Adult Skeletal Age-at-Death Estimation through Deep Random Neural Networks: A New Method and Its Computational Analysis.

Age-at-death assessment is a crucial step in the identification process of skeletal human remains. Nonetheless, in adult individuals this task is particularly difficult to achieve with reasonable accuracy due to high variability in the senescence processes. To improve the accuracy of age-at-estimation, in this work we propose a new method based on a multifactorial macroscopic analysis and deep random neural network models. A sample of 500 identified skeletons was used to establish a reference dataset (age-at-death: 19-101 years old, 250 males and 250 females). A total of 64 skeletal traits are covered in the proposed macroscopic technique. Age-at-death estimation is tackled from a function approximation perspective and a regression approach is used to infer both point and prediction interval estimates. Based on cross-validation and computational experiments, our results demonstrate that age estimation from skeletal remains can be accurately (~6 years mean absolute error) inferred across the entire adult age span and informative estimates and prediction intervals can be obtained for the elderly population. A novel software tool, DRNNAGE, was made available to the community.

Restoring anatomical complexity of a left ventricle wall as a step toward bioengineering a human heart with human induced pluripotent stem cell-derived cardiac cells.

The heart is a highly complex, multicellular solid organ with energy-demanding processes that require a dense vascular network, extensive cell-cell interactions, and extracellular matrix (ECM)-mediated crosstalk among heterogeneous cell populations. Here, we describe the regeneration of left ventricular (LV) wall using decellularized whole rabbit heart scaffolds recellularized exclusively with human induced pluripotent stem cell-derived endothelial cells, cardiomyocytes, and other cardiac cell types. Cells were sequentially delivered to the scaffold using an optimized endothelial cell:cardiomyocyte media. Macroscopic assessment after 60 days showed that the LV wall of recellularized hearts was anatomically restored to full thickness from base to apex and endocardium to epicardium. Histologic analysis of the recellularized LV wall revealed a heterogeneous pool of cardiac cells containing aligned cardiac troponin T-positive cells in close contact with ECM; vessels varied from large artery-like, surrounded by smooth muscle actin+ cells, to capillary-like. Vessel patency was demonstrated after perfusion of recellularized hearts transplanted into the femoral artery bed of a pig. The construct exhibited visible beating and responded to chronotropic drug administration. These results demonstrate the ability to tissue engineer a vascularized, full-thickness LV wall with an unparalleled level of microanatomical organization and multicellular composition, using decellularized ECM and human cardiomyocytes, endothelial cells, and other cardiac cell types. STATEMENT OF SIGNIFICANCE: Decellularized extracellular matrix (ECM) is a bioactive template for tissue engineering, but recellularizing acellular whole heart scaffolds is challenging. Here, we successfully revascularized and repopulated a large, full-thickness portion of a ventricle using human induced pluripotent stem cell-derived endothelial and cardiac cells. At 60 days, histologic studies showed that the microanatomical organization and cellular composition of this region was similar to that of the native heart. The recellularized heart showed visible beating and responded appropriately to heartbeat-altering drugs. Vessels surrounded by smooth muscle cells and endothelial cells supported blood flow through the vessels of a recellularized heart that was surgically connected to a pig femoral artery. These findings move this approach closer to the possibility of clinical translation.

Characterization of perfusion decellularized whole animal body, isolated organs, and multi-organ systems for tissue engineering applications.

To expand the application of perfusion decellularization beyond isolated single organs, we used the native vasculature of adult and neonatal rats to systemically decellularize the organs of a whole animal in situ. Acellular scaffolds were generated from kidney, liver, lower limb, heart-lung system, and a whole animal body, demonstrating that perfusion decellularization technology is applicable to any perfusable tissue, independent of age. Biochemical and histological analyses demonstrated that organs and organ systems (heart-lung pair and lower limb) were successfully decellularized, retaining their extracellular matrix (ECM) structure and organ-specific composition, as evidenced by differences in organ-specific scaffold stiffness. Altogether, we demonstrated that organs, organ systems and whole animal bodies can be perfusion decellularized while retaining ECM components and biomechanics.

Determining the maximum length of logical rules in a classifier and visual comparison of results.

Supervised learning problems can be faced by using a wide variety of approaches supported in machine learning. In recent years there has been an increasing interest in using the evolutionary computation paradigm as the classifier search method, helping the technique of applied machine learning. In this context, the knowledge representation in form of logical rules has been one of the most accepted machine learning approaches, because of its level of expressiveness. This paper proposes an evolutionary framework for rule-based classifier induction and is based on the idea of sequential covering. We introduce genetic programming as the search method for classification-rules. From this approach, we have given results on subjects as maximum rule length, number of rules needed in a classifier and the rule intersection problem. The experiments developed on benchmark clinical data resulted in a methodology to follow in the learning method evaluation. Moreover, the results achieved compared to other methods have shown that our proposal can be very useful in data analysis and classification coming from the medical domain.•The method is based on genetic programming techniques to find rules holding each class in a dataset.•The method is approached to solve the problem of rule intersection from different classes.•The method states the maximum length of a rule to generalize.

A Comparative Study of Bio-Inspired Odour Source Localisation Strategies from the State-Action Perspective.

Locating odour sources with robots is an interesting problem with many important real-world applications. In the past years, the robotics community has adapted several bio-inspired strategies to search for odour sources in a variety of environments. This work studies and compares some of the most common strategies from a behavioural perspective with the aim of knowing: (1) how different are the behaviours exhibited by the strategies for the same perceptual state; and (2) which are the most consensual actions for each perceptual state in each environment. The first step of this analysis consists of clustering the perceptual states, and building histograms of the actions taken for each cluster. In case of (1), a histogram is made for each strategy separately, whereas for (2), a single histogram containing the actions of all strategies is produced for each cluster of states. Finally, statistical hypotheses tests are used to find the statistically significant differences between the behaviours of the strategies in each state. The data used for performing this study was gathered from a purpose-built simulator which accurately simulates the real-world phenomena of odour dispersion and air flow, whilst being sufficiently fast to be employed in learning and evolutionary robotics experiments. This paper also proposes an xml-inspired structure for the generated datasets that are used to store the perceptual information of the robots over the course of the simulations. These datasets may be used in learning experiments to estimate the quality of a candidate solution or for measuring its novelty.

Multidimensional knapsack problem: a fitness landscape analysis.

Fitness landscape analysis techniques are used to better understand the influence of genetic representations and associated variation operators when solving a combinatorial optimization problem. Five representations are investigated for the multidimensional knapsack problem. Common mutation operators, such as bit-flip mutation, are employed to generate fitness landscapes. Measures such as fitness distance correlation and autocorrelation are applied to examine the landscapes associated with the tested genetic encodings. Furthermore, additional experiments are made to observe the effects of adding heuristics and local optimization to the representations. Encodings with a strong heuristic bias are more efficient, and the addition of local optimization techniques further enhances their performance.

Evolution of a multi-agent system in a cyclical environment.

The synchronisation phenomena in biological systems is a current and recurring subject of scientific study. This topic, namely that of circadian clocks, served as inspiration to develop an agent-based simulation that serves the main purpose of being a proof-of-concept of the model used in the BitBang framework, that implements a modern autonomous agent model. Despite having been extensively studied, circadian clocks still have much to be investigated. Rather than wanting to learn more about the internals of this biological process, we look to study the emergence of this kind of adaptation to a daily cycle. To that end we implemented a world with a day/night cycle, and analyse the ways the agents adapt to that cycle. The results show the evolution of the agents' ability to gather food. If we look at the total number of agents over the course of an experiment, we can pinpoint the time when reproductive technology emerges. We also show that the agents adapt to the daily cycle. This circadian rhythm can be shown by analysing the variation on the agents metabolic rate, which is affected by the variation of their movement patterns. In the experiments conducted we can observe that the metabolic rate of the agents varies according to the daily cycle.

Nueva formula matema tica para el calculo de la proteinuria de 24 horas en ninos / new mathematical formula for the calculation of 24-hour proteinuria in children

Se elaboro una formula matematica para la cuantificacion de la excrecion urinaria de proteinas en 24 horas en ninos y adolescentes, derivada de la relacion proteinuria creatinuria (RPC). Se procesaron 104 muestras de orina de 24 horas, las cuales se dividieron en 2 grupos de acuerdo con la excrecion diaria de creatinina. La proteinuria de 24 horas se determino en cada grupo por el metodo tradicional y mediante la formula propuesta. Se obtuvo una excelente correlacion (r = 0,994) y una elevada concordancia en el grupo con recolecciones adecuadas de orina, no asi en el grupo con recolecciones inadecuadas (r = 0,896). Al comparar las RPC en 30 muestras de orina obtenidas en 3 diferentes periodos, con la RPC de 24 horas correspondiente se obtuvo una elevada correlacion (r > 0,99); no obstante la que guardo mayor identidad fue la del periodo comprendido entre las 7:00 a.m. y 12:00 m

Nueva fórmula matemá tica para el cálculo de la proteinuria de 24 horas en niños / new mathematical formula for the calculation of 24-hour proteinuria in children

Se elaboró una fórmula matemática para la cuantificación de la excreción urinaria de proteinas en 24 horas en niños y adolescentes, derivada de la relación proteinuria creatinuria (RPC). Se procesaron 104 muestras de orina de 24 horas, las cuales se dividieron en 2 grupos de acuerdo con la excreción diaria de creatinina. La proteinuria de 24 horas se determinó en cada grupo por el método tradicional y mediante la fórmula propuesta. Se obtuvo una excelente correlación (r = 0,994) y una elevada concordancia en el grupo con recolecciones adecuadas de orina, no así en el grupo con recolecciones inadecuadas (r = 0,896). Al comparar las RPC en 30 muestras de orina obtenidas en 3 diferentes períodos, con la RPC de 24 horas correspondiente se obtuvo una elevada correlación (r > 0,99); no obstante la que guardó mayor identidad fue la del período comprendido entre las 7:00 a.m. y 12:00 m(AU)