Training Dataset Research Articles

Data free knowledge distillation with feature synthesis and spatial consistency for image analysis.

Privacy and security concerns restrict access to original training datasets, posing significant challenges for model compression. Data-Free Knowledge Distillation (DFKD) emerges as a solution, aiming to transfer knowledge from teacher to student networks without accessing original data. Existing DFKD methods struggle to generate high-quality synthetic samples that capture the complexities of real-world data, leading to suboptimal knowledge transfer. Moreover, these approaches often fail to preserve the spatial attributes of the teacher network, resulting in shortcut learning and limited generalization.To address these issues, a novel DFKD strategy is proposed with three innovations: (1) an enhanced DCGAN generator with an attention module for synthesizing samples with improved micro-discriminative features; (2) a Multi-Scale Spatial Activation Region Consistency (MSARC) mechanism to accurately replicate the teacher's spatial attributes; and (3) an adversarial learning framework that creates a dynamic competitive environment between the generative and distillation phases. Rigorous evaluation of the method on several benchmark datasets, including CIFAR-10, CIFAR-100, Tiny-ImageNet, and medical imaging datasets such as PathMNIST, BloodMNIST, and PneumoniaMNIST, demonstrates superior performance compared to existing DFKD methods. Specifically, on CIFAR-100, the student network attains an accuracy of 77.85%, surpassing previous methods like CMI and SpaceshipNet. On BloodMNIST, the method achieves an accuracy of 80.50%, outperforming the next best method by over 5%.

NIMG-63. LEVERAGING LLMS FOR ACCURATE DIFFERENTIATION OF RADIATION NECROSIS AND TUMOR PROGRESSION IN BRAIN MRI REPORTS: A STUDY ON AUTOMATED SCORING AND CLINICAL IMPLICATIONS

Abstract BACKGROUND Accurate differentiation between radiation necrosis and tumor progression or recurrence in patients treated with stereotactic radiosurgery for brain metastases is critical for guiding clinical management. This study leverages the advanced natural language processing capabilities of Meta Llama3, an artificial intelligence (AI) large language model (LLM), combined with prompt engineering, to rapidly categorize brain magnetic resonance imaging (MRI) radiology reports. Our objective was to develop an automated scoring system to classify concern for radiation necrosis, tumor progression, equivocal findings, or stable exams. METHODS Using a comprehensive dataset of reports annotated by expert radiologists, we ran inference on a 70-billion parameter Llama3 model (temperature 0.2, top_p 0.9) with specific prompts designed to capture the nuanced language and diagnostic criteria related to radiation necrosis or tumor progression. RESULTS The first pass was performed on a training dataset of 107 reports and did not predefine the clinical conditions. This demonstrated 43.4% accuracy in scoring when compared to a human user’s classification of each report. Agreement between the human reader and Llama3 was assessed using the Gwet agreement coefficient, AC1=0.411 (99%CI 0.396-0.426). Multiple iterations of targeted prompt engineering were then employed to narrow the definition of radiation necrosis and tumor progression, with specific examples and nuanced language used to achieve a higher degree of accuracy at 72.0%, AC1=0.719 (99%CI 0.717-0.722). DISCUSSION/FUTURE DIRECTIONS This surpasses recent demonstration of lower human-LLM agreement in radiographic score assignment, with further room for calibration on a dataset of several thousand reports. Next, we will correlate automated interpretations with actual clinical management decisions for radiation necrosis (e.g., initiation of steroids, bevacizumab, Laser Interstitial Thermal Therapy, and/or repeat imaging). This automated scoring system holds significant potential for LLMs in clinical applications. Future work will focus on integrating this model into clinical workflows and expanding its capabilities to include longitudinal monitoring of patient outcomes.

A modular protein language modelling approach to immunogenicity prediction.

Neoantigen immunogenicity prediction is a highly challenging problem in the development of personalised medicines. Low reactivity rates in called neoantigens result in a difficult prediction scenario with limited training datasets. Here we describe ImmugenX, a modular protein language modelling approach to immunogenicity prediction for CD8+ reactive epitopes. ImmugenX comprises of a pMHC encoding module trained on three pMHC prediction tasks, an optional TCR encoding module and a set of context specific immunogenicity prediction head modules. Compared with state-of-the-art models for each task, ImmugenX's encoding module performs comparably or better on pMHC binding affinity, eluted ligand prediction and stability tasks. ImmugenX outperforms all compared models on pMHC immunogenicity prediction (Area under the receiver operating characteristic curve = 0.619, average precision: 0.514), with a 7% increase in average precision compared to the next best model. ImmugenX shows further improved performance on immunogenicity prediction with the integration of TCR context information. ImmugenX performance is further analysed for interpretability, which locates areas of weakness found across existing immunogenicity models and highlight possible biases in public datasets.

Proficiency of Turing test on Nigerian Major Languages; Yoruba, Igbo and Hausa Using Google Translate

Google Translate performance evaluation in translating between English and three Nigerian languages: Yoruba, Igbo, and Hausa, highlighting its strengths and limitations using the Turing Test. The study assessed whether participants could distinguish Artificial Intelligence (AI) - generated translations from human translations which reveals that Google Translate handles basic communication in these languages but struggles with more complex linguistic features. The tool adequately translates straightforward sentences in Yoruba language, but falters with idiomatic expressions, proverbs, and tonal nuances, the same problem refects in Igbo language facing challenges with more complex sentences and dialectal variations, resulting in culturally irrelevant outputs. Hausa performed better due to the availability of a larger training dataset, delivering more accurate translations for common phrases, but still struggled with cultural and contextual nuances.The findings suggest that Google Translate's limitations prevent it from passing a Turing Test with native speakers of Yoruba, Igbo, and Hausa. While understandable translations are produced for basic sentences, native speakers can easily detect inconsistencies, awkward phrasing, and a lack of cultural nuance. These issues underscore the significant gap between AI translation and human-like proficiency for African languages. Keyword: Turing test, Nigerian Major Languages; Yoruba, Igbo, Hausa Using Google Translate Abiola, O.A., Adebisi R.O. & Akinola, S.O. (2024): Proficiency of Turing test on Nigerian Major Languages; Yoruba, Igbo and Hausa Using Google Translate. Journal of Advances in Mathematical & Computational Science. Vol. 12, No. 3. Pp 1-12 Available online at www.isteams.net/mathematics-computationaljournal. dx.doi.org/10.22624/AIMS/MATHS/V12N4P1

Identification of Cactus Species Using the MobileNet Convolutional Neural Network Model

The objective of this design is to create a system capable of accurately and precisely classifying different types of succulent cacti. The system design aims to create a succulent cactus classification application using a Convolutional Neural Network (CNN) with the MobileNet architecture. The process involves collecting cactus images, dividing them into training and test datasets, and developing a CNN model to recognize patterns in succulent cactus species. The training data is used to train the CNN model, and the test data evaluates the model's accuracy. The trained model, stored in Tf.lite format, successfully classifies 15 cactus types, achieving high accuracy by employing preprocessing steps like resizing, normalization, and background removal. Over 1,200 cactus images were taken with a smartphone, categorized into 15 classes, and prepared to ensure optimal lighting, angle, background, and resolution (224x224 pixels). The MobileNet model was chosen for its high accuracy and efficiency. Hardware used includes a Samsung A54 smartphone and an Intel i7 laptop, with software such as Python, Kotlin, and Android Studio facilitating development. This design ensures the application’s accessibility, making it a valuable tool for cactus enthusiasts and the general public to easily identify different succulent cactus types. Testing of the cactus species classification program using the Convolutional Neural Network (CNN) method with the MobileNetV2 architecture yielded strong results, achieving over 90% accuracy in classifying 15 cactus species. The highest training accuracy of 0.9837 was achieved at 150 epochs without early stopping, outperforming other epoch configurations. The model successfully classified species across five main genera—Kalanchoe, Crassula, Echeveria, Haworthia, and Euphorbia. This high accuracy highlights the model's effectiveness, making it a useful tool for cactus enthusiasts and the public to accurately identify and distinguish cactus species.

Analysis of Paddy Field Changes (1989–2021) Using Landsat Images and Flooding-Assisted MLC in an Urbanizing Tropical Watershed, Vientiane, Lao PDR

Paddy fields are essential for food security and sustaining global dietary needs, yet urban expansion often encroaches on agricultural lands. Analyzing paddy fields and land use/land cover changes over time using satellite images provides critical insights for sustainable food production and balanced urban growth. However, mapping the paddy fields in tropical monsoon areas presents challenges due to persistent weather interference, monsoon-submerged fields, and a lack of training data. To address these challenges, this study proposed a flooding-assisted maximum likelihood classification (F-MLC) method. This approach utilizes accurate training datasets from intersecting flooded paddy field maps from the rainy and dry seasons, combined with the Automated Water Extraction Index (AWEI) to distinguish natural water bodies. The F-MLC method offers a robust solution for accurately mapping paddy fields and land use changes in challenging tropical monsoon climates. The classified images for 1989, 2000, 2013, and 2021 were produced and categorized into the following five major classes: urban areas, vegetation, paddy fields, water bodies, and other lands. The paddy field class derived for each year was validated using samples from various sources, contributing to the overall accuracies ranging from 83.6% to 90.4%, with a Kappa coefficient of between 0.80 and 0.88. The study highlights a significant decrease in paddy fields, while urban areas rapidly increased, replacing 23% of paddy fields between 1989 and 2021 in the watershed. This study demonstrates the potential of the F-MLC method for analyzing paddy fields and other land use changes over time in the tropical watershed. These findings underscore the urgent need for robust policy measures to protect paddy fields by clearly defining urban expansion boundaries, prioritizing paddy field preservation, and integrating these green spaces into urban development plans. Such measures are vital for ensuring a sustainable local food supply, promoting balanced urban growth, and maintaining ecological balance within the watershed.

Estimation of Crop Residue Cover Utilizing Multiple Ground Truth Survey Techniques and Multi-Satellite Regression Models

Soil erosion within agricultural landscapes has significant environmental and economic impacts and is strongly driven by reduced residue cover in agricultural fields. Large-area soil erosion models such as the Daily Erosion Project are important tools for understanding the patterns of soil erosion, but they rely on the accurate estimation of crop residue cover over large regions to infer the tillage practices, an erosion model input. Remote sensing analyses are becoming accepted as a reliable way to estimate crop residue cover, but most use localized training datasets that may not scale well outside small study areas. An alternative source of training data may be commonly conducted tillage surveys that capture information via rapid “windshield” surveys. In this study, we utilized the Google Earth Engine to assess the utility of three crop residue survey types (windshield tillage surveys, windshield binned residue surveys, and photo analysis surveys) and one synthetic survey (retroactively binned photo analysis data) as sources of training data for crop residue cover regressions. We found that neither windshield-based survey method was able to produce reliable regressions but that they can produce reasonable distinctions between low-residue and high-residue fields. On the other hand, both photo analysis and retroactively binned photo analysis survey data were able to produce reliable regressions with r2 values of 0.57 and 0.56, respectively. Overall, this study demonstrates that photo analysis surveys are the most reliable dataset to use when creating crop residue cover models, but we also acknowledge that these surveys are expensive to conduct and suggest some ways these surveys could be made more efficient in the future.

Self-supervised learning for chest computed tomography: training strategies and effect on downstream applications.

Self-supervised pre-training can reduce the amount of labeled training data needed by pre-learning fundamental visual characteristics of the medical imaging data. We investigate several self-supervised training strategies for chest computed tomography exams and their effects on downstream applications. We benchmark five well-known self-supervision strategies (masked image region prediction, next slice prediction, rotation prediction, flip prediction, and denoising) on 15M chest computed tomography (CT) slices collected from four sites of the Mayo Clinic enterprise, United States. These models were evaluated for two downstream tasks on public datasets: pulmonary embolism (PE) detection (classification) and lung nodule segmentation. Image embeddings generated by these models were also evaluated for prediction of patient age, race, and gender to study inherent biases in models' understanding of chest CT exams. The use of pre-training weights especially masked region prediction-based weights, improved performance, and reduced computational effort needed for downstream tasks compared with task-specific state-of-the-art (SOTA) models. Performance improvement for PE detection was observed for training dataset sizes as large as with a maximum gain of 5% over SOTA. The segmentation model initialized with pre-training weights learned twice as fast as the randomly initialized model. While gender and age predictors built using self-supervised training weights showed no performance improvement over randomly initialized predictors, the race predictor experienced a 10% performance boost when using self-supervised training weights. We released self-supervised models and weights under an open-source academic license. These models can then be fine-tuned with limited task-specific annotated data for a variety of downstream imaging tasks, thus accelerating research in biomedical imaging informatics.

Investigating the Fairness of Deep Learning Models in Breast Cancer Diagnosis Based on Race and Ethnicity

Breast cancer is the leading cancer affecting women globally. Despite deep learning models making significant strides in diagnosing and treating this disease, ensuring fair outcomes across diverse populations presents a challenge, particularly when certain demographic groups are underrepresented in training datasets. Addressing the fairness of AI models across varied demographic backgrounds is crucial. This study analyzes demographic representation within the publicly accessible Emory Breast Imaging Dataset (EMBED), which includes de-identified mammography and clinical data. We spotlight the data disparities among racial and ethnic groups and assess the biases in mammography image classification models trained on this dataset, specifically ResNet-50 and Swin Transformer V2. Our evaluation of classification accuracies across these groups reveals significant variations in model performance, highlighting concerns regarding the fairness of AI diagnostic tools. This paper emphasizes the imperative need for fairness in AI and suggests directions for future research aimed at increasing the inclusiveness and dependability of these technologies in healthcare settings. Code is available at: https://github.com/kuanhuang0624/EMBEDFairModels.

Radian Scaling and Its Application to Enhance Electricity Load Forecasting in Smart Cities Against Concept Drift

In a real-world implementation, machine learning models frequently experience concept drift when forecasting the electricity load. This is due to seasonal changes influencing the scale, mean, and median values found in the input data, changing their distribution. Several methods have been proposed to solve this, such as implementing automated model retraining, feature engineering, and ensemble learning. The biggest drawback, however, is that they are too complex for simple implementation in existing projects. Since the drifted data follow the same pattern as the training dataset in terms of having different scale, mean, and median values, radian scaling was proposed as a new way to scale without relying on these values. It works by converting the difference between the two sequential values into a radian for the model to compute, removing the bounding, and allowing the model to forecast beyond the training dataset scale. In the experiment, not only does the constrained gated recurrent unit model with radian scaling have shorter average training epochs, but it also lowers the average root mean square error from 158.63 to 43.375, outperforming the best existing normalization method by 72.657%.

Deep-GB: A novel deep learning model for globular protein prediction using CNN-BiLSTM architecture and enhanced PSSM with trisection strategy.

Globular proteins (GPs) play vital roles in a wide range of biological processes, encompassing enzymatic catalysis and immune responses. Enzymes, among these globular proteins, facilitate biochemical reactions, while others, such as haemoglobin, contribute to essential physiological functions such as oxygen transport. Given the importance of these considerations, accurately identifying Globular proteins is essential. To address the need for precise GP identification, this research introduces an innovative approach that employs a hybrid-based deep learning model called Deep-GP. We generated two datasets based on primary sequences and developed a novel feature descriptor called, Consensus Sequence-based Trisection-Position Specific Scoring Matrix (CST-PSSM). The model training phase involved the application of deep learning techniques, including the bidirectional long short-term memory network (BiLSTM), gated recurrent unit (GRU), and convolutional neural network (CNN). The BiLSTM and CNN were hybridised for ensemble learning. The CST-PSSM-based ensemble model achieved the most accurate predictive outcomes, outperforming other competitive predictors across both training and testing datasets. This demonstrates the potential of harnessing deep learning for precise GB prediction as a robust tool to expedite research, streamline drug discovery, and unveil novel therapeutic targets.

A Comprehensive Comparative Study of Intuitive Physics Modeling in Machine Learning Trained with Cartoon and Realistic Data

Abstract. This study delves into the influence of training data typesspecifically cartoon versus realistic visual datasetson the development of intuitive physics modeling in machine learning. Intuitive physics, the inherent human ability to understand and predict the physical properties and dynamics of objects, presents a significant challenge for current AI systems to replicate accurately. Leveraging YOLOv5, a cutting-edge object detection model, this research systematically evaluates the cognitive understanding and performance of AI models trained on distinct types of visual data. The findings reveal that the visual complexity inherent in the training datasets plays a crucial role in shaping the model's ability to generalize and accurately perform intuitive physics tasks. Models trained on cartoon datasets exhibited different learning patterns and generalization capabilities compared to those trained on realistic data, providing valuable insights into the role of data representation in AI training. This research offers both theoretical advancements in understanding AI's cognitive limitations and practical guidance for designing AI systems that can interact with the physical world more effectively. Ultimately, the study contributes to bridging the gap between human cognition and machine learning, pushing the boundaries of what AI can achieve in modeling complex, real-world phenomena.

Deep learning for 3D vascular segmentation in hierarchical phase contrast tomography: a case study on kidney

Automated blood vessel segmentation is critical for biomedical image analysis, as vessel morphology changes are associated with numerous pathologies. Still, precise segmentation is difficult due to the complexity of vascular structures, anatomical variations across patients, the scarcity of annotated public datasets, and the quality of images. Our goal is to provide a foundation on the topic and identify a robust baseline model for application to vascular segmentation using a new imaging modality, Hierarchical Phase-Contrast Tomography (HiP-CT). We begin with an extensive review of current machine-learning approaches for vascular segmentation across various organs. Our work introduces a meticulously curated training dataset, verified by double annotators, consisting of vascular data from three kidneys imaged using HiP-CT as part of the Human Organ Atlas Project. HiP-CT pioneered at the European Synchrotron Radiation Facility in 2020, revolutionizes 3D organ imaging by offering a resolution of around 20 μm/voxel and enabling highly detailed localised zooms up to 1–2 μm/voxel without physical sectioning. We leverage the nnU-Net framework to evaluate model performance on this high-resolution dataset, using both known and novel samples, and implementing metrics tailored for vascular structures. Our comprehensive review and empirical analysis on HiP-CT data sets a new standard for evaluating machine learning models in high-resolution organ imaging. Our three experiments yielded Dice similarity coefficient (DSC) scores of 0.9523, 0.9410, and 0.8585, respectively. Nevertheless, DSC primarily assesses voxel-to-voxel concordance, overlooking several crucial characteristics of the vessels and should not be the sole metric for deciding the performance of vascular segmentation. Our results show that while segmentations yielded reasonably high scores-such as centerline DSC ranging from 0.82 to 0.88, certain errors persisted. Specifically, large vessels that collapsed due to the lack of hydrostatic pressure (HiP-CT is an ex vivo technique) were segmented poorly. Moreover, decreased connectivity in finer vessels and higher segmentation errors at vessel boundaries were observed. Such errors, particularly in significant vessels, obstruct the understanding of the structures by interrupting vascular tree connectivity. Our study establishes the benchmark across various evaluation metrics, for vascular segmentation of HiP-CT imaging data, an imaging technology that has the potential to substantively shift our understanding of human vascular networks.

Use of biomarkers of metals to improve prediction performance of cardiovascular disease mortality

BackgroundWhether including additional environmental risk factors improves cardiovascular disease (CVD) prediction is unclear. We attempted to improve CVD mortality prediction performance beyond traditional CVD risk factors by additionally using metals measured in the urine and blood and with statistical machine learning methods.MethodsOur sample included 7,085 U.S. adults aged 40 years or older from the National Health and Nutrition Examination Survey 2003–2004 through 2015–2016, linked with the National Death Index through December 31, 2019. Data were randomly split into a 50/50 training dataset used to construct CVD mortality prediction models (n = 3542) and testing dataset used as validation to assess prediction performance (n = 3543). Relative to the traditional risk factors (age, sex, race/ethnicity, smoking status, systolic blood pressure, total and high-density lipoprotein cholesterol, hypertension, and diabetes), we compared models with an additional 17 blood and urinary metal concentrations. To build the prediction models, we used Cox proportional hazards, elastic-net (ENET) penalized Cox, and random survival forest methods.Results420 participants died from CVD with 8.8 mean years of follow-up. Blood lead, cadmium, and mercury were associated (p < 0.005) with CVD mortality. Including these blood metals in a Cox model, initially containing only traditional risk factors, raised the C-index from 0.845 to 0.847. Additionally, the Net Reclassification Index showed that 23% of participants received a more accurate risk prediction. Further inclusion of urinary metals improved risk reclassification but not risk discrimination.ConclusionsIncorporating blood metals slightly improved CVD mortality risk discrimination, while blood and urinary metals enhanced risk reclassification, highlighting their potential utility in improving cardiovascular risk assessments.

Prediction of viral oncoproteins through the combination of generative adversarial networks and machine learning techniques

Viral oncoproteins play crucial roles in transforming normal cells into cancer cells, representing a significant factor in the etiology of various cancers. Traditionally, identifying these oncoproteins is both time-consuming and costly. With advancements in computational biology, bioinformatics tools based on machine learning have emerged as effective methods for predicting biological activities. Here, for the first time, we propose an innovative approach that combines Generative Adversarial Networks (GANs) with supervised learning methods to enhance the accuracy and generalizability of viral oncoprotein prediction. Our methodology evaluated multiple machine learning models, including Random Forest, Multilayer Perceptron, Light Gradient Boosting Machine, eXtreme Gradient Boosting, and Support Vector Machine. In ten-fold cross-validation on our training dataset, the GAN-enhanced Random Forest model demonstrated superior performance metrics: 0.976 accuracy, 0.976 F1 score, 0.977 precision, 0.976 sensitivity, and 1.0 AUC. During independent testing, this model achieved 0.982 accuracy, 0.982 F1 score, 0.982 precision, 0.982 sensitivity, and 1.0 AUC. These results establish our new tool, VirOncoTarget, accessible via a web application. We anticipate that VirOncoTarget will be a valuable resource for researchers, enabling rapid and reliable viral oncoprotein prediction and advancing our understanding of their role in cancer biology.

Application of Prompt Learning Models in Identifying the Collaborative Problem Solving Skills in an Online Task

Collaborative problem solving (CPS) competence is considered one of the essential 21st-century skills. To facilitate the assessment and learning of CPS competence, researchers have proposed a series of frameworks to conceptualize CPS and explored ways to make sense of the complex processes involved in collaborative problem solving. However, encoding explicit behaviors into subskills within the frameworks of CPS skills is still a challenging task. Traditional studies have relied on manual coding to decipher behavioral data for CPS, but such coding methods can be very time-consuming and cannot support real-time analyses. Scholars have begun to explore approaches for constructing automatic coding models. Nevertheless, the existing models built using machine learning or deep learning techniques depend on a large amount of training data and have relatively low accuracy. To address these problems, this paper proposes a prompt-based learning pre-trained model. The model can achieve high performance even with limited training data. In this study, three experiments were conducted, and the results showed that our model not only produced the highest accuracy, macro F1 score, and kappa values on large training sets, but also performed the best on small training sets of the CPS behavioral data. The application of the proposed prompt-based learning pre-trained model contributes to the CPS skills coding task and can also be used for other CSCW coding tasks to replace manual coding.

Strictly Conservative Neural Implicits

AbstractWe describe a method to convert 3D shapes into neural implicit form such that the shape is approximated in a guaranteed conservative manner. This means the input shape is strictly contained inside the neural implicit or, alternatively, vice versa. Such conservative approximations are of interest in a variety of applications, including collision detection, occlusion culling, or intersection testing. Our approach is the first to guarantee conservativeness in this context of neural implicits. We support input given as mesh, voxel set, or implicit function. Adaptive affine arithmetic is employed in the neural network fitting process, enabling the reasoning over infinite sets of points despite using a finite set of training data. Combined with an interior point style optimization approach this yields the desired guarantee.

Towards Integrating Federated Learning with Split Learning via Spatio-temporal Graph Framework for Brain Disease Prediction.

Functional Magnetic Resonance Imaging (fMRI) is used for extracting blood oxygen signals from brain regions to map brain functional connectivity for brain disease prediction. Despite its effectiveness, fMRI has not been widely used: on the one hand, collecting and labeling the data is time-consuming and costly, which limits the amount of valid data collected at a single healthcare site; on the other hand, integrating data from multiple sites is challenging due to data privacy restrictions. To address these issues, we propose a novel, integrated Federated learning and Split learning Spatio-temporal Graph framework (FS2G). Specifically, we introduce federated learning and split learning techniques to split a spatio-temporal model into a client temporal model and a server spatial model. In the client temporal model, we propose a time-aware mechanism to focus on changes in brain functional states and use an InceptionTime model to extract information about changes in the brain states of each subject. In the server spatial model, we propose a united graph convolutional network to integrate multiple graph convolutional networks. Integrating federated learning and split learning, FS2G can utilize multi-site fMRI data without violating data privacy protection and reduce the risk of overfitting as it is capable of learning from limited training data sets. Moreover, it boosts the extraction of spatio-temporal features of fMRI using spatio-temporal graph networks. Experiments on ABIDE and ADHD200 datasets demonstrate that our proposed method outperforms state-of-the-art methods. In addition, we explore biomarkers associated with brain disease prediction using community discovery algorithms using intermediate results of FS2G. The source code is available at https://github.com/yutian0315/FS2G.

COVID-19 pandemic microplastics environmental impacts predicted by deep random forest (DRF) predictive model

BackgroundMicroplastic pollution is a pressing issue with far-reaching environmental and public health consequences. This study delves into the intricacies of predicting microplastic pollution during the COVID-19 pandemic in Tehran, Iran.MethodsThe research introduces a rigorous comparative analysis that evaluates the predictive prowess of the Deep Random Forest algorithm and established benchmarks, such as Random Forest, Decision Trees, Gradient Boosting, AdaBoost, and Support Vector Machine. The evaluation process encompasses a meticulous 70–30 training–testing split of the main data set. Performance is assessed by analysis metrics, including ROC and statistical errors. The primary data set encompasses distinct categories, including household wastes, hospital wastes, clinics wastes, and unknown-originated susceptible waste which is categorized in Infected items, PPEs, SUPs, Test kits, Medical packages, Unknown-originated pandemic mircoplastic waste. Deliberately, this data set was partitioned into training and testing subsets, ensuring the robustness and reliability of subsequent analyses. Approximately 70% of the main database was allocated to the training data set, with the remaining 30% constituting the testing data set.ResultsThe findings underscore the proposed algorithm’s supremacy, boasting an impressive AUC = 0.941. This exceptional score reflects the model’s precision in categorizing microplastics. These results have profound implications for environmental management and public health during pandemics.ConclusionsThe study positions the proposed model as a potent tool for microplastic pollution prediction, encouraging further research to refine predictive models and tap into new data sources for a more comprehensive understanding of microplastic dynamics in urban settings.

DeepLigType: Predicting Ligand Types of Protein-Ligand Binding Sites Using a Deep Learning Model.

The analysis of protein-ligand binding sites plays a crucial role in the initial stages of drug discovery. Accurately predicting the ligand types that are likely to bind to protein-ligand binding sites enables more informed decision making in drug design. Our study, DeepLigType, determines protein-ligand binding sites using Fpocket and then predicts the ligand type of these pockets with the deep learning model, Convolutional Block Attention Module (CBAM) with ResNet. CBAM-ResNet has been trained to accurately predict five distinct ligand types. We classified protein-ligand binding sites into five different categories according to the type of response ligands cause when they bind to their target proteins, which are antagonist, agonist, activator, inhibitor, and others. We created a novel dataset, referred to as LigType5, from the widely recognized PDBbind and scPDB dataset for training and testing our model. While the literature mostly focuses on the specificity and characteristic analysis of protein binding sites by experimental (laboratory-based) methods, we propose a computational method with the DeepLigType architecture. DeepLigType demonstrated an accuracy of 74.30% and an AUC of 0.83 in ligand type prediction on a novel test dataset using the CBAM-ResNet deep learning model. For access to the code implementation of this research, please visit our GitHub repository at https://github.com/drorhunvural/DeepLigType.