Biarchetype Analysis: Simultaneous Learning of Observations and Features Based on Extremes.

We introduce a novel exploratory technique, termed biarchetype analysis, which extends archetype analysis to simultaneously identify archetypes of both observations and features. This innovative unsupervised machine learning tool aims to represent observations and features through instances of pure types, or biarchetypes, which are easily interpretable as they embody mixtures of observations and features. Furthermore, the observations and features are expressed as mixtures of the biarchetypes, which makes the structure of the data easier to understand. We propose an algorithm to solve biarchetype analysis. Although clustering is not the primary aim of this technique, biarchetype analysis is demonstrated to offer significant advantages over biclustering methods, particularly in terms of interpretability. This is attributed to biarchetypes being extreme instances, in contrast to the centroids produced by biclustering, which inherently enhances human comprehension. The application of biarchetype analysis across various machine learning challenges underscores its value, and both the source code and examples are readily accessible in R and Python at https://github.com/aleixalcacer/JA-BIAA.

Métodos computacionales para estudio de la anemia drepanocítica / Study of sickle cell anemia using computacional methods

El procesamiento de imágenes digitales y la visión por computador son ampliamente utilizados en medicina actualmente y son de gran interés las propuestas de nuevos métodos de análisis automatizado de imágenes digitales o mejorar la eficiencia de los existentes. En este trabajo se desarrollaron métodos nuevos para estudiar computacionalmente a través de imágenes de muestras de sangre la drepanocitosis, dolencia con alta incidencia mundial y en Cuba, sobre todo en la región oriental. Se propusieron nuevos métodos de análisis de formas, obtenidos a partir de resultados clásicos de geometría integral y nuevas propuestas de visión por computador para evaluar trastornos neurofisiológicos asociados a través del estudio de las expresiones faciales del paciente. La validación estadística realizada comprobó la superioridad de estos métodos sobre otros, se determinó que son válidos para ser introducidos en software de apoyo para mejorar la calidad de la atención médica(AU)

Digital image processing and computer vision are frequently used in medicine at present and the proposals of new methods of automatic analysis of digital images or the efficiency improvement of the existing are of great interest. In this work new methods to computationally study sickle cell disease through blood samples images are developed, an illness with high incidence in the world and in Cuba, mainly in the eastern region. New shape analysis methods obtained from classical results of integral geometry and new computer vision proposals for evaluate neuro physiological disorders associated with this illness through the study of the facial expressions of the patient were proposed. The statistical validation realized confirmed the superiority of these methods on previous proposals, which is why they are valid to be introduced in support software to improve the quality of the medical attention(AU)

A new extrinsic sample mean in the shape space with applications to the boundaries of anatomical structures.

Shape analysis is of great importance in many fields of medical imaging and computational biology. In this paper, we consider the shape space as the set of smooth planar immersed curves in R(2) (parameterized curves) and, using the property of being isometric to a classical manifold immersed in a Euclidean space, we introduce a new extrinsic sample mean and a new extrinsic variance for a finite set of shapes, which are not necessarily star shaped. This is a fundamental tool in medical image analysis, for instance, to assess uncertainties that arise in locating anatomical structures such as the prostate and the bladder. We apply it to a dataset consisting of parallel planar axial CT sections of human prostate, in order to study the variability between boundaries that have been manually delineated by several observers.

Numerical quantification of Perkinsus marinus in the American oyster Crassostrea virginica (Gmelin, 1791) (Mollusca: Bivalvia) by modern stereology.

Species of Perkinsus are responsible for high mortalities of bivalve molluscs world-wide. Techniques to accurately estimate parasites in tissues are required to improve understanding of perkinsosis. This study quantifies the number and tissue distribution of Perkinsus marinus in Crassostrea virginica by modern stereology and immunohistochemistry. Mean total number of trophozoites were (mean +/- SE) 11.80 +/- 3.91 million and 11.55 +/- 3.88 million for the optical disector and optical fractionator methods, respectively. The mean empirical error between both stereological approaches was 3.8 +/- 1.0%. Trophozoites were detected intracellularly in the following tissues: intestine (30.1%), Leydig tissue (21.3%), hemocytes (14.9%), digestive gland (11.4%), gills (6.1%), connective tissues (5.7%), gonads (4.1%), palps (2.2%), muscle (1.9%), mantle connective (0.8%), pericardium (0.7%), mantle epithelium (0.1%), and heart (0.1%). The remaining 0.6% were found extracellularly. Percentages of trophozoite stages were (mean +/- SE): large, log-phase trophonts, i.e., signet rings, 97.0 +/- 1.2%; meronts, 2.0 +/- 0.9%; clusters of small, log-phase trophonts, i.e., merozoites, 1.0 +/- 0.5%. Levels of infection in hemocytes and Leydig tissue were representative of total parasite intensity. These techniques are a powerful tool to follow parasite distribution and invasion, and to further explore mechanisms of Perkinsus spp. pathogenesis in bivalves.

New variance expressions for systematic sampling: the filtering approach.

We present a collection of variance models for estimators obtained by geometric systematic sampling with test points, quadrats, and n-boxes in general, on a bounded domain in n-dimensional Euclidean space R(n), n = 1, 2, ... , and for systematic rays and sectors on the circle. The approach adopted - termed the filtering approach - is new and different from the current transitive approach. This report is only preliminary, however, because it includes only variance models in terms of the covariogram of the measurement function. The estimation step is in preparation.