Multi-class Model Research Articles

Abstract B065: FabricaTM: A large-scale data simulation platform isolates tumor signal from cell-free DNA and improves tissue of origin prediction accuracy

Abstract Cell-free DNA (cfDNA) liquid biopsy is a promising non-invasive method for disease detection, but data availability and the complexity of cfDNA composition pose barriers to model development. Cancer prediction models trained on cfDNA data from clinically collected blood samples often struggle to isolate tumor-derived signals from confounding factors, and tissue of origin (TOO) models suffer from small sample sizes of rarer tumor types. To address these challenges, we developed FabricaTM, a platform to generate diverse, large-scale, high-fidelity simulated cfDNA datasets for robust and accurate machine learning (ML) model training. A key advantage of FabricaTM is its ability to enhance ML models by matching confounding variables between non-cancer and cancer samples. We first synthesized age-balanced non-cancer samples, then simulated tumor DNA shedding into circulation from 502 reference biopsy samples across 16 indications, optimizing for final tumor content distributions ranging from 0.001 to 14.2%. All reference non-cancer and biopsy samples consisted of aligned bisulfite sequencing reads from a custom target hybrid capture panel. This approach expanded a non-cancer dataset of 177 samples to 1,212 simulated samples and generated 6,400 simulated cancer cfDNA samples matching the non-cancers in age distribution. We then assessed this dataset for equivalence with clinical data and ability to improve cancer prediction. Non-linear dimensionality reduction and clustering of methylation signal showed simulated samples cluster indistinguishably from 2,290 samples (1,149 non-cancer, 1,141 treatment-naïve cancer) from the CORE-HH clinical study (NCT05435066). A binary cancer prediction model trained on the FabricaTM-generated dataset and evaluated on the CORE-HH dataset showed reduced reliance on age-associated signal (R2=0.15) and increased tuning towards tumor content, with minor cost to sensitivity (<4% at 95% specificity), relative to a model trained and cross-validated on the CORE-HH data (R2=0.60). To evaluate the benefit of these data for TOO prediction, we trained multiclass TOO models using FabricaTM's tumor biopsy reference data with and without the addition of our simulated data for 10 target indication groupings and benchmarked their performance on our clinical dataset. Training with both data types yielded a 10% increase in balanced accuracy. Moreover, peak performance was achieved at different training sample numbers per indication, illustrating FabricaTM's potential to estimate sample size requirements during study design. Our findings suggest FabricaTM-generated data can augment limited datasets to train ML models to identify tumor-specific biomarkers obscured by technical and demographic biases in real-world data. The age-balancing approach is readily applied to other confounders to help models learn true disease signatures. The platform is also adaptable to other diseases and biofluids (e.g., Alzheimer's disease, urine), allowing for extension to diagnostic and screening development beyond cancer and blood across the care spectrum. Citation Format: Kade Pettie, Shiva Farashahi, Jackson Killian, Dorna Kashef, Josh Hubbell, Feras Hantash, Jocelyn Charlton, Kieran Chacko. FabricaTM: A large-scale data simulation platform isolates tumor signal from cell-free DNA and improves tissue of origin prediction accuracy [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr B065.

NIMG-76. INCORPORATING IMAGING FEATURES OF NORMAL BRAIN TO ENHANCE SPATIALLY DIFFERENTIATING TREATMENT-INDUCED EFFECTS FROM RECURRENT GLIOMA WITH AI

Abstract INTRODUCTION Although AI-based approaches have been applied to discriminate between tumor recurrence (rTumor) and treatment-induced effects (TxE) from chemoradiotherapy in patients with recurrent glioma, these models typically either do not account for within-lesion mixtures of rTumor and TxE or generalize to independent test-sets with >80% accuracy. We hypothesize that incorporating imaging features of normal-appearing brain (NAB) into existing AI-based models that leverage tissue-samples with known locations on MRI and physiological imaging will significantly improve performance for distinguishing regions of TxE from rTumor and allow for visualization of rTumor beyond the T2-lesion. METHODS Using 254 pathology-confirmed tissue-samples with coordinates on diffusion-weighted, perfusion-weighted, and anatomical MRI from 135 glioma suspected of progression, we modified previously-developed binary, AI-based prediction models to learn multi-class targets consisting of rTumor, TxE, and contralateral NAB-regions (3/patient) with imaging signatures of CSF, white-matter, and grey-matter (659 total 10mm-isotropic samples). The multi-class ensemble model was trained with 4 unique repeats of constrained-5-fold-cross-validation and evaluated on a held-out test set (30 patients). This process was repeated 20 times, and the performance was compared to corresponding binary AI-based models. RESULTS NAB was easiest to distinguish from individual and combined classes (AUC-ROCs>0.98). Multi-class model performance for predicting TxE increased by 27% to 0.93±0.03 (p<0.00001) when discriminating between NAB and rTumor as compared to just rTumor with the binary models. While TxE vs rTumor alone was most difficult to predict with the multi-class model (AUC-ROC=0.73) achieving equivalent performance to binary models, there was 67% less variance in test performance. This approach allowed for the quantification of whole-brain probability maps of rTumor, TxE, and NAB along with a combined 3-channel colormap, facilitating the visualization of rTumor predictions beyond the T2-lesion. CONCLUSIONS Including NAB imaging features improved performance of AI-based models of rTumor/TxE, while expanding the predicted area to produce more interpretable whole-brain probability maps. Current work is evaluating their utility in prospectively guiding sampling of rTumor tissue and predicting survival.

A Novel Machine Learning Model and a Web Portal for Predicting the Human Skin Sensitization Effects of Chemical Agents

Skin sensitization is a significant concern for chemical safety assessments. Traditional animal assays often fail to predict human responses accurately, and ethical constraints limit the collection of human data, necessitating a need for reliable in silico models of skin sensitization prediction. This study introduces HuSSPred, an in silico tool based on the Human Predictive Patch Test (HPPT). HuSSPred aims to enhance the reliability of predicting human skin sensitization effects for chemical agents to support their regulatory assessment. We have curated an extensive HPPT database and performed chemical space analysis and grouping. Binary and multiclass QSAR models were developed with Bayesian hyperparameter optimization. Model performance was evaluated via five-fold cross-validation. We performed model validation with reference data from the Defined Approaches for Skin Sensitization (DASS) app. HuSSPred models demonstrated strong predictive performance with CCR ranging from 55 to 88%, sensitivity between 48 and 89%, and specificity between 37 and 92%. The positive predictive value (PPV) ranged from 84 to 97%, versus negative predictive value (NPV) from 22 to 65%, and coverage was between 75 and 93%. Our models exhibited comparable or improved performance compared to existing tools, and the external validation showed the high accuracy and sensitivity of the developed models. HuSSPred provides a reliable, open-access, and ethical alternative to traditional testing for skin sensitization. Its high accuracy and reasonable coverage make it a valuable resource for regulatory assessments, aligning with the 3Rs principles. The publicly accessible HuSSPred web tool offers a user-friendly interface for predicting skin sensitization based on chemical structure.

Topological embedding and directional feature importance in ensemble classifiers for multi-class classification

Cancer is the second leading cause of disease-related death worldwide, and machine learning-based identification of novel biomarkers is crucial for improving early detection and treatment of various cancers. A key challenge in applying machine learning to high-dimensional data is deriving important features in an interpretable manner to provide meaningful insights into the underlying biological mechanismsWe developed a class-based directional feature importance (CLIFI) metric for decision tree methods and demonstrated its use for the The Cancer Genome Atlas proteomics data. The CLIFI metric was incorporated into four algorithms, Random Forest (RF), LAtent VAriable Stochastic Ensemble of Trees (LAVASET), and Gradient Boosted Decision Trees (GBDTs), and a new extension incorporating the LAVA step into GBDTs (LAVABOOST). Both LAVA methods incorporate topological information from protein interactions into the decision function.The different models' performance in classifying 28 cancers resulted in F1-scores of 92.6% (RF), 92.0% (LAVASET), 89.3% (LAVABOOST) and 85.7% (GBDT), with no method outperforming all others for individual cancer type prediction. The CLIFI metric enables visualisation of the model's decision-making functions. The resulting CLIFI value distributions indicated heterogeneity in the expression of several proteins (MYH11, ERα, BCL2) across different cancer types (including brain glioma, breast, kidney, thyroid and prostate cancer) aligning with the original raw expression data.In conclusion, we have developed an integrated, directional feature importance metric for multi-class decision tree-based classification models that facilitates interpretable feature importance assessment. The CLIFI metric can be combined with incorporating topological information into the decision functions of models to introduce inductive bias, enhancing interpretability.

Deep Learning Approach for Fetal Health Prediction

Monitoring the foetus's health is crucial during pregnancy to avoid complications that may worsen the course of pregnancy and delivery. Cardiotocography (CTG) is a tool that provides complex information by monitoring the foetus's heart rate signal. Obstetricians visually interpret these signals to predict potential risks and draw clinical inferences. The interpretation, however, relies on the expertise of the obstetrician, leading to a significant false positive rate. Thus, the study uses deep learning techniques to effectively identify foetal health states as 'Normal', 'Suspect' and 'Pathological'. The dataset used is CTG data drawn from the Kaggle repository. The optimal subset of features is obtained by comparing traditional feature selection and meta-heuristic-based feature selection techniques. Deep learning algorithms, namely, Convolution Neural Networks, Artificial Neural Networks, and Radial Basis Function Networks, are applied to train multi-class classification models that predict foetal health status. The models are then evaluated using performance metrics.

Prediction of Dynamic Toxicity of Nanoparticles Using Machine Learning.

Predicting the toxicity of nanoparticles plays an important role in biomedical nanotechnologies, in particular in the creation of new drugs. Safety analysis of nanoparticles can identify potentially harmful effects on living organisms and the environment. Advanced machine learning models are used to predict the toxicity of nanoparticles in a nutrient solution. In this article, we performed a comparative analysis of the current state of research in the field of nanoparticle toxicity analysis using machine learning methods; we trained a regression model for predicting the quantitative toxicity of nanoparticles depending on their concentration in the nutrient solution at a fixed point in time with the achieved metrics values of MSE = 2.19 and RMSE = 1.48; we trained a multi-class classification model for predicting the toxicity class of nanoparticles depending on their concentration in the nutrient solution at a fixed point in time with the achieved metrics values of Accuracy = 0.9756, Recall = 0.9623, F1-Score = 0.9640, and Log Loss = 0.1855. As a result of the analysis, we concluded the good predictive ability of the trained models. The optimal dosages for the nanoparticles under study were determined as follows: ZnO = 9.5 × 10-5 mg/mL; Fe3O4 = 0.1 mg/mL; SiO2 = 1 mg/mL. The most significant features of predictive models are the diameter of the nanoparticle and the nanoparticle concentration in the nutrient solution.

Enhancing Adversarial Robustness through Stable Adversarial Training

Deep neural network models are vulnerable to attacks from adversarial methods, such as gradient attacks. Evening small perturbations can cause significant differences in their predictions. Adversarial training (AT) aims to improve the model’s adversarial robustness against gradient attacks by generating adversarial samples and optimizing the adversarial training objective function of the model. Existing methods mainly focus on improving robust accuracy, balancing natural and robust accuracy and suppressing robust overfitting. They rarely consider the AT problem from the characteristics of deep neural networks themselves, such as the stability properties under certain conditions. From a mathematical perspective, deep neural networks with stable training processes may have a better ability to suppress overfitting, as their training process is smoother and avoids sudden drops in performance. We provide a proof of the existence of Ulam stability for deep neural networks. Ulam stability not only determines the existence of the solution for an operator inequality, but it also provides an error bound between the exact and approximate solutions. The feature subspace of a deep neural network with Ulam stability can be accurately characterized and constrained by a function with special properties and a controlled error boundary constant. This restricted feature subspace leads to a more stable training process. Based on these properties, we propose an adversarial training framework called Ulam stability adversarial training (US-AT). This framework can incorporate different Ulam stability conditions and benchmark AT models, optimize the construction of the optimal feature subspace, and consistently improve the model’s robustness and training stability. US-AT is simple and easy to use, and it can be easily integrated with existing multi-class AT models, such as GradAlign and TRADES. Experimental results show that US-AT methods can consistently improve the robust accuracy and training stability of benchmark models.

LMCD-OR: a large-scale, multilevel categorized diagnostic dataset for oral radiography

In recent years, digital dentistry has increasingly utilized advanced image analysis techniques, such as image classification and disease diagnosis, to improve clinical outcomes. Despite these advances, the lack of comprehensive benchmark datasets is a significant barrier. To address this gap, our research team develop LMCD-OR, a substantial collection of oral radiograph images designed to support extensive artificial intelligence (AI)-driven diagnostics. LMCD-OR comprises 3,818 digital imaging and communications in medicine (DICOM) oral X-ray images from local medical institutions that are meticulously annotated to provide broad category information for both primary dental outpatient services and detailed secondary disease diagnoses. This dataset is engineered to train and validate multiclassification models to improve the precision and scope of oral disease diagnostics. To ensure robust dataset validation, we employ four cutting-edge visual neural network classification models as benchmarks. These models are tested against rigorous performance metrics, demonstrating the ability of the dataset to support advanced image classification and disease diagnosis tasks. LMCD-OR is publicly available at http://dentaldataset.zeroacademy.net.

Hyperparameter selection for dataset-constrained semantic segmentation: Practical machine learning optimization.

This paper provides a pedagogical example for systematic machine learning optimization in small dataset image segmentation, emphasizing hyperparameter selections. A simple process is presented for medical physicists to examine hyperparameter optimization. This is also applied to a case-study, demonstrating the benefit of the method. An unrestricted public Computed Tomography (CT) dataset, with binary organ segmentation, was used to develop a multiclass segmentation model. To start the optimization process, a preliminary manual search of hyperparameters was conducted and from there a grid search identified the most influential result metrics. A total of 658 different models were trained in 2100h, using 13160 effective patients. The quantity of results was analyzed using random forest regression, identifying relative hyperparameter impact. Metric implied segmentation quality (accuracy 96.8%, precision 95.1%) and visual inspection were found to be mismatched. In this work batch normalization was most important, but performance varied with hyperparameters and metrics selected. Targeted grid-search optimization and random forest analysis of relative hyperparameter importance, was an easily implementable sensitivity analysis approach. The proposed optimization method gives a systematic and quantitative approach to something intuitively understood, that hyperparameters change model performance. Even just grid search optimization with random forest analysis presented here can be informative within hardware and data quality/availability limitations, adding confidence to model validity and minimize decision-making risks. By providing a guided methodology, this work helps medical physicists to improve their model optimization, irrespective of specific challenges posed by datasets and model design.

Equipping computational pathology systems with artifact processing pipelines: a showcase for computation and performance trade-offs

BackgroundHistopathology is a gold standard for cancer diagnosis. It involves extracting tissue specimens from suspicious areas to prepare a glass slide for a microscopic examination. However, histological tissue processing procedures result in the introduction of artifacts, which are ultimately transferred to the digitized version of glass slides, known as whole slide images (WSIs). Artifacts are diagnostically irrelevant areas and may result in wrong predictions from deep learning (DL) algorithms. Therefore, detecting and excluding artifacts in the computational pathology (CPATH) system is essential for reliable automated diagnosis.MethodsIn this paper, we propose a mixture of experts (MoE) scheme for detecting five notable artifacts, including damaged tissue, blur, folded tissue, air bubbles, and histologically irrelevant blood from WSIs. First, we train independent binary DL models as experts to capture particular artifact morphology. Then, we ensemble their predictions using a fusion mechanism. We apply probabilistic thresholding over the final probability distribution to improve the sensitivity of the MoE. We developed four DL pipelines to evaluate computational and performance trade-offs. These include two MoEs and two multiclass models of state-of-the-art deep convolutional neural networks (DCNNs) and vision transformers (ViTs). These DL pipelines are quantitatively and qualitatively evaluated on external and out-of-distribution (OoD) data to assess generalizability and robustness for artifact detection application.ResultsWe extensively evaluated the proposed MoE and multiclass models. DCNNs-based MoE and ViTs-based MoE schemes outperformed simpler multiclass models and were tested on datasets from different hospitals and cancer types, where MoE using (MobileNet) DCNNs yielded the best results. The proposed MoE yields 86.15 % F1 and 97.93% sensitivity scores on unseen data, retaining less computational cost for inference than MoE using ViTs. This best performance of MoEs comes with relatively higher computational trade-offs than multiclass models. Furthermore, we apply post-processing to create an artifact segmentation mask, a potential artifact-free RoI map, a quality report, and an artifact-refined WSI for further computational analysis. During the qualitative evaluation, field experts assessed the predictive performance of MoEs over OoD WSIs. They rated artifact detection and artifact-free area preservation, where the highest agreement translated to a Cohen Kappa of 0.82, indicating substantial agreement for the overall diagnostic usability of the DCNN-based MoE scheme.ConclusionsThe proposed artifact detection pipeline will not only ensure reliable CPATH predictions but may also provide quality control. In this work, the best-performing pipeline for artifact detection is MoE with DCNNs. Our detailed experiments show that there is always a trade-off between performance and computational complexity, and no straightforward DL solution equally suits all types of data and applications. The code and HistoArtifacts dataset can be found online at Github and Zenodo, respectively.

Identifying patient subgroups in MASLD and MASH-associated fibrosis: molecular profiles and implications for drug development

The incidence of MASLD and MASH-associated fibrosis is rapidly increasing worldwide. Drug therapy is hampered by large patient variability and partial representation of human MASH fibrosis in preclinical models. Here, we investigated the mechanisms underlying patient heterogeneity using a discovery dataset and validated in distinct human transcriptomic datasets, to improve patient stratification and translation into subgroup specific patterns. Patient stratification was performed using weighted gene co-expression network analysis (WGCNA) in a large public transcriptomic discovery dataset (n = 216). Differential expression analysis was performed using DESeq2 to obtain differentially expressed genes (DEGs). Ingenuity Pathway analysis was used for functional annotation. The discovery dataset showed relevant fibrosis-related mechanisms representative of disease heterogeneity. Biological complexity embedded in genes signature was used to stratify discovery dataset into six subgroups of various sizes. Of note, subgroup-specific DEGs show differences in directionality in canonical pathways (e.g. Collagen biosynthesis, cytokine signaling) across subgroups. Finally, a multiclass classification model was trained and validated in two datasets. In summary, our work shows a potential alternative for patient population stratification based on heterogeneity in MASLD-MASH mechanisms. Future research is warranted to further characterize patient subgroups and identify protein targets for virtual screening and/or in vitro validation in preclinical models.

Machine learning-based classification of varicocoele grading: A promising approach for diagnosis and treatment optimization.

Varicocoele is a correctable cause of male infertility. Although physical examination is still being used in diagnosis and grading, it gives conflicting results when compared to ultrasonography-based varicocoele grading. We aimed to develop a multi-class machine learning model for the grading of varicocoeles based on ultrasonographic measurements. Between January and May 2024, we enrolled unilateral varicocoele patients at an infertility clinic, assessing their varicocoele stages using the Dubin and Amelar system. We measured vascular diameter and reflux time at the testicular apex and the subinguinal region ultrasonography in both the supine and standing positions. Using these measurements, we developed four multi-class machine learning models, evaluating their performance metrics and determining which patient position and projection were most influential in varicocoele grading. We included 248 patients with unilateral varicocoele in the study, their average age was 26.61±4.95 years old. Of these, 212 had left-sided and 36 had right-sided varicocoeles. According to the Dubin and Amelar system, there were 66 grade I, 96 grade II, and 86 grade III varicocoeles. Among the models we created, the random forest (RF) model performed best, with an overall accuracy of 0.81±0.06, an F1 score of 0.79±0.02, a sensitivity of 0.69±0.02, and a specificity of 0.8±0.03. Vascular diameter measurement at the testicular apex in the supine position had the most impact on grading across all models. In support vector machine and multi-layer perceptron models, reflux time measurements from the subinguinal projection in the standing position contributed the most, while in RF and k-nearest neighbors models, measurements from the subinguinal projection in the supine position were the most influential. Machine learning methods have demonstrated superior accuracy in predicting disease compared to traditional statistical regressions and nomograms. These advancements hold promise for clinically automated prediction of varicocoele grades in patients. Tailored varicocoele grading for individuals has the potential to enhance treatment effectiveness and overall quality of life.

Is Automated Machine Learning useful for ocular toxoplasmosis identification and classification of the inflammatory activity?

PurposeTo evaluate the performance of automated machine learning (AutoML) models in diagnosing ocular toxoplasmosis (OT) and classifying its inflammatory activity from fundus photographs. DesignCross-sectional study. MethodsFundus photographs from OT patients in two Colombian referral centers and an open-source OT database were classified into active OT, inactive OT, and no OT. Image quality assessment excluded images with artifacts but included blurry images due to vitritis. Photos were uploaded to Amazon Web Services S3 and Google Cloud Bucket. Two models were developed on each platform: a binary model (active/inactive OT vs. no OT) and a multiclass model (active OT, inactive OT, and no OT). Datasets were split into 70% for training, 20% for testing, and 10% for validation. Sensitivity, specificity, precision, accuracy, F1-score, the area under the precision-recall curve (AUPRC), and Cohen's Kappa were calculated for each platform and model. An external validation using an open-source Image bank was performed. ResultsThe binary model on AWS showed a sensitivity of 0.97, specificity of 0.98, and AUPRC of 1.00, while the Google Cloud binary model had a sensitivity of 0.82, specificity of 0.91, and AUPRC of 0.91. The multiclass model on AWS achieved an F1 score of 0.88, with Cohen's Kappa of 0.81, while the Google Cloud model reached an F1 score of 0.88, with Cohen's Kappa of 0.82. External validation for Google Cloud achieved an accuracy of 87.5% and 80.3% in the binary and multiclass models, respectively ConclusionsAutoML is a powerful tool for diagnosing OT and classifying inflammatory activity, potentially guiding diagnosis and treatment decisions.

Deep Learning to Predict the Future Growth of Geographic Atrophy from Fundus Autofluorescence

PurposeThe region of growth (ROG) of geographic atrophy (GA) throughout the macular area has an impact on visual outcomes. Here, we developed multiple deep learning (DL) models to predict the 1-year ROG of GA lesions using fundus autofluorescence (FAF) images. DesignIn this retrospective analysis, 3 types of models were developed using FAF images collected 6 months after baseline to predict the GA lesion area (segmented lesion mask) at 1.5 years, FAF images collected at baseline and 6 months to predict the GA lesion at 1.5 years, and FAF images collected 6 months after baseline to predict the GA lesion at 1 and 1.5 years. The 1-year ROG from the 6-month visit was derived by taking the difference between the GA lesion area (segmented lesion mask) at the 1.5-year and 6-month visits. ParticipantsPatients enrolled in the following lampalizumab clinical trials and prospective observational studies: NCT02247479, NCT02247531, NCT02479386, and NCT02399072. MethodsDatasets of study eyes from 597 patients were split into model training (310), validation (78), and test sets (209), stratified by baseline or initial lesion area, lesion growth rate, foveal involvement, and focality. DL experiments were performed using the 2-dimensional U-Net, whole-lesion and multiclass models were developed. Main Outcome MeasuresThe performance of the models was evaluated by calculating the Dice score, coefficient of determination (R2), and the squared Pearson correlation coefficient (r2) between the true and derived GA lesion 1-year ROG. ResultsThe model using baseline and 6-month FAF images to predict GA lesion enlargement at 1.5 years had the best performance for the derived 1-year ROG. Mean Dice scores were 0.73, 0.68, and 0.70, respectively, in the training, validation, and test sets. The R2 (0.77, 0.53, and 0.79) and r2 (0.83, 0.61, and 0.79) showed similar trends across the 3 sets. ConclusionsThese findings show the potential of using baseline and/or 6-month visit FAF images to predict 1-year GA ROG using a DL approach. This work could potentially help support decision making in clinical trials and more informed treatment decisions in clinical practice.

Unveiling the Effectiveness of Multi-Classification Algorithms: A Comprehensive Investigation

Multi classes portray alishead to a ton of genuine computer-based intelligence applications that need the ability to perceive numerous different classes immediately. In multiclass arrangement issues, we really want to manage multiple classes which imply the calculation which we are utilizing ought to be equipped for working with different classes. The rear different models accessible for this and a few strategies are likewise accessible that can make support vector machines and Calculated relapse equipped for managing multiple classes. In this examination paper, we present a multiclass grouping strategy from AI (ML) to foresee the six classes of Dermatology sickness expectation. In particular, the One-Refrains One (OVO) and One-Sections Rest (OVR) systems are assessed under Support Vector Machine (SVM) and Logistic Regression (LR) calculations. The current review intends store cognize fitting finding class for dermatology patients through building a multi-class expectation model. Our exploratory outcomes showed that the overall model is LR under the OVR technique accomplishing the most noteworthy Exactness 98.2% when contrasted with LR with OVO and SVM with OVR strategy accomplishing normal precision when contrasted with OVO. This proposed model shows the similar and promising outcomes that can upgrade the expectation execution model formulate classification.

Automatic sleep staging based on 24/7 EEG SubQ (UNEEG medical) data displays strong agreement with polysomnography in healthy adults

Goal and aimsPerformance evaluation of automatic sleep staging on two-channel subcutaneous electroencephalography. Focus technologyUNEEG medical’s 24/7 electroencephalography SubQ (the SubQ device) with deep learning model U-SleepSQ. Reference method/technologyManually scored hypnograms from polysomnographic recordings. SampleTwenty-two healthy adults with 1-6 recordings per participant. The clinical study was registered at ClinicalTrials.gov with the identifier NCT04513743. DesignFine-tuning of U-Sleep in 11-fold cross-participant validation on 22 healthy adults. The resultant model was called U-SleepSQ. Core analyticsBland-Altman analysis of sleep parameters. Advanced multiclass model performance metrics: stage-specific accuracy, specificity, sensitivity, kappa (κ), and F1 score. Additionally, Cohen’s κ coefficient and macro F1 score. Longitudinal and participant-level performance evaluation. Additional analytics and exploratory analysesExploration of model confidence quantification. Performance vs. age, sex, body mass index, SubQ implantation hemisphere, normalized entropy, transition index, and scores from the following three questionnaires: Morningness-Eveningness Questionnaire, World Health Organization’s 5-item Well-being Index, and Major Depression Inventory. Core outcomesThere was a strong agreement between the focus and reference method/technology. Important supplemental outcomesThe confidence score was a promising metric for estimating the reliability of each hypnogram classified by the system. Core conclusionThe U-SleepSQ model classified hypnograms for healthy participants soon after implantation and longitudinally with a strong agreement with the gold standard of manually scored polysomnographics, exhibiting negligible temporal variation.

Game-Based Assessment of Peripheral Neuropathy Combining Sensor-Equipped Insoles, Video Games, and AI: Proof-of-Concept Study.

Detecting peripheral neuropathy (PNP) is crucial in preventing complications such as foot ulceration. Clinical examinations for PNP are infrequently provided to patients at high risk due to restrictions on facilities, care providers, or time. A gamified health assessment approach combining wearable sensors holds the potential to address these challenges and provide individuals with instantaneous feedback on their health status. We aimed to develop and evaluate an application that assesses PNP through video games controlled by pressure sensor-equipped insoles. In the proof-of-concept exploratory cohort study, a complete game-based framework that allowed the study participant to play 4 video games solely by modulating plantar pressure values was established in an outpatient clinic setting. Foot plantar pressures were measured by the sensor-equipped insole and transferred via Bluetooth to an Android tablet for game control in real time. Game results and sensor data were delivered to the study server for visualization and analysis. Each session lasted about 15 minutes. In total, 299 patients with diabetes mellitus and 30 with metabolic syndrome were tested using the game application. Patients' game performance was initially assessed by hypothesis-driven key capabilities that consisted of reaction time, sensation, skillfulness, balance, endurance, and muscle strength. Subsequently, specific game features were extracted from gaming data sets and compared with nerve conduction study findings, neuropathy symptoms, or disability scores. Multiple machine learning algorithms were applied to 70% (n=122) of acquired data to train predictive models for PNP, while the remaining data were held out for final model evaluation. Overall, clinically evident PNP was present in 247 of 329 (75.1%) participants, with 88 (26.7%) individuals showing asymmetric nerve deficits. In a subcohort (n=37) undergoing nerve conduction study as the gold standard, sensory and motor nerve conduction velocities and nerve amplitudes in lower extremities significantly correlated with 79 game features (|R|>0.4, highest R value +0.65; P<.001; adjusted R2=0.36). Within another subcohort (n=173) with normal cognition and matched covariates (age, sex, BMI, etc), hypothesis-driven key capabilities and specific game features were significantly correlated with the presence of PNP. Predictive models using selected game features achieved 76.1% (left) and 81.7% (right foot) accuracy for PNP detection. Multiclass models yielded an area under the receiver operating characteristic curve of 0.76 (left foot) and 0.72 (right foot) for assessing nerve damage patterns (small, large, or mixed nerve fiber damage). The game-based application presents a promising avenue for PNP screening and classification. Evaluation in expanded cohorts may iteratively optimize artificial intelligence model efficacy. The integration of engaging motivational elements and automated data interpretation will support acceptance as a telemedical application.

Applications of Machine Learning: Classifying Wheat Seeds through Computer Neural Networks

In the past few decades, the incompetent processes of traditional mechanical seed sorting have given rise to new, innovative methods, along with the integration of smart technology in agriculture. The present study created a multi-classification machine learning model that classifies wheat seeds based on given characteristics by giving highly accurate predictions for future samples based on given data. Due to various uses of different types of seeds, effective sorting is an integral preparatory stage for food processing. The dataset used in the study included seven features (directly measured characteristics of seeds), three labels (types of seeds), and around two hundred examples. The seven features include area, perimeter, compactness, kernel length, kernel width, asymmetry coefficient, and kernel groove. The first model was created through neural networks trained on all seven characteristics, reaching an overall accuracy of 0.94. Implementing a decision tree method, feature selection was applied to the data, which distinguished kernel groove, kernel width, and asymmetry coefficient as the three most important features. A new model was constructed with only the three selected features, which reduced the risk of overfitting, eliminated noise, and yielded a higher accuracy (0.96) than the previous model. Both multi-classification models used the softmax activation function and results were compared; feature selection noticeably reduced any inaccuracies, including categorical cross entropy loss. The main incentive of the study was to develop an effective model to sort wheat seeds (to replace traditional mechanical sorting) and provide reasonable recommendations for practical use through an analysis of feature selection.

E-WebGuard: Enhanced neural architectures for precision web attack detection

Web applications have become a favored tool for organizations to disseminate vast amounts of information to the public. With the increasing adoption and inherent openness of these applications, there is an observed surge in web-based attacks exploited by adversaries. However, most of the web attack detection works are based on public datasets that are outdated or do not cover a sufficient quantity of web application attacks. Furthermore, most of them are binary detection (i.e., normal or attack) and there is little work on multi-class web attack detection. This highlights the crucial need for automated web attack detection models to bolster web security. In this study, a suite of integrated machine learning and deep learning models is designed to detect web attacks. Specifically, this study employs the Character-level Support Vector Machine (Char-SVM), Character-level Long Short-Term Memory (Char-LSTM), Convolutional Neural Network - SVM (CNN-SVM), and CNN-Bi-LSTM models to differentiate between standard HTTP requests and HTTP-based attacks in both the CSIC 2010 and SR-BH 2020 datasets. Note that the CSIC 2010 dataset involves binary classification, while the SR-BH 2020 dataset involves multi-class classification, specifically with 13 classes. Notably, the input data is first converted to the character level before being fed into any of the proposed model architectures. In the binary classification task, the Char-SVM model with a linear kernel outperforms other models, achieving an accuracy rate of 99.60%. The CNN-Bi-LSTM model closely follows with a 99.41% accuracy, surpassing the performance of the CNN-LSTM model presented in previous research. In the context of multi-class classification, the CNN-Bi-LSTM model demonstrates outstanding performance with a 99.63% accuracy rate. Furthermore, the multi-class classification models, namely Char-LSTM and CNN-Bi-LSTM, achieve validation accuracies above 98%, outperforming the two machine learning-based methods mentioned in the original research.

Unequal area facility layout problem considering transporters interaction– a queuing theory and machine learning approach

ABSTRACT This study presents a novel analytical framework that merges queueing theory with deep neural networks to optimize facility layout and transporter selection in manufacturing systems. It addresses critical factors such as stochastic service times of facilities, random demand, transporter capacity, speed, and transportation batch size. Three objectives are considered: minimizing material handling costs (MHC), work-in-process (WIP), and the interaction probability of transporters (IP). The latter objective focuses on reducing instances where transporters cross paths to prevent accidents or disruptions. WIP is computed using a multi-class open queueing network model, while IP is determined using a deep neural network. The model facilitates the identification of facilities and the assignment of suitable transporters, considering empty transporter travels to minimize MHC and WIP. Results from the model are compared to a simulation model for validation across various scenarios, demonstrating acceptable accuracy. Additionally, a multi-objective meta-heuristic optimization algorithm is employed to solve the model. The effectiveness of the optimization method is evaluated against other approaches, highlighting its applicability in enhancing manufacturing system performance.