André Queiroz de Andrade

A new training algorithm using artificial neural networks to classify gender-specific dynamic gait patterns

The aim of this study was to present a new training algorithm using artificial neural networks called multi-objective least absolute shrinkage and selection operator (MOBJ-LASSO) applied to the classification of dynamic gait patterns. The movement pattern is identified by 20 characteristics from the three components of the ground reaction force which are used as input information for the neural networks in gender-specific gait classification. The classification performance between MOBJ-LASSO (97.4%) and multi-objective algorithm (MOBJ) (97.1%) is similar, but the MOBJ-LASSO algorithm achieved more improved results than the MOBJ because it is able to eliminate the inputs and automatically select the parameters of the neural network. Thus, it is an effective tool for data mining using neural networks. From 20 inputs used for training, MOBJ-LASSO selected the first and second peaks of the vertical force and the force peak in the antero-posterior direction as the variables that classify the gait patterns of the different genders.

Process attributes in bio-ontologies

BackgroundBiomedical processes can provide essential information about the (mal-) functioning of an organism and are thus frequently represented in biomedical terminologies and ontologies, including the GO Biological Process branch. These processes often need to be described and categorised in terms of their attributes, such as rates or regularities. The adequate representation of such process attributes has been a contentious issue in bio-ontologies recently; and domain ontologies have correspondingly developed ad hoc workarounds that compromise interoperability and logical consistency.ResultsWe present a design pattern for the representation of process attributes that is compatible with upper ontology frameworks such as BFO and BioTop. Our solution rests on two key tenets: firstly, that many of the sorts of process attributes which are biomedically interesting can be characterised by the ways that repeated parts of such processes constitute, in combination, an overall process; secondly, that entities for which a full logical definition can be assigned do not need to be treated as primitive within a formal ontology framework. We apply this approach to the challenge of modelling and automatically classifying examples of normal and abnormal rates and patterns of heart beating processes, and discuss the expressivity required in the underlying ontology representation language. We provide full definitions for process attributes at increasing levels of domain complexity.ConclusionsWe show that a logical definition of process attributes is feasible, though limited by the expressivity of DL languages so that the creation of primitives is still necessary. This finding may endorse current formal upper-ontology frameworks as a way of ensuring consistency, interoperability and clarity.

A realism-based analysis of the OpenEHR entry model

We describe a proposal of improvement for clinical models of the OpenEHR standard according to the realism-based assumptions. In order to reach such proposal, we analyze OpenEHR entry model and realism-based ontologies created specifically for Medicine. Thus, we check our approach in a test bed of real medical records. Finally, we suggest a new taxonomy with the aim to improve the entry model, and offer conclusions from our underway research.

The impact of web-based digital health interventions on weight loss and lifestyle habits changes in overweight and obese adults: a systematic review and meta-analysis (Preprint)

Background Obesity is a highly prevalent condition with important health implications. Face-to-face interventions to treat obesity demand a large number of human resources and time, generating a great burden to individuals and health system. In this context, the internet is an attractive tool for delivering weight loss programs due to anonymity, 24-hour-accessibility, scalability, and reachability associated with Web-based programs. Objective We aimed to investigate the effectiveness of Web-based digital health interventions, excluding hybrid interventions and non-Web-based technologies such as text messaging, short message service, in comparison to nontechnology active or inactive (wait list) interventions on weight loss and lifestyle habit changes in individuals with overweight and obesity. Methods We searched PubMed or Medline, SciELO, Lilacs, PsychNet, and Web of Science up to July 2018, as well as references of previous reviews for randomized trials that compared Web-based digital health interventions to offline interventions. Anthropometric changes such as weight, body mass index (BMI), waist, and body fat and lifestyle habit changes in adults with overweight and obesity were the outcomes of interest. Random effects meta-analysis and meta-regression were performed for mean differences (MDs) in weight. We rated the risk of bias for each study and the quality of evidence across studies using the Grades of Recommendation, Assessment, Development, and Evaluation approach. Results Among the 4071 articles retrieved, 11 were included. Weight (MD −0.77 kg, 95% CI −2.16 to 0.62; 1497 participants; moderate certainty evidence) and BMI (MD −0.12 kg/m2; 95% CI −0.64 to 0.41; 1244 participants; moderate certainty evidence) changes were not different between Web-based and offline interventions. Compared to offline interventions, digital interventions led to a greater short-term (<6 months follow-up) weight loss (MD −2.13 kg, 95% CI −2.71 to −1.55; 393 participants; high certainty evidence), but not in the long-term (MD −0.17 kg, 95% CI −2.10 to 1.76; 1104 participants; moderate certainty evidence). Meta-analysis was not possible for lifestyle habit changes. High risk of attrition bias was identified in 5 studies. For weight and BMI outcomes, the certainty of evidence was moderate mainly due to high heterogeneity, which was mainly attributable to control group differences across studies (R2=79%). Conclusions Web-based digital interventions led to greater short-term but not long-term weight loss than offline interventions in overweight and obese adults. Heterogeneity was high across studies, and high attrition rates suggested that engagement is a major issue in Web-based interventions.