14 October 2012 in Progress in Artificial Intelligence
27 March 2003 in Journal of Animal Breeding and Genetics
Summary: The aim of this work is to highlight the need of monitoring small populations to conserve their genetic variability by using a set of parameters to characterize both the structure of populations and management practices. As a representative example we analyse the pedigree information of the endangered Xalda sheep breed of Asturias. The herdbook of Xalda sheep included a total of 805 animals and 62 herds. The number of founders was 329. Nowadays, there are 562 live animals and 26 active herds. The breed is in risk of losing genetic diversity because of the abusive use of certain individuals as parents. The effective number of founder animals is 81.1. The effective number of founder herds is 9.9. The average value of inbreeding in the whole Xalda population was 1.5%. The average relatedness (AR) coefficient reached 1.8% in the whole pedigree. The genetic representation of the lines of founders is unbalanced. Inbreeding trends and effective size do not provide realistic information concerning the risk of loss of diversity as a result of the shallowness of the genealogical information. We suggest the monitoring of the breed using AR to unbalance the genetic contributions of specific individuals, equalizing the genetic representation of the founders and lines in the population. In addition, AR can suggest the introduction of new, under‐represented animals in herds showing high average AR values relative to the population. Our results can be useful to improve the development of conservation initiatives involving open herdbooks to avoid the risk of loss of genetic diversity caused by incorrect management practices.
4 December 2024 in Computer Methods and Programs in Biomedicine
Ultrasound (US) is a medical imaging modality that plays a crucial role in the early detection of breast cancer. The emergence of numerous deep learning systems has offered promising avenues for the segmentation and classification of breast cancer tumors in US images. However, challenges such as the absence of data standardization, the exclusion of non-tumor images during training, and the narrow view of single-task methodologies have hindered the practical applicability of these systems, often resulting in biased outcomes. This study aims to explore the potential of multi-task systems in enhancing the detection of breast cancer lesions. To address these limitations, our research introduces an end-to-end multi-task framework designed to leverage the inherent correlations between breast cancer lesion classification and segmentation tasks. Additionally, a comprehensive analysis of a widely utilized public breast cancer ultrasound dataset named BUSI was carried out, identifying its irregularities and devising an algorithm tailored for detecting duplicated images in it. Experiments are conducted utilizing the curated dataset to minimize potential biases in outcomes. Our multi-task framework exhibits superior performance in breast cancer respecting single-task approaches, achieving improvements close to 15% in segmentation and classification. Moreover, a comparative analysis against the state-of-the-art reveals statistically significant enhancements across both tasks. The experimental findings underscore the efficacy of multi-task techniques, showcasing better generalization capabilities when considering all image types: benign, malignant, and non-tumor images. Consequently, our methodology represents an advance towards more general architectures with real clinical applications in the breast cancer field.
1 February 2015 in Pattern Recognition
27 November 2016 in Limnology and Oceanography: Methods
In recent decades, the automatic study and analysis of plankton communities using imaging techniques has advanced significantly. The effectiveness of these automated systems appears to have improved, reaching acceptable levels of accuracy. However, plankton ecologists often find that classification systems do not work as well as expected when applied to new samples. This paper proposes a methodology to assess the efficacy of learned models which takes into account the fact that the data distribution (the plankton composition of the sample) can vary between the model building phase and the production phase. As opposed to most validation methods that consider the individual organism as the unit of validation, our approach uses a validation‐by‐sample, which is more appropriate when the objective is to estimate the abundance of different morphological groups. We argue that, in these cases, the base unit to correctly estimate the error is the sample, not the individual. Thus, model assessment processes require groups of samples with sufficient variability in order to provide precise error estimates.
1 October 2017 in Engineering Applications of Artificial Intelligence