Training Data Research Articles

Artificial intelligence and machine learning for early cancer prediction and response

Worldwide, cancer claims more lives than any other disease. Although cancer detection, prognosis, and treatment have all advanced, one major obstacle is the lack of personalized, data-driven care. Artificial intelligence (AI), used to forecast and automate numerous malignancies, has emerged as a viable solution for increasing healthcare accuracy and patient outcomes. Artificial intelligence applications in cancer include risk evaluation, early detection, patient prognosis prediction, and treatment selection based on comprehensive data. Machine learning (ML), a form of artificial intelligence that allows computers to learn from training data, is very successful in predicting breast, brain, lung, liver, and prostate cancers. Indeed, AI and machine learning have outperformed physicians in predicting cancer. These technologies can potentially enhance the diagnosis, prediction, and quality of life of patients with a wide range of disorders, not just cancer. As a result, it is critical to enhance existing AI and ML technologies and create new programs to help patients. This article discusses the use of AI and machine learning algorithms in cancer prediction, including its present uses, future possibilities, and limitations.

Estimation of the mean effective pressure of a spark ignition internal combustion engine using a neural network, considering the wall-wetting dynamics

The management and development of internal combustion engines stand as critical pursuits within the automotive and related industries. Utilizing cylinder pressure as feedback, engine controllers rely on intricate systems to regulate performance. However, due to the inherent complexity and nonlinearity of engines, direct measurement of cylinder pressure through pressure sensors is costly and computationally demanding. Consequently, the need for accurate and detailed engine models becomes paramount. Neural networks offer a promising avenue for simulating internal combustion engines, combining speed and precision. By treating the engine as an enigmatic entity, neural networks can construct detailed models. This study aims to employ two types of neural networks—multilayer perceptron and radial basis functions—to train and build a model of an internal combustion engine. These networks will simulate and estimate the engine's mean suitable pressure, allowing for a comparison of their effectiveness. Prior to implementing the neural network architecture, an engine model was constructed in MATLAB to gather necessary training data. This preliminary step ensured a robust foundation for subsequent network design and implementation. In summary, this research focuses on leveraging neural networks to model internal combustion engines, utilizing both multilayer perceptron and radial basis functions to simulate engine behavior and estimate mean suitable pressure.

Semi-Supervised Correction Model for Turbulence-Distorted Images

Significant progress has been made in addressing turbulence distortion in recent years, but persistent challenges remain. Firstly, existing methods heavily rely on fully supervised optimization strategies and synthetic datasets, posing difficulties in effectively utilizing unlabeled real data for training. Secondly, most approaches construct networks in a straightforward manner, overlooking the representation model of phase distortion and point spread function (PSF) in spatial and channel dimensions. This oversight restricts the potential for distortion correction. To address these challenges, this paper proposes a semi-supervised atmospheric turbulence correction method based on the mean-teacher framework. Our approach imposes constraints on the unlabeled data of student networks using pseudo-labels generated by teacher networks, thereby enhancing the generalization ability by leveraging information from unlabeled data. Furthermore, we introduce to use no-reference image quality assessment criterion to select the most reliable pseudo-label for each unlabeled sample by predicting physical parameters that indicating the level of degradation. Additionally, we propose to combine sliding window-based self-attention with channel attention to facilitate local-global context interaction. This design is inspired by the representation of phase distortion and PSF, which can be characterized by coefficients and basis functions corresponding to the channel-wise representation of convolutional neural network features. Moreover, the base functions exhibit spatial correlation, akin to Zenike and Airy disks. Experimental results show that the proposed method surpasses state-of-the-art models.

Towards Consistency in Pediatric Brain Tumor Measurements: Challenges, Solutions, and the Role of AI-Based Segmentation.

MR imaging is central to the assessment of tumor burden and changes over time in neuro-oncology. Several response assessment guidelines have been set forth by the Response Assessment in Pediatric Neuro-Oncology (RAPNO) working groups in different tumor histologies; however, the visual delineation of tumor components using MRIs is not always straightforward, and complexities not currently addressed by these criteria can introduce inter- and intra-observer variability in manual assessments. Differentiation of non-enhancing tumor from peritumoral edema, mild enhancement from absence of enhancement, and various cystic components can be challenging; particularly given a lack of sufficient and uniform imaging protocols in clinical practice. Automated tumor segmentation with artificial intelligence (AI) may be able to provide more objective delineations, but rely on accurate and consistent training data created manually (ground truth). Herein, this paper reviews existing challenges and potential solutions to identifying and defining subregions of pediatric brain tumors (PBTs) that are not explicitly addressed by current guidelines. The goal is to assert the importance of defining and adopting criteria for addressing these challenges, as it will be critical to achieving standardized tumor measurements and reproducible response assessment in PBTs, ultimately leading to more precise outcome metrics and accurate comparisons among clinical studies.

Non-linear Kernel Optimisation of Support Vector Machine Algorithm for Online Marketplace Sentiment Analysis

Twitter is a social media platform that is very important in the digital world. Fast communication and interaction make Twitter a vital information center in sentiment analysis. The purpose of this research is to classify public opinion about the presence of marketplaces in Indonesia, both positive and negative sentiments, using a Non-linear SVM algorithm based on 1276 tweets. This research involves the stages of data pre-processing, labeling, feature extraction using TF-IDF, and data division into three scenarios: 80% training data and 20% test data, 50% training data and 50% test data scenario, and 20% training data and 80% test data scenario. The last process, GridSearchCV, combines cross-validation and non-linear SVM parameters for model evaluation using a confusion matrix. The best SVM model resulting from the scenario was 80% training and 20% test data, with hyperparameters Gamma = 100 and C = 0.01, achieving 89% accuracy. When tested on never-before-seen data, the accuracy increased to 90%, with an f1-score of 91%, precision of 88%, and recall of 95% on negative sentiments. In conclusion, evaluating the performance of non-linear SVM kernels with a combination of hyperparameter values can improve accuracy, especially on public response information about online marketplaces and public sentiment.

FlexDTI: flexible diffusion gradient encoding scheme-based highly efficient diffusion tensor imaging using deep learning

Objective. Most deep neural network-based diffusion tensor imaging methods require the diffusion gradients’ number and directions in the data to be reconstructed to match those in the training data. This work aims to develop and evaluate a novel dynamic-convolution-based method called FlexDTI for highly efficient diffusion tensor reconstruction with flexible diffusion encoding gradient scheme. Approach. FlexDTI was developed to achieve high-quality DTI parametric mapping with flexible number and directions of diffusion encoding gradients. The method used dynamic convolution kernels to embed diffusion gradient direction information into feature maps of the corresponding diffusion signal. Furthermore, it realized the generalization of a flexible number of diffusion gradient directions by setting the maximum number of input channels of the network. The network was trained and tested using datasets from the Human Connectome Project and local hospitals. Results from FlexDTI and other advanced tensor parameter estimation methods were compared. Main results. Compared to other methods, FlexDTI successfully achieves high-quality diffusion tensor-derived parameters even if the number and directions of diffusion encoding gradients change. It reduces normalized root mean squared error by about 50% on fractional anisotropy and 15% on mean diffusivity, compared with the state-of-the-art deep learning method with flexible diffusion encoding gradient scheme. Significance. FlexDTI can well learn diffusion gradient direction information to achieve generalized DTI reconstruction with flexible diffusion gradient scheme. Both flexibility and reconstruction quality can be taken into account in this network.

Limitations of Large Language Models in Propaganda Detection Task

Propaganda in the digital era is often associated with online news. In this study, we focused on the use of large language models and their detection of propaganda techniques in the electronic press to investigate whether it is a noteworthy replacement for human annotators. We prepared prompts for generative pre-trained transformer models to find spans in news articles where propaganda techniques appear and name them. Our study was divided into three experiments on different datasets—two based on an annotated SemEval2020 Task 11 corpora and one on an unannotated subset of the Polish Online News Corpus, which we claim to be an even bigger challenge as an example of an under-resourced language. Reproduction of the results of the first experiment resulted in a higher recall of 64.53% than the original run, and the highest precision of 81.82% was achieved for gpt-4-1106-preview CoT. None of our attempts outperformed the baseline F1 score. One of the attempts with gpt-4-0125-preview on original SemEval2020 Task 11 achieved an almost 20% F1 score, but it was below the baseline, which oscillated around 50%. Part of our work that was dedicated to Polish articles showed that gpt-4-0125-preview had a 74% accuracy in the binary detection of propaganda techniques and 69% in propaganda technique classification. The results for SemEval2020 show that the outputs of generative models tend to be unpredictable and are hardly reproducible for propaganda detection. For the time being, these are unreliable methods for this task, but we believe they can help to generate more training data.

Improving Stroke Detection with Hybrid Sampling and Cascade Generalization

The prevalence of stroke in Indonesia has increased. One survey in Indonesia that contains information about the health conditions of the Indonesian people is the Indonesian Family Life Survey (IFLS). The proportion of respondents who had a stroke and non-stroke in IFLS5 showed an imbalance with an extreme level of imbalance; hence, this research aims to overcome this problem with SMOTE, SMOTE-Tomek Link, and SMOTE-ENN; then, the balanced dataset is classified using the ensemble and cascade approaches to improve the detection of stroke risk and to identify the important variables. However, the stroke respondents were still challenging to classify after imbalance class handling, presumably because of the large amount of data before and after balancing. The solution is to balance the training data with various percentages. The results showed the best percentage is applied to 5% of the training data, balanced by the SMOTE-ENN, and the ensemble method with the cascade approach increases the sensitivity and balanced accuracy values. Random forest and logistic regression combine models that produce the best performance, with a classification tree as the final model. The important variables obtained from this combination are the addition of probability from random forest, logistic regression, history of hypertension, age, and physical activity.

A Projection Approach to Local Regression with Variable-Dimension Covariates

Incomplete covariate vectors are known to be problematic for estimation and inferences on model parameters, but their impact on prediction performance is less understood. We develop an imputation-free method that builds on a random partition model admitting variable-dimension covariates. Cluster-specific response models further incorporate covariates via linear predictors, facilitating estimation of smooth prediction surfaces with relatively few clusters. We exploit marginalization techniques of Gaussian kernels to analytically project response distributions according to any pattern of missing covariates, yielding a local regression with internally consistent uncertainty propagation that utilizes only one set of coefficients per cluster. Aggressive shrinkage of these coefficients regulates uncertainty due to missing covariates. The method allows in- and out-of-sample prediction for any missingness pattern, even if the pattern in a new subject’s incomplete covariate vector was not seen in the training data. We develop an MCMC algorithm for posterior sampling that improves a computationally expensive update for latent cluster allocation. Finally, we demonstrate the model’s effectiveness for nonlinear point and density prediction under various circumstances by comparing with other recent methods for regression of variable dimensions on synthetic and real data. Supplemental materials are available online.

Bagging Algorithm Based on Possibilistic Data Interpolation for Brain-Computer Interface

Recently, brain-computer interfaces (BCIs) and brain-machine interfaces have garnered the attention of researchers. Based on connections with external devices, external computers and machines can be controlled by brain signals measured via near-infrared spectroscopy (NIRS) or electroencephalograph devices. Herein, we propose a novel bagging algorithm that generates interpolation data around misclassified data using a possibilistic function, to be applied to BCIs. In contrast to AdaBoost, which is a conventional ensemble learning method that increases the weight of misclassified data to incorporate them with high probability to the next datasets, we generate interpolation data using a membership function centered on misclassified data and incorporate them into the next datasets simultaneously. The interpolated data are known as virtual data herein. By adding the virtual data to the training data, the volume of the training data becomes sufficient for adjusting the discriminate boundary more accurately. Because the membership function is defined as a possibility distribution, this method is named the bagging algorithm based on the possibility distribution. Herein, we formulate a bagging-type ensemble learning based on the possibility distribution and discuss the usefulness of the proposed method for solving simple calculations using NIRS data.

Synthetic aperture radar automatic target recognition based on cost‐sensitive awareness generative adversarial network for imbalanced data

AbstractIn military contexts, synthetic aperture radar (SAR) automatic target recognition (ATR) models frequently encounter the challenge of imbalanced data, resulting in a noticeable degradation in the recognition performance. Therefore, the authors propose a cost‐sensitive awareness generative adversarial network (CAGAN) model, aiming to improve the robustness of ATR models under imbalanced data. Firstly, the authors introduce a convolutional neural network (DCNN) to extract features. Then, the synthetic minority over‐sampling technique (SMOTE) is applied to achieve feature‐level balancing for the minority category. Finally, a CAGAN model is designed to perform the final classification task. In this process, the GAN‐based adversarial training mechanism enriches the diversity of training samples, making the ATR model more comprehensive in understanding different categories. In addition, the cost matrix increases the penalty for misclassification results and further improves the classification accuracy. Simultaneously, the cost‐sensitive awareness can accurately adjust the cost matrix through training data, thus reducing dependence on expert knowledge and improving the generalisation performance of the ATR model. This model is an end‐to‐end ATR scheme, which has been experimentally validated on the MSTAR and OpenSARship datasets. Compared to other methods, the proposed method exhibits strong robustness in dealing with various imbalanced scenarios and significant generalisation capability across different datasets.

Training a Dataset Simulated Using RGB Images for an End-to-End Event-Based DoLP Recovery Network

Event cameras are bio-inspired neuromorphic sensors that have emerged in recent years, with advantages such as high temporal resolutions, high dynamic ranges, low latency, and low power consumption. Event cameras can be used to build event-based imaging polarimeters, overcoming the limited frame rates and low dynamic ranges of existing systems. Since events cannot provide absolute brightness intensity in different angles of polarization (AoPs), degree of linear polarization (DoLP) recovery in non-division-of-time (non-DoT) event-based imaging polarimeters is an ill-posed problem. Thus, we need a data-driven deep learning approach. Deep learning requires large amounts of data for training, and constructing a dataset for event-based non-DoT imaging polarimeters requires significant resources, scenarios, and time. We propose a method for generating datasets using simulated polarization distributions from existing red–green–blue images. Combined with event simulator V2E, the proposed method can easily construct large datasets for network training. We also propose an end-to-end event-based DoLP recovery network to solve the problem of DoLP recovery using event-based non-DoT imaging polarimeters. Finally, we construct a division-of-time event-based imaging polarimeter simulating an event-based four-channel non-DoT imaging polarimeter. Using real-world polarization events and DoLP ground truths, we demonstrate the effectiveness of the proposed simulation method and network.

Construction and optimization of non-parametric analysis model for meter coefficients via back propagation neural network

This study addresses the drawbacks of traditional methods used in meter coefficient analysis, which are low accuracy and long processing time. A new method based on non-parametric analysis using the Back Propagation (BP) neural network is proposed to overcome these limitations. The study explores the classification and pattern recognition capabilities of the BP neural network by analyzing its non-parametric model and optimization methods. For model construction, the study uses the United Kingdom Domestic Appliance-Level Electricity dataset’s meter readings and related data for training and testing the proposed model. The non-parametric analysis model is used for data pre-processing, feature extraction, and normalization to obtain the training and testing datasets. Experimental tests compare the proposed non-parametric analysis model based on the BP neural network with the traditional Least Squares Method (LSM). The results demonstrate that the proposed model significantly improves the accuracy indicators such as mean absolute error (MAE) and mean relative error (MRE) when compared with the LSM method. The proposed model achieves an MAE of 0.025 and an MRE of 1.32% in the testing dataset, while the LSM method has an MAE of 0.043 and an MRE of 2.56% in the same dataset. Therefore, the proposed non-parametric analysis model based on the BP neural network can achieve higher accuracy in meter coefficient analysis when compared with the traditional LSM method. This study provides a novel non-parametric analysis method with practical reference value for the electricity industry in energy metering and load forecasting.

A Deep Learning-Based Crop Disease Diagnosis Method Using Multimodal Mixup Augmentation

With the widespread adoption of smart farms and continuous advancements in IoT (Internet of Things) technology, acquiring diverse additional data has become increasingly convenient. Consequently, studies relevant to deep learning models that leverage multimodal data for crop disease diagnosis and associated data augmentation methods are significantly growing. We propose a comprehensive deep learning model that predicts crop type, detects disease presence, and assesses disease severity at the same time. We utilize multimodal data comprising crop images and environmental variables such as temperature, humidity, and dew points. We confirmed that the results of diagnosing crop diseases using multimodal data improved 2.58%p performance compared to using crop images only. We also propose a multimodal-based mixup augmentation method capable of utilizing both image and environmental data. In this study, multimodal data refer to data from multiple sources, and multimodal mixup is a data augmentation technique that combines multimodal data for training. This expands the conventional mixup technique that was originally applied solely to image data. Our multimodal mixup augmentation method showcases a performance improvement of 1.33%p compared to the original mixup method.

A Three-Zone Identification Method for Coal Mine Area Based on DS Evidence Theory

As coal ore and other resources are continuously mined, a three-zone structure is formed underground consisting of a sagging zone, fault zone, and caving zone. The use of well-logging data to identify the three zones is important for production safety and environmental management. Owing to the scarcity of data that can reflect three zones in normal coal mining, conventional identification and prediction methods face challenges when extracting data features, incurring a degree of uncertainty within prediction results. Accordingly, the accurate identification of the three zones has become a critical objective in daily production. To address this issue, we developed a method called a method called backpropagation neural networks with Dempster–Shafer (DS) evidence theory. Initially, we preprocessed the training data and deployed two backpropagation neural networks (BPNNs) to predict the three zones according to two parameters. According to these prediction results, the local and global credibility of each prediction is calculated and used to obtain the basic probability assignment function required for the DS evidence theory. Finally, the DS evidence theory is used to fuse the two BPNNs prediction results, thereby producing the final prediction results. The proposed method was demonstrated to improve prediction accuracy by 6.4% compared to a conventional neural network.

Sampling the user controls in neural modeling of audio devices

This work studies neural modeling of nonlinear parametric audio circuits, focusing on how the diversity of settings of the target device user controls seen during training affects network generalization. To study the problem, a large corpus of training datasets is synthetically generated using SPICE simulations of two distinct devices, an analog equalizer and an analog distortion pedal. A proven recurrent neural network architecture is trained using each dataset. The difference in the datasets is in the sampling resolution of the device user controls and in their overall size. Based on objective and subjective evaluation of the trained models, a sampling resolution of five for the device parameters is found to be sufficient to capture the behavior of the target systems for the types of devices considered during the study. This result is desirable, since a dense sampling grid can be impractical to realize in the general case when no automated way of setting the device parameters is available, while collecting large amounts of data using a sparse grid only incurs small additional costs. Thus, the result provides guidance for efficient collection of training data for neural modeling of other similar audio devices.

Versatile multiple object tracking in sparse 2D/3D videos via deformable image registration.

Tracking body parts in behaving animals, extracting fluorescence signals from cells embedded in deforming tissue, and analyzing cell migration patterns during development all require tracking objects with partially correlated motion. As dataset sizes increase, manual tracking of objects becomes prohibitively inefficient and slow, necessitating automated and semi-automated computational tools. Unfortunately, existing methods for multiple object tracking (MOT) are either developed for specific datasets and hence do not generalize well to other datasets, or require large amounts of training data that are not readily available. This is further exacerbated when tracking fluorescent sources in moving and deforming tissues, where the lack of unique features and sparsely populated images create a challenging environment, especially for modern deep learning techniques. By leveraging technology recently developed for spatial transformer networks, we propose ZephIR, an image registration framework for semi-supervised MOT in 2D and 3D videos. ZephIR can generalize to a wide range of biological systems by incorporating adjustable parameters that encode spatial (sparsity, texture, rigidity) and temporal priors of a given data class. We demonstrate the accuracy and versatility of our approach in a variety of applications, including tracking the body parts of a behaving mouse and neurons in the brain of a freely moving C. elegans. We provide an open-source package along with a web-based graphical user interface that allows users to provide small numbers of annotations to interactively improve tracking results.

Site Identification and Next Choice Protocol for Hit-to-Lead Optimization.

Time efficiency and cost savings are major challenges in drug discovery and development. In this process, the hit-to-lead stage is expected to improve efficiency because it primarily exploits the trial-and-error approach of medicinal chemists. This study proposes a site identification and next choice (SINCHO) protocol to improve the hit-to-lead efficiency. This protocol selects an anchor atom and growth site pair, which is desirable for a hit-to-lead strategy starting from a 3D complex structure. We developed and fine-tuned the protocol using a training data set and assessed it using a test data set of the preceding hit-to-lead strategy. The protocol was tested for experimentally determined structures and molecular dynamics (MD) ensembles. The protocol had a high prediction accuracy for applying MD ensembles, owing to the consideration of protein flexibility. The SINCHO protocol enables medicinal chemists to visualize and modify functional groups in a hit-to-lead manner.

Dynamical Tests of a Deep-Learning Weather Prediction Model

Abstract Global deep-learning weather prediction models have recently been shown to produce forecasts that rival those from physics-based models run at operational centers. It is unclear whether these models have encoded atmospheric dynamics, or simply pattern matching that produces the smallest forecast error. Answering this question is crucial to establishing the utility of these models as tools for basic science. Here we subject one such model, Pangu-Weather, to a set of four classical dynamical experiments that do not resemble the model training data. Localized perturbations to the model output and the initial conditions are added to steady time-averaged conditions, to assess the propagation speed and structural evolution of signals away from the local source. Perturbing the model physics by adding a steady tropical heat source results in a classical Matsuno–Gill response near the heating, and planetary waves that radiate into the extratropics. A localized disturbance on the winter-averaged North Pacific jet stream produces realistic extratropical cyclones and fronts, including the spontaneous emergence of polar lows. Perturbing the 500hPa height field alone yields adjustment from a state of rest to one of wind–pressure balance over ∼6 hours. Localized subtropical low pressure systems produce Atlantic hurricanes, provided the initial amplitude exceeds about 4 hPa, and setting the initial humidity to zero eliminates hurricane development. We conclude that the model encodes realistic physics in all experiments, and suggest it can be used as a tool for rapidly testing a wide range of hypotheses.

A Comparative Study of Students Graduation Analysis Using Classification Methods in Undergraduate Electrical Engineering Tidar University

This research aimed to classify achievement factors for electrical engineering students at Tidar University using K-Means and Agglomerative Clustering classification algorithms. The goal was to understand if any parameters influence high-achieving student performance. The Indonesian government and private sector for university students provide significant education funds. Student scholarships are awarded based primarily on GPA and entry path, overburdening staff and causing confusion during distribution to eligible recipients. A system was needed to accommodate additional eligible criteria. The researcher selected factors to identify engineering student performance, including school origin, entry path, tuition fees, and GPA. These inputs could determine graduation status. The results compared calculation methods based on collected data accuracy, processing times, and characterizing clustered data to determine the best classification method. Agglomerative Hierarchical Clustering performed better. Accuracy testing on 600 training data points yielded 73.94% for improved K-means and 90.42% for AHC. The Average processing time was 674.92 seconds for improved K-means and 554.35 seconds for AHC. Silhouette testing also characterized calculation methods, with improved K-means scoring best at 0.654 and AHC at 0.787 using two clusters.