OBJECTIVE: Development and validation of a radiomics model for predicting occult locally advanced esophageal squamous cell carcinoma (LA-ESCC) on computed tomography (CT) radiomic features before implementation of treatment.
METHODS: The study retrospectively collected 574 patients with esophageal squamous cell carcinoma (ESCC) from two medical centers, which were divided into three cohorts for training, internal and external validation. After delineating volume of interest (VOI), radiomics features were extracted and subjected to feature selection using three robust methods. Subsequently, 10 machine learning models were constructed, among which the optimal model was utilized to establish a radiomics signature. Furthermore, a predictive nomogram incorporating both clinical and radiomics signatures was developed. The performance of these models was evaluated through receiver operating characteristic curves, calibration curves, decision curve analysis as well as measures including accuracy, sensitivity, and specificity.
RESULTS: A total of 19 radiomics features were selected. The multilayer perceptron (MLP), which was found to be optimal, achieved an AUC of 0.919, 0.864 and 0.882 in the training, internal and external validation cohorts, respectively. Similarly, MLP showed good accuracy in distinguish occult LA-ESCC in subgroup of cT1-2N0M0 diagnosed by clinicians with 0.803 and 0.789 in two validation cohorts respectively. By incorporating the radiomics signature with clinical signature, a predictive nomogram demonstrated superior prediction performance with an AUC of 0.877 and accuracy of 0.85 in external validation cohort.
CONCLUSIONS: The radiomics and machine learning model can offers improved accuracy in prediction of occult LA-ESCC, providing valuable assistance to clinicians when choosing treatment plans.

The widespread use of devices like mobile phones and wearables allows for automatic monitoring of human daily activities, generating vast datasets that offer insights into long-term human behavior. A structured and controlled data collection process is essential to unlock the full potential of this information. While wearable sensors for physical activity monitoring have gained significant traction in healthcare, sports science, and fitness applications, securing diverse and comprehensive datasets for research and algorithm development poses a notable challenge. In this proof-of-concept study, we underscore the significance of semantic representation in enhancing data interoperability and facilitating advanced analytics for physical activity sensor observations. Our approach focuses on enhancing the usability of physical activity datasets by employing a medical-grade (CE certified) sensor to generate synthetic datasets. Additionally, we provide insights into ethical considerations related to synthetic datasets. The study conducts a comparative analysis between real and synthetic activity datasets, assessing their effectiveness in mitigating model bias and promoting fairness in predictive analysis. We have created an ontology for semantically representing observations from physical activity sensors and conducted predictive analysis on data collected using MOX2-5 activity sensors. Until now, there has been a lack of publicly available datasets for physical activity collected with MOX2-5 activity monitoring medical grade (CE certified) device. The MOX2-5 captures and transmits high-resolution data, including activity intensity, weight-bearing, sedentary, standing, low, moderate, and vigorous physical activity, as well as steps per minute. Our dataset consists of physical activity data collected from 16 adults (Male: 12; Female: 4) over a period of 30-45 days (approximately 1.5 months), yielding a relatively small volume of 539 records. To address this limitation, we employ various synthetic data generation methods, such as Gaussian Capula (GC), Conditional Tabular General Adversarial Network (CTGAN), and Tabular General Adversarial Network (TABGAN), to augment the dataset with synthetic data. For both the authentic and synthetic datasets, we have developed a Multilayer Perceptron (MLP) classification model for accurately classifying daily physical activity levels. The findings underscore the effectiveness of semantic ontology in semantic search, knowledge representation, data integration, reasoning, and capturing meaningful relationships between data. The analysis supports the hypothesis that the efficiency of predictive models improves as the volume of additional synthetic training data increases. Ontology and Generative AI hold the potential to expedite advancements in behavioral monitoring research. The data presented, encompassing both real MOX2-5 and its synthetic counterpart, serves as a valuable resource for developing robust methods in activity type classification. Furthermore, it opens avenues for exploration into research directions related to synthetic data, including model efficiency, detection of generated data, and considerations regarding data privacy.

Purpose: For systematic athletic training and targeted talent development, it is essential to know the physical fitness and motor competencies of top athletes in detail. However, it can be difficult to identify differences in performance requirements and thus to provide adequate support, especially for sports that at first glance appear to have similar demands-such as track and field throwing disciplines. Therefore, the aim of the study was to examine the physical fitness and motor competence of top athletes from different throwing disciplines and to check whether the athletes\' performance parameters match the specific requirements of the respective sport. Methods: The study involved 289 male youth athletes (aged 14-18 years) across four distinct throwing disciplines: shot put (n = 101), hammer throw (n = 16), discus throw (n = 63), and javelin throw (n = 109). The performance evaluation comprised three anthropometric measurements and twelve motor performance prerequisites applicable to the throwing disciplines. Discriminant analysis and neural networks (Multilayer Perceptron) were implemented to determine the possibility of distinguishing among athletes from the four sports. Results: The study\'s findings indicate that in male throwing athletes, disparities in general physical fitness and motor proficiency assessments discern the majority of talented young athletes based on their specific sport (discriminant analysis: 68.2%; multilayer perceptron analysis: 72.2%). This remains applicable irrespective of the classification method employed. Discus throwers possessed a height advantage, while shot putters and hammer throwers exhibited superior arm strength. Javelin throwers displayed better explosive strength and sprinting speed. Except for the hammer throwers, all events demonstrated a high level of explosive power in the medicine ball forward or backward throw test, which was especially crucial for shot put and discus athletes. Conclusion: The significance of physical fitness and motor competence tests in identifying and transferring talented athletes in track and field throwing disciplines has been affirmed. Using linear and non-linear classification methods, most athletes could be assigned to their correct sport. However, this also shows that slightly different training and talent identification is required for each of these sports. Furthermore, non-linear analysis methods can provide useful support for the development processes in junior competitive sports.

Unsupervised defect detection methods have garnered substantial attention in industrial defect detection owing to their capacity to circumvent complex fault sample collection. However, these models grapple with establishing a robust boundary between normal and abnormal conditions in intricate scenarios, leading to a heightened frequency of false-positive predictions. Spurious alerts exacerbate the work of reconfirmation and impede the widespread adoption of unsupervised anomaly detection models in industrial applications. To this end, we delve into the sole available data source in unsupervised defect detection models, the unsupervised training dataset, to introduce a solution called the False Alarm Identification (FAI) method aimed at learning the distribution of potential false alarms using anomaly-free images. It exploits a multi-layer perceptron to capture the semantic information of potential false alarms from a detector trained on anomaly-free training images at the object level. During the testing phase, the FAI model operates as a post-processing module applied after the baseline detection algorithm. The FAI algorithm determines whether each positive patch predicted by the normalizing flow algorithm is a false alarm by its semantic features. When a positive prediction is identified as a false alarm, the corresponding pixel-wise predictions are set to negative. The effectiveness of the FAI method is demonstrated by two state-of-the-art normalizing flow algorithms on extensive industrial applications.

UNASSIGNED: Identifying patients with hepatocellular carcinoma (HCC) at high risk of recurrence after hepatectomy can help to implement timely interventional treatment. This study aimed to develop a machine learning (ML) model to predict the recurrence risk of HCC patients after hepatectomy.
UNASSIGNED: We retrospectively collected 315 HCC patients who underwent radical hepatectomy at the Third Affiliated Hospital of Sun Yat-sen University from April 2013 to October 2017, and randomly divided them into the training and validation sets at a ratio of 7:3. According to the postoperative recurrence of HCC patients, the patients were divided into recurrence group and non-recurrence group, and univariate and multivariate logistic regression were performed for the two groups. We applied six machine learning algorithms to construct the prediction models and performed internal validation by 10-fold cross-validation. Shapley additive explanations (SHAP) method was applied to interpret the machine learning model. We also built a web calculator based on the best machine learning model to personalize the assessment of the recurrence risk of HCC patients after hepatectomy.
UNASSIGNED: A total of 13 variables were included in the machine learning models. The multilayer perceptron (MLP) machine learning model was proved to achieve optimal predictive value in test set (AUC = 0.680). The SHAP method displayed that γ-glutamyl transpeptidase (GGT), fibrinogen, neutrophil, aspartate aminotransferase (AST) and total bilirubin (TB) were the top 5 important factors for recurrence risk of HCC patients after hepatectomy. In addition, we further demonstrated the reliability of the model by analyzing two patients. Finally, we successfully constructed an online web prediction calculator based on the MLP machine learning model.
UNASSIGNED: MLP was an optimal machine learning model for predicting the recurrence risk of HCC patients after hepatectomy. This predictive model can help identify HCC patients at high recurrence risk after hepatectomy to provide early and personalized treatment.

Artificial neural networks can solve various tasks in computer vision, such as image classification, object detection, and general recognition. Our comparative study deals with four types of artificial neural networks-multilayer perceptrons, probabilistic neural networks, radial basis function neural networks, and convolutional neural networks-and investigates their ability to classify 2D matrix codes (Data Matrix codes, QR codes, and Aztec codes) as well as their rotation. The paper presents the basic building blocks of these artificial neural networks and their architecture and compares the classification accuracy of 2D matrix codes under different configurations of these neural networks. A dataset of 3000 synthetic code samples was used to train and test the neural networks. When the neural networks were trained on the full dataset, the convolutional neural network showed its superiority, followed by the RBF neural network and the multilayer perceptron.

This study concerns dispersed data stored in independent local tables with different sets of attributes. The paper proposes a new method for training a single neural network-a multilayer perceptron based on dispersed data. The idea is to train local models that have identical structures based on local tables; however, due to different sets of conditional attributes present in local tables, it is necessary to generate some artificial objects to train local models. The paper presents a study on the use of varying parameter values in the proposed method of creating artificial objects to train local models. The paper presents an exhaustive comparison in terms of the number of artificial objects generated based on a single original object, the degree of data dispersion, data balancing, and different network structures-the number of neurons in the hidden layer. It was found that for data sets with a large number of objects, a smaller number of artificial objects is optimal. For smaller data sets, a greater number of artificial objects (three or four) produces better results. For large data sets, data balancing and the degree of dispersion have no significant impact on quality of classification. Rather, a greater number of neurons in the hidden layer produces better results (ranging from three to five times the number of neurons in the input layer).

The use of models capable of forecasting the production of photovoltaic (PV) energy is essential to guarantee the best possible integration of this energy source into traditional distribution grids. Long Short-Term Memory networks (LSTMs) are commonly used for this purpose, but their use may not be the better option due to their great computational complexity and slower inference and training time. Thus, in this work, we seek to evaluate the use of neural networks MLPs (Multilayer Perceptron), Recurrent Neural Networks (RNNs), and LSTMs, for the forecast of 5 min of photovoltaic energy production. Each iteration of the predictions uses the last 120 min of data collected from the PV system (power, irradiation, and PV cell temperature), measured from 2019 to mid-2022 in Maceió (Brazil). In addition, Bayesian hyperparameters optimization was used to obtain the best of each model and compare them on an equal footing. Results showed that the MLP performs satisfactorily, requiring much less time to train and forecast, indicating that they can be a better option when dealing with a very short-term forecast in specific contexts, for example, in systems with little computational resources.

Microarrays are applications of electrical engineering and technology in biology that allow simultaneous measurement of expression of numerous genes, and they can be used to analyze specific diseases. This study undertakes classification analyses of various microarrays to compare the performances of classification algorithms over different data traits. The datasets were classified into test and control groups based on five utilized machine learning methods, including MultiLayer Perceptron (MLP), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and k-Nearest Neighbors (KNN), and the resulting accuracies were compared. k-fold cross-validation was used in evaluating the performance and the result was analyzed by comparing the performances of the five machine learning methods. Through the experiments, it was observed that the two tree-based methods, DT and RF, showed similar trends in results and the remaining three methods, MLP, SVM, and DT, showed similar trends. DT and RF generally showed worse performance than other methods except for one dataset. This suggests that, for the effective classification of microarray data, selecting a classification algorithm that is suitable for data traits is crucial to ensure optimum performance.

Revealing associations among various structural and functional patterns of the brain can yield highly informative results about the healthy and disordered brain. Studies using neuroimaging data have more recently begun to utilize the information within as well as across various functional and anatomical domains (i.e., groups of brain networks). However, most whole-brain approaches assume similar complexity of interactions throughout the brain. Here we investigate the hypothesis that interactions between brain networks capture varying amounts of complexity, and that we can better capture this information by varying the complexity of the model subspace structure based on available training data. To do this, we employ a Bayesian optimization-based framework known as the Tree Parzen Estimator (TPE) to identify, exploit and analyze patterns of variation in the information encoded by temporal information extracted from functional magnetic resonance imaging (fMRI) subdomains of the brain. Using a repeated cross-validation procedure on a schizophrenia classification task, we demonstrate evidence that interactions between specific functional subdomains are better characterized by more sophisticated model architectures compared to less complicated ones required by the others for optimally contributing towards classification and understanding the brain\'s functional interactions. We show that functional subdomains known to be involved in schizophrenia require more complex architectures to optimally unravel discriminatory information about the disorder. Our study points to the need for adaptive, hierarchical learning frameworks that cater differently to the features from different subdomains, not only for a better prediction but also for enabling the identification of features predicting the outcome of interest.