Nivio Ziviani

Dehua Chen, Amir Jalilifard, Adriano Veloso et al.

A pharmacological effect of a drug on cells, organs and systems refers to the specific biochemical interaction produced by a drug substance, which is called its mechanism of action. Drug repositioning (or drug repurposing) is a fundamental problem for the identification of new opportunities for the use of already approved or failed drugs. In this paper, we present a method based on a multi-relation unsupervised graph embedding model that learns latent representations for drugs and diseases so that the distance between these representations reveals repositioning opportunities. Once representations for drugs and diseases are obtained we learn the likelihood of new links (that is, new indications) between drugs and diseases. Known drug indications are used for learning a model that predicts potential indications. Compared with existing unsupervised graph embedding methods our method shows superior prediction performance in terms of area under the ROC curve, and we present examples of repositioning opportunities found on recent biomedical literature that were also predicted by our method.

Eduardo Nigri, Nivio Ziviani, Fabio Cappabianco et al.

Deep Convolutional Neural Networks (CNNs) are becoming prominent models for semi-automated diagnosis of Alzheimer's Disease (AD) using brain Magnetic Resonance Imaging (MRI). Although being highly accurate, deep CNN models lack transparency and interpretability, precluding adequate clinical reasoning and not complying with most current regulatory demands. One popular choice for explaining deep image models is occluding regions of the image to isolate their influence on the prediction. However, existing methods for occluding patches of brain scans generate images outside the distribution to which the model was trained for, thus leading to unreliable explanations. In this paper, we propose an alternative explanation method that is specifically designed for the brain scan task. Our method, which we refer to as Swap Test, produces heatmaps that depict the areas of the brain that are most indicative of AD, providing interpretability for the model's decisions in a format understandable to clinicians. Experimental results using an axiomatic evaluation show that the proposed method is more suitable for explaining the diagnosis of AD using MRI while the opposite trend was observed when using a typical occlusion test. Therefore, we believe our method may address the inherent black-box nature of deep neural networks that are capable of diagnosing AD.

Gianlucca Zuin, Adriano Veloso, João Cândido Portinari et al.

In this paper, we deal with the problem of automatic tag recommendation for painting artworks. Diachronic descriptions containing deviations on the vocabulary used to describe each painting usually occur when the work is done by many experts over time. The objective of this work is to provide a framework that produces a more accurate and homogeneous set of tags for each painting in a large collection. To validate our method we build a model based on a weakly-supervised neural network for over $5{,}300$ paintings with hand-labeled descriptions made by experts for the paintings of the Brazilian painter Candido Portinari. This work takes place with the Portinari Project which started in 1979 intending to recover and catalog the paintings of the Brazilian painter. The Portinari paintings at that time were in private collections and museums spread around the world and thus inaccessible to the public. The descriptions of each painting were made by a large number of collaborators over 40 years as the paintings were recovered and these diachronic descriptions caused deviations on the vocabulary used to describe each painting. Our proposed framework consists of (i) a neural network that receives as input the image of each painting and uses frequent itemsets as possible tags, and (ii) a clustering step in which we group related tags based on the output of the pre-trained classifiers.

Tiago Alves, Alberto Laender, Adriano Veloso et al.

Early recognition of risky trajectories during an Intensive Care Unit (ICU) stay is one of the key steps towards improving patient survival. Learning trajectories from physiological signals continuously measured during an ICU stay requires learning time-series features that are robust and discriminative across diverse patient populations. Patients within different ICU populations (referred here as domains) vary by age, conditions and interventions. Thus, mortality prediction models using patient data from a particular ICU population may perform suboptimally in other populations because the features used to train such models have different distributions across the groups. In this paper, we explore domain adaptation strategies in order to learn mortality prediction models that extract and transfer complex temporal features from multivariate time-series ICU data. Features are extracted in a way that the state of the patient in a certain time depends on the previous state. This enables dynamic predictions and creates a mortality risk space that describes the risk of a patient at a particular time. Experiments based on cross-ICU populations reveals that our model outperforms all considered baselines. Gains in terms of AUC range from 4% to 8% for early predictions when compared with a recent state-of-the-art representative for ICU mortality prediction. In particular, models for the Cardiac ICU population achieve AUC numbers as high as 0.88, showing excellent clinical utility for early mortality prediction. Finally, we present an explanation of factors contributing to the possible ICU outcomes, so that our models can be used to complement clinical reasoning.

Tiago Pimentel, Marianne Monteiro, Adriano Veloso et al.

Anomalies are intuitively easy for human experts to understand, but they are hard to define mathematically. Therefore, in order to have performance guarantees in unsupervised anomaly detection, priors need to be assumed on what the anomalies are. By contrast, active learning provides the necessary priors through appropriate expert feedback. Thus, in this work we present an active learning method that can be built upon existing deep learning solutions for unsupervised anomaly detection, so that outliers can be separated from normal data effectively. We introduce a new layer that can be easily attached to any deep learning model designed for unsupervised anomaly detection to transform it into an active method. We report results on both synthetic and real anomaly detection datasets, using multi-layer perceptrons and autoencoder architectures empowered with the proposed active layer, and we discuss their performance on finding clustered and low density anomalies.

Sabir Ribas, Alberto Ueda, Rodrygo L. T. Santos et al.

Static rankings of papers play a key role in the academic search setting. Many features are commonly used in the literature to produce such rankings, some examples are citation-based metrics, distinct applications of PageRank, among others. More recently, learning to rank techniques have been successfully applied to combine sets of features producing effective results. In this work, we propose the metric S-RCR, which is a simplified version of a metric called Relative Citation Ratio --- both based on the idea of a co-citation network. When compared to the classical version, our simplification S-RCR leads to improved efficiency with a reasonable effectiveness. We use S-RCR to rank over 120 million papers in the Microsoft Academic Graph dataset. By using this single feature, which has no parameters and does not need to be tuned, our team was able to reach the 3rd position in the first phase of the WSDM Cup 2016.

Sabir Ribas, Berthier Ribeiro-Neto, Edmundo de Souza e Silva et al.

To manage the problem of having a higher demand for resources than availability of funds, research funding agencies usually rank the major research groups in their area of knowledge. This ranking relies on a careful analysis of the research groups in terms of their size, number of PhDs graduated, research results and their impact, among other variables. While research results are not the only variable to consider, they are frequently given special attention because of the notoriety they confer to the researchers and the programs they are affiliated with. In here we introduce a new metric for quantifying publication output, called R-Score for reputation-based score, which can be used in support to the ranking of research groups or programs. The novelty is that the metric depends solely on the listings of the publications of the members of a group, with no dependency on citation counts. R-Score has some interesting properties: (a) it does not require access to the contents of published material, (b) it can be curated to produce highly accurate results, and (c) it can be naturally used to compare publication output of research groups (e.g., graduate programs) inside a same country, geographical area, or across the world. An experiment comparing the publication output of 25 CS graduate programs from Brazil suggests that R-Score can be quite useful for providing early insights into the publication patterns of the various research groups one wants to compare.