Evolved explainable classifications for lymph node metastases.

A novel evolutionary approach for Explainable Artificial Intelligence is presented: the "Evolved Explanations" model (EvEx). This methodology combines Local Interpretable Model Agnostic Explanations (LIME) with Multi-Objective Genetic Algorithms to allow for automated segmentation parameter tuning in image classification tasks. In this case, the dataset studied is Patch-Camelyon, comprised of patches from pathology whole slide images. A publicly available Convolutional Neural Network (CNN) was trained on this dataset to provide a binary classification for presence/absence of lymph node metastatic tissue. In turn, the classifications are explained by means of evolving segmentations, seeking to optimize three evaluation goals simultaneously. The final explanation is computed as the mean of all explanations generated by Pareto front individuals, evolved by the developed genetic algorithm. To enhance reproducibility and traceability of the explanations, each of them was generated from several different seeds, randomly chosen. The observed results show remarkable agreement between different seeds. Despite the stochastic nature of LIME explanations, regions of high explanation weights proved to have good agreement in the heat maps, as computed by pixel-wise relative standard deviations. The found heat maps coincide with expert medical segmentations, which demonstrates that this methodology can find high quality explanations (according to the evaluation metrics), with the novel advantage of automated parameter fine tuning. These results give additional insight into the inner workings of neural network black box decision making for medical data.

Unsupervised Machine Learning Applied to Seismic Interpretation: Towards an Unsupervised Automated Interpretation Tool.

Seismic interpretation is a fundamental process for hydrocarbon exploration. This activity comprises identifying geological information through the processing and analysis of seismic data represented by different attributes. The interpretation process presents limitations related to its high data volume, own complexity, time consumption, and uncertainties incorporated by the experts' work. Unsupervised machine learning models, by discovering underlying patterns in the data, can represent a novel approach to provide an accurate interpretation without any reference or label, eliminating the human bias. Therefore, in this work, we propose exploring multiple methodologies based on unsupervised learning algorithms to interpret seismic data. Specifically, two strategies considering classical clustering algorithms and image segmentation methods, combined with feature selection, were evaluated to select the best possible approach. Additionally, the resultant groups of the seismic data were associated with groups obtained from well logs of the same area, producing an interpretation with aggregated lithologic information. The resultant seismic groups correctly represented the main seismic facies and correlated adequately with the groups obtained from the well logs data.

Explainable Artificial Intelligence for Bias Detection in COVID CT-Scan Classifiers.

PROBLEM: An application of Explainable Artificial Intelligence Methods for COVID CT-Scan classifiers is presented. MOTIVATION: It is possible that classifiers are using spurious artifacts in dataset images to achieve high performances, and such explainable techniques can help identify this issue. AIM: For this purpose, several approaches were used in tandem, in order to create a complete overview of the classificatios. METHODOLOGY: The techniques used included GradCAM, LIME, RISE, Squaregrid, and direct Gradient approaches (Vanilla, Smooth, Integrated). MAIN RESULTS: Among the deep neural networks architectures evaluated for this image classification task, VGG16 was shown to be most affected by biases towards spurious artifacts, while DenseNet was notably more robust against them. Further impacts: Results further show that small differences in validation accuracies can cause drastic changes in explanation heatmaps for DenseNet architectures, indicating that small changes in validation accuracy may have large impacts on the biases learned by the networks. Notably, it is important to notice that the strong performance metrics achieved by all these networks (Accuracy, F1 score, AUC all in the 80 to 90% range) could give users the erroneous impression that there is no bias. However, the analysis of the explanation heatmaps highlights the bias.

A Portable Fuzzy Driver Drowsiness Estimation System.

The adequate automatic detection of driver fatigue is a very valuable approach for the prevention of traffic accidents. Devices that can determine drowsiness conditions accurately must inherently be portable, adaptable to different vehicles and drivers, and robust to conditions such as illumination changes or visual occlusion. With the advent of a new generation of computationally powerful embedded systems such as the Raspberry Pi, a new category of real-time and low-cost portable drowsiness detection systems could become standard tools. Usually, the proposed solutions using this platform are limited to the definition of thresholds for some defined drowsiness indicator or the application of computationally expensive classification models that limits their use in real-time. In this research, we propose the development of a new portable, low-cost, accurate, and robust drowsiness recognition device. The proposed device combines complementary drowsiness measures derived from a temporal window of eyes (PERCLOS, ECD) and mouth (AOT) states through a fuzzy inference system deployed in a Raspberry Pi with the capability of real-time response. The system provides three degrees of drowsiness (Low-Normal State, Medium-Drowsy State, and High-Severe Drowsiness State), and was assessed in terms of its computational performance and efficiency, resulting in a significant accuracy of 95.5% in state recognition that demonstrates the feasibility of the approach.

Prototype of Robotic Device for Mobility Assistance for the Elderly in Urban Environments.

This study aims to develop a prototype of an autonomous robotic device to assist the locomotion of the elderly in urban environments. Among the achievements presented are the control techniques used for autonomous navigation and the software tools and hardware applied in the prototype. This is an extension of a previous work, in which part of the navigation algorithm was developed and validated in a simulated environment. In this extension, the real prototype is controlled by an algorithm based on fuzzy logic to obtain standalone and more-natural navigation for the user of the device. The robotic device is intended to guide an elderly person in an urban environment autonomously, although it also has a manual navigation mode. Therefore, the device should be able to navigate smoothly without sudden manoeuvres and should respect the locomotion time of the user. Furthermore, because of the proposed environment, the device should be able to navigate in an unknown and unstructured environment. The results reveal that this prototype achieves the proposed objective, demonstrating adequate behaviour for navigation in an unknown environment and fundamental safety characteristics to assist the elderly.

Local Interpretable Model-Agnostic Explanations for Classification of Lymph Node Metastases.

An application of explainable artificial intelligence on medical data is presented. There is an increasing demand in machine learning literature for such explainable models in health-related applications. This work aims to generate explanations on how a Convolutional Neural Network (CNN) detects tumor tissue in patches extracted from histology whole slide images. This is achieved using the "locally-interpretable model-agnostic explanations" methodology. Two publicly-available convolutional neural networks trained on the Patch Camelyon Benchmark are analyzed. Three common segmentation algorithms are compared for superpixel generation, and a fourth simpler parameter-free segmentation algorithm is proposed. The main characteristics of the explanations are discussed, as well as the key patterns identified in true positive predictions. The results are compared to medical annotations and literature and suggest that the CNN predictions follow at least some aspects of human expert knowledge.

Evolving Spiking Neural Networks for Recognition of Aged Voices.

The aging of the voice, known as presbyphonia, is a natural process that can cause great change in vocal quality of the individual. This is a relevant problem to those people who use their voices professionally, and its early identification can help determine a suitable treatment to avoid its progress or even to eliminate the problem. This work focuses on the development of a new model for the identification of aging voices (independently of their chronological age), using as input attributes parameters extracted from the voice and glottal signals. The proposed model, named Quantum binary-real evolving Spiking Neural Network (QbrSNN), is based on spiking neural networks (SNNs), with an unsupervised training algorithm, and a Quantum-Inspired Evolutionary Algorithm that automatically determines the most relevant attributes and the optimal parameters that configure the SNN. The QbrSNN model was evaluated in a database composed of 120 records, containing samples from three groups of speakers. The results obtained indicate that the proposed model provides better accuracy than other approaches, with fewer input attributes.

Analysis and Classification of Voice Pathologies Using Glottal Signal Parameters.

The classification of voice diseases has many applications in health, in diseases treatment, and in the design of new medical equipment for helping doctors in diagnosing pathologies related to the voice. This work uses the parameters of the glottal signal to help the identification of two types of voice disorders related to the pathologies of the vocal folds: nodule and unilateral paralysis. The parameters of the glottal signal are obtained through a known inverse filtering method, and they are used as inputs to an Artificial Neural Network, a Support Vector Machine, and also to a Hidden Markov Model, to obtain the classification, and to compare the results, of the voice signals into three different groups: speakers with nodule in the vocal folds; speakers with unilateral paralysis of the vocal folds; and speakers with normal voices, that is, without nodule or unilateral paralysis present in the vocal folds. The database is composed of 248 voice recordings (signals of vowels production) containing samples corresponding to the three groups mentioned. In this study, a larger database was used for the classification when compared with similar studies, and its classification rate is superior to other studies, reaching 97.2%.

Intelligent multiagent coordination based on reinforcement hierarchical neuro-fuzzy models.

This paper presents the research and development of two hybrid neuro-fuzzy models for the hierarchical coordination of multiple intelligent agents. The main objective of the models is to have multiple agents interact intelligently with each other in complex systems. We developed two new models of coordination for intelligent multiagent systems, which integrates the Reinforcement Learning Hierarchical Neuro-Fuzzy model with two proposed coordination mechanisms: the MultiAgent Reinforcement Learning Hierarchical Neuro-Fuzzy with a market-driven coordination mechanism (MA-RL-HNFP-MD) and the MultiAgent Reinforcement Learning Hierarchical Neuro-Fuzzy with graph coordination (MA-RL-HNFP-CG). In order to evaluate the proposed models and verify the contribution of the proposed coordination mechanisms, two multiagent benchmark applications were developed: the pursuit game and the robot soccer simulation. The results obtained demonstrated that the proposed coordination mechanisms greatly improve the performance of the multiagent system when compared with other strategies.

Classification of vocal aging using parameters extracted from the glottal signal.

This article proposes and evaluates a method to classify vocal aging using artificial neural network (ANN) and support vector machine (SVM), using the parameters extracted from the speech signal as inputs. For each recorded speech, from a corpus of male and female speakers of different ages, the corresponding glottal signal is obtained using an inverse filtering algorithm. The Mel Frequency Cepstrum Coefficients (MFCC) also extracted from the voice signal and the features extracted from the glottal signal are supplied to an ANN and an SVM with a previous selection. The selection is performed by a wrapper approach of the most relevant parameters. Three groups are considered for the aging-voice classification: young (aged 15-30 years), adult (aged 31-60 years), and senior (aged 61-90 years). The results are compared using different possibilities: with only the parameters extracted from the glottal signal, with only the MFCC, and with a combination of both. The results demonstrate that the best classification rate is obtained using the glottal signal features, which is a novel result and the main contribution of this article.