This study investigates the application of diffusion models in medical image classification (DiffMIC), focusing on skin and oral lesions. Utilizing the datasets PAD-UFES-20 for skin cancer and P-NDB-UFES for oral cancer, the diffusion model demonstrated competitive performance compared to state-of-the-art deep learning models like Convolutional Neural Networks (CNNs) and Transformers. Specifically, for the PAD-UFES-20 dataset, the model achieved a balanced accuracy of 0.6457 for six-class classification and 0.8357 for binary classification (cancer vs. non-cancer). For the P-NDB-UFES dataset, it attained a balanced accuracy of 0.9050. These results suggest that diffusion models are viable models for classifying medical images of skin and oral lesions. In addition, we investigate the robustness of the model trained on PAD-UFES-20 for skin cancer but tested on the clinical images of the HIBA dataset.

In this paper, we propose an approach to support the diagnosis of urinary tract diseases, with a focus on bladder cancer, using SHAP (SHapley Additive exPlanations)-based feature selection to enhance the transparency and effectiveness of predictive models. Six binary classification scenarios were developed to distinguish bladder cancer from other urological and oncological conditions. The algorithms XGBoost, LightGBM, and CatBoost were employed, with hyperparameter optimization performed using Optuna and class balancing with the SMOTE technique. The selection of predictive variables was guided by importance values through SHAP-based feature selection while maintaining or even improving performance metrics such as balanced accuracy, precision, and specificity. The use of explainability techniques (SHAP) for feature selection proved to be an effective approach. The proposed methodology may contribute to the development of more transparent, reliable, and efficient clinical decision support systems, optimizing screening and early diagnosis of urinary tract diseases.

Cancer is one of the leading causes of death worldwide, and early diagnosis of the disease is one of the most important factors in reducing mortality or increasing lifespan. Traditionally, healthcare experts use various sources of information to determine a diagnosis, often including some form of imaging along with clinical and demographic data. In this work, we propose a method to improve fusion of medical images and multi-field complementary data for classification in small datasets using deep learning models. To achieve that, we introduce a novel complementary data extraction block using hyperbolic space feature enhancement by Poincar & eacute; transformation and a mechanism for multi-field feature interactions. We evaluate it using datasets for the diagnosis of skin cancer (PAD-UFES-20) and oral cavity cancer (NDB-UFES). The experimental results show statistically significant improvement in performance on PAD-UFES-20 dataset of the proposed model over baseline. The SingleS Poincar & eacute; Conv-Concat with MetaBlock fusion model using PiT image backbone achieved performance of in balanced accuracy metric. For NBD-UFES dataset, the ViT architecture with SingleDS Poincar & eacute; Conv-Concat using MetaBlock fusion () obtained the best mean among all the results evaluated. The results indicate the feasibility of the proposed approach. The source code is public available at GitHub at https://github.com/lmlima/Hyperbolic-space-feature-enhancement-and-interaction-mechanism-for-classification.

OBJECTIVE Creation of a database based on artificial intelligence (AI) to diagnose malignant and potentially malignant disorders. STUDY DESIGN This is a retrospective and descriptive study conducted in a reference center for oral diagnosis in Espírito Santo. Clinical, sociodemographic, pathological, and histopathological images of cases of oral leukoplakia and squamous cell carcinoma (OSCC) diagnosed in this institution in the period 2010-2021 were used. All data were stored in a self-system and used for the training of a computer-aided diagnostic system (CADS). RESULTS A total of 69 patients were included, generating 237 images, 146 of leukoplakia and 91 of OSCC. The validation of the results attested to the feasibility of CADS in distinguishing between leukoplakia and OSCC, and the presence or absence of dysplasia in the samples through tests of precision, balanced accuracy, and area under the ROC curve. CONCLUSION The creation of the database allowed the development of CADS. Thus, the use of AI in head and neck pathologies can be translated soon, into more effective strategies to assist in the diagnosis of patients affected by these lesions in addition to assisting in the secondary prevention of OSCC. Thanks for financial support to the research: UFES and CNPq.

Variational autoencoders (VAEs) are generative models known for learning compact and continuous latent representations of data. While they have proven effective in various applications, using latent representations for classification tasks presents challenges. Typically, a straightforward approach involves concatenating the mean and variance vectors and inputting them into a shallow neural network. In this paper, we introduce a novel approach for variational autoencoders, named VAE-GNA, which integrates Gaussian neurons into the latent space along with attention mechanisms. These neurons directly process mean and variance values through a suitable modified sigmoid function, not only improving classification, but also optimizing the training of the VAE in extracting features, in synergy with the classification network. Additionally, we investigate both additive and multiplicative attention mechanisms to enhance the model’s capabilities. We applied the proposed method to automatic cancer detection using near-infrared (NIR) spectral data, showing that the experimental results of VAE-GNA surpass established baselines for spectral datasets. The results obtained indicate the feasibility and effectiveness of our approach.

Skin lesions are classified in benign or malignant. Among the malignant, melanoma is a very aggressive cancer and the major cause of deaths. So, early diagnosis of skin cancer is very desired. In the last few years, there is a growing interest in computer aided diagnostic (CAD) using most image and clinical data of the lesion. These sources of information present limitations due to their inability to provide information of the molecular structure of the lesion. NIR spectroscopy may provide an alternative source of information to automated CAD of skin lesions. The most commonly used techniques and classification algorithms used in spectroscopy are Principal Component Analysis (PCA), Partial Least Squares - Discriminant Analysis (PLS-DA), and Support Vector Machines (SVM). Nonetheless, there is a growing interest in applying the modern techniques of machine and deep learning (MDL) to spectroscopy. One of the main limitations to apply MDL to spectroscopy is the lack of public datasets. Since there is no public dataset of NIR spectral data to skin lesions, as far as we know, an effort has been made and a new dataset named NIR-SC-UFES, has been collected, annotated and analyzed generating the gold-standard for classification of NIR spectral data to skin cancer. Next, the machine learning algorithms XGBoost, CatBoost, LightGBM, 1D-convolutional neural network (1D-CNN) were investigated to classify cancer and non-cancer skin lesions. Experimental results indicate the best performance obtained by LightGBM with pre-processing using standard normal variate (SNV), feature extraction providing values of 0.839 for balanced accuracy, 0.851 for recall, 0.852 for precision, and 0.850 for F-score. The obtained results indicate the first steps in CAD of skin lesions aiming the automated triage of patients with skin lesions in vivo using NIR spectral data.

The diagnosis of oral squamous cell carcinoma or oral leukoplakia and the presence or absence of oral epithelial dysplasia is carried by pathologists. In recent years, deep learning has been presented to deal with the automated detection of various pathologies using digital images. One of the main limitations to applying deep learning to histopathological images is the lack of public datasets. In order to fill this gap, a joint effort has been made and a new dataset of histopathological images of oral cancer, named P-NDB-UFES, has been collected, annotated, and analyzed by oral pathologists generating the gold-standard for classification. This dataset is composed of 3763 images of patches of histopathological images with oral squamous cell carcinoma (29%), dysplasia (51.29%), and without dysplasia (18.79%). Next, convolutional neural network (CNN), transformers neural networks, and few-shot learning approaches (i.e., Siamese, Triplet, and ProtoNet) were investigated to classify oral squamous cell carcinoma and the presence or absence of oral dysplasia. Experimental results indicate that the CNNs and transformers models, in general, have no statistically significant difference, with only DenseNet-121 outperforming transformers at a balanced accuracy (BCC) of 91.91%, recall, and precision of 91.93%. Few-shot learning methods were inferior when compared to other methods, with different configurations having statistical differences among themselves. For ProtoNet architectures, the usage of hyperbolic space showed to have a similar behavior to Euclidean distance, however, these results were heavily influenced by the optimizer used.

Organizations are made up of teams, and team formation is a crucial attribution for decision-makers. Some studies indicate the increased correlation between team arrangement and performance. Moreover, some social aspects of team members’ interaction, such as cohesion and diversity have a strong relationship with team performance and innovative practices. In this paper we present a sociometric metric to measure cohesion, and propose the Krippendorff’s disagreement metric to compute intra-group diversity by measuring the disagreement among observations. So, a novel bi-objective formulation is presented by maximizing cohesion and disagreement. In order to generate an approximate Pareto front, four multi-objective evolutionary algorithms (MOEA) are tested NSGA-II, U-NSGA-III, SMS-EMOA and SPEA2 have been applied to a benchmark, consisting of a team formation problem with 7 instances with the same structure but different numbers of individuals and groups. On average the four algorithms provided a very close performance generating good Pareto approximations indicating the feasibility for this class of problem.

This paper aims to investigate the mathematical problem-solving capabilities of Chat Generative Pre-Trained Transformer (ChatGPT) in case of Bayesian reasoning. The study draws inspiration from Zhu & Gigerenzer's research in 2006, which posed the question: Can children reason the Bayesian way? In the pursuit of answering this question, a set of 10 Bayesian reasoning problems were presented. The results of their work revealed that children's ability to reason effectively using Bayesian principles is contingent upon a well-structured information representation. In this paper, we present the same set of 10 Bayesian reasoning problems to ChatGPT. Remarkably, the results demonstrate that ChatGPT provides the right solutions to all problems.

Data streams are important sources of information nowadays, and with the popularization of mobile devices and sensor systems that collect all kinds of data, more and more information is generated at an ever increasing speed. This growth in data supply poses some problems for traditional machine learning algorithms. Tasks such as data classification, regression, or data clustering presents some limitations regarding very large datasets, data streams, or variations in data. The high cost of labeling instances for training classification algorithms makes it difficult to use fully supervised algorithms. Unbalanced datasets tend to cause algorithms to ignore one or more classes. Moreover, concept drifts in data streams require algorithms to be retrained from time to time. In order to tackle such problems mentioned, a semi-supervised and online algorithm based on Extreme Learning Machine (ELM) called SSOE-FP-ELM is proposed and detailed. Experimental results show that the proposed algorithm outperform others in the literature in accuracy, generalization ability and concept drift detection and recovery, showing suitable alternatives for data streams classification.

The gold standard for the diagnosis of oral cancer is the microscopic analysis of specimens removed preferentially through incisional biopsies of oral mucosa with a clinically detected suspicious lesion. This dataset contains captured histopathological images of oral squamous cell carcinoma and leukoplakia. A total of 237 images were captured, 89 leukoplakia with dysplasia images, 57 leukoplakia without dysplasia images and 91 carcinoma images. The images were captured with an optical light microscope, using 10x and 40x objectives, attached to a microscope camera and visualized through a software. The images were saved in PNG format at 2048 × 1536 size pixels and they refer to hematoxylin-eosin stained histopathologic slides from biopsies performed between 2010 and 2021 in patients managed at the Oral Diagnosis project (NDB) of the Federal University of Espírito Santo (UFES). Oral leukoplakias were represented by samples with and without epithelial dysplasia. Since the diagnosis considers socio-demographic data (gender, age and skin color) as well as clinical data (tobacco use, alcohol consumption, sun exposure, fundamental lesion, type of biopsy, lesion color, lesion surface and lesion diagnosis), this information was also collected. So, our aim by releasing this dataset NDB-UFES is to provide a new dataset to be used by researchers in Artificial Intelligence (machine and deep learning) to develop tools to assist clinicians and pathologists in the automated diagnosis of oral potentially malignant disorders and oral squamous cell carcinoma.

Machine and deep learning algorithms have increasingly been applied to solve problems in various areas of knowledge. Among these areas, Chemometrics has been benefited from the application of these algorithms in spectral data analysis. Commonly, algorithms such as Support Vector Machines and Partial Least Squares are applied to spectral datasets to perform classification and regression tasks. In this paper, we present a 1D convolutional neural networks (1D-CNN) to evaluate the effectiveness on spectral data obtained from spectroscopy. In most cases, the spectrum signals are noisy and present overlap among classes. Firstly, we perform extensive experiments including 1D-CNN compared to machine learning algorithms and standard algorithms used in Chemometrics on spectral data classification for the most known datasets available in the literature. Next, spectral samples of the SARS-COV2 virus, which causes the COVID-19, have recently been collected via spectroscopy was used as a case study. Experimental results indicate superior performance of 1D-CNN over machine learning algorithms and standard algorithms, obtaining an average accuracy of 96.5%, specificity of 98%, and sensitivity of 94%. The promissing obtained results indicate the feasibility to use 1D-CNN in automated systems to diagnose COVID-19 and other viral diseases in the future.

Background Oral cancer is one of the most common types of cancer in men causing mortality if not diagnosed early. In recent years, computer-aided diagnosis (CAD) using artificial intelligence techniques, in particular, deep neural networks have been investigated and several approaches have been proposed to deal with the automated detection of various pathologies using digital images. Recent studies indicate that the fusion of images with the patient’s clinical information is important for the final clinical diagnosis. As such dataset does not yet exist for oral cancer, as far as the authors are aware, a new dataset was collected consisting of histopathological images, demographic and clinical data. This study evaluated the importance of complementary data to histopathological image analysis of oral leukoplakia and carcinoma for CAD.Methods A new dataset (NDB-UFES) was collected from 2011 to 2021 consisting of histopathological images and information. The 237 samples were curated and analyzed by oral pathologists generating the gold standard for classification. State-of-the-art image fusion architectures and complementary data (Concatenation, Mutual Attention, MetaBlock and MetaNet) using the latest deep learning backbones were investigated for 4 distinct tasks to identify oral squamous cell carcinoma, leukoplakia with dysplasia and leukoplakia without dysplasia. We evaluate them using balanced accuracy, precision, recall and area under the ROC curve metrics.Results Experimental results indicate that the best models present balanced accuracy of 83.24% using images, demographic and clinical information with MetaBlock fusion and ResNetV2 backbone. It represents an improvement in performance of 30.68% (19.54 pp) in the task to differentiate samples diagnosed with oral squamous cell carcinoma and leukoplakia with or without dysplasia.Conclusion This study indicates that cured demographic and clinical data may positively influence the performance of artificial intelligence models in automated classification of oral cancer.

In the last few years, deep neural networks have achieved promising results in several fields. However, one of the main limitations of these methods is the need for large-scale datasets to properly generalize. Few-shot learning methods emerged as an attempt to solve this shortcoming. Among the few-shot learning methods, there is a class of methods known as embedding learning or metric learning. These methods tackle the classification problem by learning to compare, needing fewer training data. One of the main problems in plant diseases and pests recognition is the lack of large public datasets available. Due to this difficulty, the field emerges as an intriguing application to evaluate the few-shot learning methods. The field is also relevant due to the social and economic importance of agriculture in several countries. In this work, datasets consisting of biotic stresses in coffee leaves are used as a case study to evaluate the performance of few-shot learning in classification and severity estimation tasks. We achieved competitive results compared with the ones reported in the literature in the classification task, with accuracy values close to 96%. Furthermore, we achieved superior results in the severity estimation task, obtaining 6.74% greater accuracy than the baseline.

Student dropout represents a social, resource and time loss for everyone involved. By identifying students with the potential to evade, it is possible to take the necessary measures to prevent that from happening. This paper investigates different techniques for classification using a real students dataset that is multivariate, sequential and imbalanced. To address imbalance, an approach using weighted loss function and an approach using synthetic data generated through a combination of variational autoencoder and the Adaptive Synthetic (ADASYN) technique were investigated. In addition, we propose two models for predicting evasion using LSTM and 1D CNN networks to take advantage of the multivariate sequential property. The experiments showed that the approach only with LSTM architecture in the dynamic module using the preprocessed original dataset and without the use of weighted loss function obtained the best mean sensitivity (0.9676) among the investigated proposals. The LSTM architecture in the dynamic module using the preprocessed original dataset with weighted loss function had better specificity (0.9558) and area under curve (0.9735). Both had the same mean accuracy of 0.9549.

In recent years, deep learning methods have been introduced for segmentation and classification of leaf lesions caused by pests and pathogens. Among the commonly used approaches, convolutional neural networks have provided results with high accuracy. The purpose of this work is to present an effective and practical system capable of segmenting and classifying different types of leaf lesions and estimating the severity of stress caused by biotic agents in coffee leaves using convolutional neural networks. The proposed approach consists of two stages: a semantic segmentation stage with severity calculation and a symptom lesion classification stage. Each stage was tested separately, highlighting the positive and negative points of each one. We obtained very good results for the severity estimation, suggesting that the model can estimate severity values very close to the real values. For the biotic stress classification, the accuracy rates were greater than 97%. Due to the promising results obtained, an App for Android platform was developed and implemented, consisting of semantic segmentation and severity calculation, as well as symptom classification to assist both specialists and farmers to identify and quantify biotic stresses using images of coffee leaves acquired by smartphone.

Through the analysis of cultural, socioeconomic and academic performance aspects it is possible to map the profile of the students and their motivations to drop out. This article aims to create a computational tool for data visualization that allows drawing the profile of students to support educational institutions managers in the definition of dropout avoidance policies. We present a method to treat data collected by higher education institutions over the years, analyze them to understand the dropout and provide that information to the university and the general public. Eight questions were proposed to clarify the dropout from the Federal University of Esp\'irito Santo, Brazil. The questions were answered through the dashboard that helps to understand the causes of dropout. It is expected that this tool can be used by others educational institutions to draw student profiles contributing to possible resolution of the problem.

Deciding when to buy or sell a stock is not an easy task because the market is hard to predict, being influenced by political and economic factors. Thus, methodologies based on computational intelligence have been applied to this challenging problem. In this work, every day the stocks are ranked by technique for order preference by similarity to ideal solution (TOPSIS) using technical analysis criteria, and the most suitable stock is selected for purchase. Even so, it may occur that the market is not favorable to purchase on certain days, or even, the TOPSIS make an incorrect selection. To improve the selection, another method should be used. So, a hybrid model composed of empirical mode decomposition (EMD) and extreme learning machine (ELM) is proposed. The EMD decomposes the series into several sub-series, and thus the main omponent (trend) is extracted. This component is processed by the ELM, which performs the prediction of the next element of component. If the value predicted by the ELM is greater than the last value, then the purchase of the stock is confirmed. The method was applied in a universe of 50 stocks in the Brazilian market. The selection made by TOPSIS showed promising results when compared to the random selection and the return generated by the Bovespa index. Confirmation with the EMD-ELM hybrid model was able to increase the percentage of profit tradings.

Skin cancer is one of the most common types of cancer in the world. Different computer-aided diagnosis systems have been proposed to tackle skin lesion diagnosis, most of them based on deep convolutional neural networks. However, recent advances in computer vision achieved state-of-the-art results in many tasks, notably transformer-based networks. We explore and evaluate advances in computer vision architectures, training methods and multimodal feature fusion for skin lesion diagnosis task. Experiments show that PiT ( 0.800 ± 0.006 ), CoaT ( 0.780 ± 0.024 ) and ViT ( 0.771 ± 0.018 ) transformer-based backbone models with MetaBlock fusion achieved state-of-the-art results for balanced accuracy on PAD-UFES-20 dataset.

Skin lesions diagnostic is a challenging problem due to the variety of visual aspects of the lesions. The clinical analysis of skin lesions relies on the visual information as well as on the complementary information provided by the patient. Since dermatologists make use of visual cues and patient clinical information, we investigate if the combination of features from convolutional neural networks (CNN), handcrafted features and patient clinical information can improve the performance of automated diagnosis of skin cancer. Most works on skin lesion diagnosis in the literature use dermoscopic images without patient clinical information. In order to address this problem, we used a clinical image dataset of skin lesions with patient information collected via smartphone named PAD-UFES-20. With the proposed fusion architecture we show that the results using clinical features as a complement to the CNN and handcrafted features improve the classification in terms of balanced accuracy by 7.1 % for cancer and by 3.2 % for melanoma as compared with only features extracted from a CNN. In addition, our findings show that combining only handcrafted features with deep features did not improve the results, indicating the importance of using clinical metadata for skin lesion classification.