Predicting effluent quality parameters for wastewater treatment plant: A machine learning-based methodology.

Wastewater Treatment Plants (WWTPs) present complex biochemical processes of high variability and difficult prediction. This study presents an innovative approach using Machine Learning (ML) models to predict wastewater quality parameters. In particular, the models are applied to datasets from both a simulated wastewater treatment plant (WWTP), using DHI WEST software (WEST WWTP), and a real-world WWTP database from Santa Catarina Brewery AMBEV, located in Lages/SC - Brazil (AMBEV WWTP). A distinctive aspect is the evaluation of predictive performance in continuous data scenarios and the impact of changes in WWTP operations on predictive model performance, including changes in plant layout. For both plants, three different scenarios were addressed, and the quality of predictions by random forest (RF), support vector machine (SVM), and multilayer perceptron (MLP) models were evaluated. The prediction quality by the MLP model reached an R2 of 0.72 for TN prediction in the WEST WWTP output, and the RF model better adapted to the real data of the AMBEV WWTP, despite the significant discrepancy observed between the real and the predicted data. Techniques such as Partial Dependence Plots (PDP) and Permutation Importance (PI) were used to assess the importance of features, particularly in the simulated WEST tool scenario, showing a strong correlation of prediction results with influent parameters related to nitrogen content. The results of this study highlight the importance of collecting and storing high-quality data and the need for information on changes in WWTP operation for predictive model performance. These contributions advance the understanding of predictive modeling for wastewater quality and provide valuable insights for future practice in wastewater treatment.

A Trust-Based Methodology to Evaluate Deep Learning Models for Automatic Diagnosis of Ocular Toxoplasmosis from Fundus Images.

In the automatic diagnosis of ocular toxoplasmosis (OT), Deep Learning (DL) has arisen as a powerful and promising approach for diagnosis. However, despite the good performance of the models, decision rules should be interpretable to elicit trust from the medical community. Therefore, the development of an evaluation methodology to assess DL models based on interpretability methods is a challenging task that is necessary to extend the use of AI among clinicians. In this work, we propose a novel methodology to quantify the similarity between the decision rules used by a DL model and an ophthalmologist, based on the assumption that doctors are more likely to trust a prediction that was based on decision rules they can understand. Given an eye fundus image with OT, the proposed methodology compares the segmentation mask of OT lesions labeled by an ophthalmologist with the attribution matrix produced by interpretability methods. Furthermore, an open dataset that includes the eye fundus images and the segmentation masks is shared with the community. The proposal was tested on three different DL architectures. The results suggest that complex models tend to perform worse in terms of likelihood to be trusted while achieving better results in sensitivity and specificity.

Driftage: a multi-agent system framework for concept drift detection.

BACKGROUND: The amount of data and behavior changes in society happens at a swift pace in this interconnected world. Consequently, machine learning algorithms lose accuracy because they do not know these new patterns. This change in the data pattern is known as concept drift. There exist many approaches for dealing with these drifts. Usually, these methods are costly to implement because they require (i) knowledge of drift detection algorithms, (ii) software engineering strategies, and (iii) continuous maintenance concerning new drifts. RESULTS: This article proposes to create Driftage: a new framework using multi-agent systems to simplify the implementation of concept drift detectors considerably and divide concept drift detection responsibilities between agents, enhancing explainability of each part of drift detection. As a case study, we illustrate our strategy using a muscle activity monitor of electromyography. We show a reduction in the number of false-positive drifts detected, improving detection interpretability, and enabling concept drift detectors' interactivity with other knowledge bases. CONCLUSION: We conclude that using Driftage, arises a new paradigm to implement concept drift algorithms with multi-agent architecture that contributes to split drift detection responsability, algorithms interpretability and more dynamic algorithms adaptation.

The London Chest Activity of Daily Living scale cut-off point to discriminate functional status in patients with chronic obstructive pulmonary disease.

OBJECTIVE: To determine the cut-off point for the London Chest Activity of Daily Living scale in order to better discriminate functional status. Secondarily, to determine which of the scores (total or %total) is better associated with clinical outcomes of a pulmonary rehabilitation program. METHODS: Sixty-one patients with chronic obstructive pulmonary disease performed the following tests: spirometry; Chronic Obstructive Pulmonary Disease Assessment Test; Saint George's Respiratory Questionnaire; modified Medical Research Council, the body-mass index, airflow obstruction, dyspnea, and exercise capacity index; six-minute walk test; physical activity in daily life assessment and London Chest Activity of Daily Living scale. Thirty-eight patients were evaluated pre- and post-pulmonary rehabilitation . The cut-off point was determined using the receiver operating characteristic curve with six-minute walk test (cut-off point: 82%pred), modified Medical Research Council (cut-off point: 2), level of physical (in)activity (cut-off point: 80min per day in physical activity ≥3 metabolic equivalent of task) and presence/absence of severe physical inactivity (cut-off point: 4580 steps per day) as anchors. RESULTS: A cut-off point found for all anchors was 28%: modified Medical Research Council [sensitivity=83%; specificity=72%; area under the curve=0.80]; level of physical (in)activity [sensitivity=65%; specificity=59%; area under the curve=0.67] and classification of severe physical inactivity [sensitivity=70%; specificity=62%; area under the curve=0.70]. The patients who scored ≤28% in %total score of London Chest Activity of Daily Living had lower modified Medical Research Council , Chronic Obstructive Pulmonary Disease Assessment Test, Saint George's Respiratory Questionnaire, body-mass index, airflow obstruction, dyspnea and exercise capacity index and sitting time than who scored >28%, and higher forced expiratory volume in the first second, time in physical activity ≥3 metabolic equivalent of task, steps per day and six-minute walk distance. The %total score of London Chest Activity of Daily Living correlated better with clinical outcomes than the total score. CONCLUSIONS: The cut-off point of 28% is sensitive and specific to distinguish the functional status in patients with chronic obstructive pulmonary disease. The %total score of the London Chest Activity of Daily Living reflects better outcomes of chronic obstructive pulmonary disease when compared to total score.

Classification of diabetes-related retinal diseases using a deep learning approach in optical coherence tomography.

BACKGROUND AND OBJECTIVES: Spectral Domain Optical Coherence Tomography (SD-OCT) is a volumetric imaging technique that allows measuring patterns between layers such as small amounts of fluid. Since 2012, automatic medical image analysis performance has steadily increased through the use of deep learning models that automatically learn relevant features for specific tasks, instead of designing visual features manually. Nevertheless, providing insights and interpretation of the predictions made by the model is still a challenge. This paper describes a deep learning model able to detect medically interpretable information in relevant images from a volume to classify diabetes-related retinal diseases. METHODS: This article presents a new deep learning model, OCT-NET, which is a customized convolutional neural network for processing scans extracted from optical coherence tomography volumes. OCT-NET is applied to the classification of three conditions seen in SD-OCT volumes. Additionally, the proposed model includes a feedback stage that highlights the areas of the scans to support the interpretation of the results. This information is potentially useful for a medical specialist while assessing the prediction produced by the model. RESULTS: The proposed model was tested on the public SERI-CUHK and A2A SD-OCT data sets containing healthy, diabetic retinopathy, diabetic macular edema and age-related macular degeneration. The experimental evaluation shows that the proposed method outperforms conventional convolutional deep learning models from the state of the art reported on the SERI+CUHK and A2A SD-OCT data sets with a precision of 93% and an area under the ROC curve (AUC) of 0.99 respectively. CONCLUSIONS: The proposed method is able to classify the three studied retinal diseases with high accuracy. One advantage of the method is its ability to produce interpretable clinical information in the form of highlighting the regions of the image that most contribute to the classifier decision.

Temporalidade e Interpretabilidade na Análise de Narrativas / Temporality and Interpretability: Analysis of Narratives

Resumo Destacamos a importância de manter uma consistência vertical nos estudos que versam sobre narrativas para a compreensão da psique humana. Nessa direção, os princípios da temporalidade e sua íntima relação com a interpretabilidade, propostos por autores clássicos como Bruner, Polkinghorne, Ricoeur e Sarbin, são tomados como guias na manutenção dessa consistência. Para ilustrá-la, analisamos duas narrativas versando sobre temas diversos: experiência escolar e supervisão acadêmica em psicoterapia. Na primeira, os processos de segmentação e encadeamento do tempo engajado no mundo da ação constituem elementos-chave de análise na significação da experiência escolar. Na segunda, o interdiscurso entre a estagiária e a supervisora se interliga e cria significado pela própria inter-relação entre tempos narrativos diversos.

Abstract This paper highlights the importance of maintaining vertical consistence in narrative studies investigating human psyche. The principles of temporality and interpretability, proposed by classic authors as Bruner, Polkinghorne, Ricoeur e Sarbin, are taken as guidelines to maintain this consistence. To illustrate this idea, two narratives dealing with different themes are analyzed: one referring to school experience and the other to academic supervision of psychotherapy. For the first one the processes of segmentation and entanglement of the engaging time in the world of action compose the key-elements to analyze the meaning of school experience. For the second, different narrating times - one of the trainee and the other of the supervisor - create an interdiscourse from which meaning emerges as a consequence of the interrelationships of diverse narrating times.

Dual energy imaging and intracycle motion correction for CT coronary angiography in patients with intermediate to high likelihood of coronary artery disease.

We explored whether intracycle motion correction algorithms (MCAs) might be applicable to dual energy computed tomography coronary angiography in patients with intermediate to high likelihood of coronary artery disease. MCA reconstructions were associated with higher interpretability rates (96.7% vs. 87.9%, P<.001), image quality scores (4.12 ± 0.9 vs. 3.76 ± 1.0; P<.0001), and diagnostic performance [area under the curve of 0.95 (95% confidence interval [CI] 0.92-0.97) vs. 0.89 (95% CI 0.86-0.92); P<.0001] compared to conventional reconstructions. In conclusion, application of intracycle MCA reconstructions to dual energy computed tomography acquisitions was feasible and resulted in significantly higher image quality scores, interpretability, and diagnostic performance.

Uma análise das evidências de validade brasileiras do PMK / An analysis of the evidences of Brazilian validities about the PMK

Tendo em vista a proposta e a quantidade de medidas provenientes do PMK, foram colocados como objeto de análise os estudos de evidência de validade fornecidos pelo Manual. De fato, o interesse é buscar os fundamentos da interpretabilidade dos resultados de um instrumento com uma pretensão dessa magnitude. Como resultado, encontrou-se que as análises fatoriais realizadas não fornecem subsídios para a estrutura proposta para a interpretação das distintas medidas do PMK. Ao lado disso, os estudos com grupo-critério apresentados é, quando muito, incipiente e não fornece evidências de interpretabilidade para quase nenhum dos denominados seis fatores. As evidências apresentadas são confusas, há incoerências entre correlação e análise de variância e a interpretabilidade do teste não fica demonstrada pelas pesquisas.

Considering the proposal and amount of measures concerning with the PMK, the analysis of the studies of evidence of validity supplied by the Manual were placed under investigation. In fact, the interest is to search for foundations of the interpretability of the results of a test with a pretentiousness of this magnitude. As a result, the factorial analyzes do not support the proposed structure for the interpretation of the different measures of the PMK. Besides, the presented studies with criterion group were as much as incipient and do not furnish evidences of interpretability for almost none of the named six factors. The showed evidences are confusing, there are incoherencies between the concept of correlation and variance analysis and the interpretability of the test is not demonstrated by the researches.

Uma análise das evidências de validade brasileiras do PMK / An analysis of the evidences of Brazilian validities about the PMK

Tendo em vista a proposta e a quantidade de medidas provenientes do PMK, foram colocados como objeto de análise os estudos de evidência de validade fornecidos pelo Manual. De fato, o interesse é buscar os fundamentos da interpretabilidade dos resultados de um instrumento com uma pretensão dessa magnitude. Como resultado, encontrou-se que as análises fatoriais realizadas não fornecem subsídios para a estrutura proposta para a interpretação das distintas medidas do PMK. Ao lado disso, os estudos com grupo-critério apresentados é, quando muito, incipiente e não fornece evidências de interpretabilidade para quase nenhum dos denominados seis fatores. As evidências apresentadas são confusas, há incoerências entre correlação e análise de variância e a interpretabilidade do teste não fica demonstrada pelas pesquisas.(AU)

Considering the proposal and amount of measures concerning with the PMK, the analysis of the studies of evidence of validity supplied by the Manual were placed under investigation. In fact, the interest is to search for foundations of the interpretability of the results of a test with a pretentiousness of this magnitude. As a result, the factorial analyzes do not support the proposed structure for the interpretation of the different measures of the PMK. Besides, the presented studies with criterion group were as much as incipient and do not furnish evidences of interpretability for almost none of the named six factors. The showed evidences are confusing, there are incoherencies between the concept of correlation and variance analysis and the interpretability of the test is not demonstrated by the researches.(AU)