions needed to serve the menu and its price, are conflicting and therefore a trade-off has to be established by means of the set of Pareto solutions. Any such a solution delivers a menu in term of the recipes composing each daily meal. The application of the presented methodology to the case of cycle menus for nursing homes is investigated. The case study shows that the menu's environmental impact is generally in inverse proportion to its price. Nevertheless, it is possible to obtain a menu with a significantly reduced environmental impact at an affordable extra cost.

International Journal of Tomography Simulation, Jan 20, 2010

Diagnosis by medical images implies the expert ability of recognizing patterns of interest in ter... more Diagnosis by medical images implies the expert ability of recognizing patterns of interest in terms of some features like gray (or color) level intensity, shape attributes, texture. The image segmentation algorithms constitute a valid support in the analysis of medical images by providing reliable computer tools able to separate the objects of interest from the background. There is a great deal of segmentation algorithms depending on the mathematical model adopted for the information to be retrieved from data. They span from very simple and fast threshold procedures, to local signal processing like edge detection, to sophisticated ones based on global optimization methods. This work describes a region growing algorithm that falls within the last framework; it is based on a novel image model. It is formulated in the discrete domain to deal directly with the image data without approximation schemes required by the formulation in the continuum domain, typical of the variational methods. The segmentation procedure is efficient and reliable, allowing a hierarchical processing also in term of the signal components. It can easily take into account a wide range of situations occurring in the medical environment, going from the analysis of angiographies to the analysis of CT scan images of human body organs.

Ductile irons offer a wide range of mechanical properties at a lower cost than the older malleabl... more Ductile irons offer a wide range of mechanical properties at a lower cost than the older malleable iron. These properties mainly depend on the shape characteristics of the metal matrix microstructure and on the graphite elements morphology; these geometrical features are currently evaluated by the experts visual inspection. This work provides an automatic procedure for a reliable estimation of standard parameters of the material microstructure morphology based on a novel image segmentation technique. The procedure has been validated versus standard segmentation techniques, and successfully tested on specimens of different kinds of ductile irons of a typical production.

International Geoscience and Remote Sensing Symposium (IGARSS), 2011

... Silvia Canale, Alberto De Santis, Daniela Iacoviello, Fiora Pirri, Simone Sagratella Sapienza... more ... Silvia Canale, Alberto De Santis, Daniela Iacoviello, Fiora Pirri, Simone Sagratella Sapienza Universit`a di Roma, Dipartimento di Informatica e ... [11] Elena Angiati, Giorgio Boni, Laura Candela, Fabio Castelli, Silvana G. Dellepiane, Fabio Delogu, Fabio Pintus, Roberto Rudari ...

Signal, Image and Video Processing, 2007

Models and algorithms in image processing are usually defined in the continuum and then applied t... more Models and algorithms in image processing are usually defined in the continuum and then applied to discrete data, that is the signal samples over a lattice. In particular, the set up in the continuum of the segmentation problem allows a fine formulation basically through either a variational approach or a moving interfaces approach. In any case, the image segmentation is