Potential and limitations of machine meta-learning (ensemble) methods for predicting COVID-19 mortality in a large inhospital Brazilian dataset.

The majority of early prediction scores and methods to predict COVID-19 mortality are bound by methodological flaws and technological limitations (e.g., the use of a single prediction model). Our aim is to provide a thorough comparative study that tackles those methodological issues, considering multiple techniques to build mortality prediction models, including modern machine learning (neural) algorithms and traditional statistical techniques, as well as meta-learning (ensemble) approaches. This study used a dataset from a multicenter cohort of 10,897 adult Brazilian COVID-19 patients, admitted from March/2020 to November/2021, including patients [median age 60 (interquartile range 48-71), 46% women]. We also proposed new original population-based meta-features that have not been devised in the literature. Stacking has shown to achieve the best results reported in the literature for the death prediction task, improving over previous state-of-the-art by more than 46% in Recall for predicting death, with AUROC 0.826 and MacroF1 of 65.4%. The newly proposed meta-features were highly discriminative of death, but fell short in producing large improvements in final prediction performance, demonstrating that we are possibly on the limits of the prediction capabilities that can be achieved with the current set of ML techniques and (meta-)features. Finally, we investigated how the trained models perform on different hospitals, showing that there are indeed large differences in classifier performance between different hospitals, further making the case that errors are produced by factors that cannot be modeled with the current predictors.

A associação entre os transtornos de linguagem receptivo e expressivo e os transtornos de aprendizagem / The association between receptive and expressive language disorders and learning disorders

O aprendizado e a linguagem são processos neuronais complexos, os quais possuem íntima ligação entre si. Diante disso, o transtorno de linguagem (TL) também está fortemente correlacionado aos transtornos de aprendizagem (TA), uma associação inicialmente observada no projeto de extensão "Construindo Histórias...", da PUC Minas, e confirmada pela literatura. Foi realizada uma busca de publicações dos últimos 15 anos, nas bases PubMed, LILACS, Up to Date e MedLine, utilizando os descritores Language Development Disorders, Child Languages, Language Development, Learning e Learning Disorders e suas traduções em português. Foram selecionadas 18 publicações e também o foi utilizado o livro Transtornos da aprendizagem. Os TL por definição são mudanças nos padrões normais do aprendizado da linguagem e podem ser classificados em expressivo e receptivo. Como resultado desse quadro, a criança apresenta dificuldades em compreender e de se expressar, o que está diretamente relacionado ao processo de aprendizagem e pode resultar em um TA. O diagnóstico desses transtornos deve ser realizado de maneira individualizada e observando a interação de fatores genéticos, fisiológicos, socioculturais e escolares. Conclui-se que os transtornos de linguagem e de aprendizado não afetam apenas a vida acadêmica, mas também a vida social, comportamental e psicológica e, por isso, é de extrema importância que ocorra um foco maior na detecção precoce desses transtornos, para que, assim, a criança receba um tratamento adequado e, consequentemente, tenha um prognóstico melhor. (AU)

Learning and language are complex neuronal processes, which have an intimate connection with each other. Therefore, language development disorder (LDD) or language impairment is also strongly correlated with learning disorders (LD), an association initially observed in the extension project "Constructing stories" at PUC Minas and confirmed by the literature. A search for publications from the last 15 years was carried out in the following outlets: PubMed, LILACS, Up to Date and MedLine using the descriptors Language Development Disorders, child languages, Language Development, Learning and Learning Disorders and their Portuguese version. Eighteen publications were selected and the book Transtornos da aprendizagem, 2016 was also used. LDD by definition are changes in normal language learning patterns and can be classified as expressive and receptive. Because of this situation, children have difficulties in understanding and expressing themselves, this is directly related to the learning process, which can result in an LD. The diagnosis of these disorders must be done individually and observing the interaction of genetic, physiological, sociocultural and school factors. It is worth noting that language and learning disorders not only affect academic life, but also social, behavioral and psychological life, and therefore it is extremely important that there is a greater focus on the early detection of these disorders, thus, children receive an appropriate treatment and, consequently, have a better prognosis. (AU)