Search Results (15)

We develop SAFIRE (Self-Attention Fuzzy-Inspired Rule Estimator), a differentiable fuzzy-inspired rule-scoring surrogate for binary medical tabular classification coupling multi-head self-attention, Gaussian membership functions, and Hard Concrete gates for continuous rule scoring. We position SAFIRE as a smooth surrogate of the discrete $L_0$-regularised rule-selection problem and establish five mathematical results and one complexity remark: (1) the relaxed objective is differentiable almost everywhere under positive Gaussian widths (enforced by a Softplus reparameterisation) and fixed batch-normalisation statistics; (2) the deterministic-inference active threshold is strictly stricter than the expected-nonzero training threshold, identifying Hard Concrete gates as continuous rule-scoring devices rather than automatic pruning mechanisms; (3) per-sample forward complexity identifies attention and rule layers as the dominant terms; (4) the Softplus–BatchNorm–linear rule operator violates all four triangular-norm axioms—with necessary and sufficient conditions per axiom and a no-finite-parameterisation impossibility result—while a Softplus reparameterisation restores coordinate-wise monotonicity; (5) a margin-based upper bound characterises disagreement between the full classifier and a top-k rule-only surrogate; and (6) the Softplus-reparameterised constrained variant is provably coordinate-wise monotone with explicit asymptotic regimes. Evaluated on four University of California, Irvine (UCI), medical binary tabular benchmarks under repeated stratified cross-validation, SAFIRE-Prog is statistically competitive with strong interpretable, modern, and gradient-boosting baselines, with one Bonferroni-significant gain over RuleFit on the Diabetic Retinopathy Debrecen corpus. The 48-configuration Hard Concrete sweep, constrained-variant comparison, and a top-k fidelity analysis (per-fold range 0.73–0.95) provide quantitative companion measurements for the mathematical framework. A supplementary large-scale hospital electronic health record (EHR) benchmark (Diabetes 130-US Hospitals, $n=101,766$) shows the rule-scoring mechanism scales to ∼${10}^5$ records and, under severe class imbalance, statistically matches gradient boosting on accuracy while significantly exceeding it on macro-F1. The results offer a mathematically auditable pathway towards interpretable, auditable rule scoring for medical tabular classification, with rule signatures defined in a projected latent space rather than over raw clinical variables. Full article

In many university and healthcare projects, models are built for very different data types such as tables, institutional time series, and medical images, but they are deployed as separate applications. In this work, that separation made testing and maintenance difficult because each module had its own pipeline and runtime requirements. This paper presents an integrated AI lakehouse-style implementation that runs three model pipelines inside one containerized backend. For medical imaging, we used MRI datasets from IEEE DataPort: a four-class classification set with 7012 images (5708 train/1304 test) and a segmentation set with 3063 image–mask pairs. The classification model (ResNet50 transfer learning) is evaluated using a proper train–validation–test protocol across multiple splits (80/10/10, 70/10/20, 60/10/30, and 10/30/60), achieving a test accuracy of 99.00% under the standard 80/10/10 split. Additionally, a patient-level evaluation is conducted using an external glioma dataset to provide a more realistic assessment without data leakage. The segmentation model (DeepLabV3-ResNet50) achieved 83.09% validation mIoU and 88.79% Dice score. For university KPI forecasting, we used annual IPEDS and NSF HERD data from 2010 to 2023 for three universities (BSU, EOU, and UAB). To examine the effect of preprocessing on forecasting performance, two case studies are conducted. In the first case, linear interpolation is applied to generate semester-level data. In the second case, the original annual data is used directly without interpolation. Random Forest regression and ARIMA models are evaluated using MAE, RMSE, MAPE, and $R^2$. The results showed that interpolation improved apparent forecasting performance due to smoothing, while evaluation on the original annual data provided a more realistic assessment of model behavior. To further validate the framework on a larger dataset, an additional case study is conducted using a student dropout dataset. For water potability, we trained and compared multiple tabular classifiers on a large dataset (1,048,575 samples). A Random Forest model (100 trees, max depth 10) achieved 85.86% test accuracy and high recall for unsafe samples (0.8447). All modules are served via FastAPI and deployed together using Docker, with workflow automation routing requests to the correct endpoint. System-level benchmarking indicates that the backend maintains stable throughput and latency under concurrent requests. Full article

Background/Objectives: Postoperative pulmonary complications (PPCs) following abdominal surgery are associated with prolonged hospitalization, delayed recovery, and increased mortality. Because nursing surveillance is essential for early detection and timely intervention, this study aimed to develop a holistic nursing surveillance decision support system integrating PPC risk prediction with structured nursing action recommendations. Methods: In this retrospective cohort study, electronic medical record (EMR) data from approximately 6900 adult patients who underwent abdominal surgery at a single institution between January 2015 and September 2023 were analyzed. The study protocol was approved by the Institutional Review Board, and the requirement for informed consent was waived because of the retrospective study design. PPC risk was predicted using a tabular multilayer perceptron (MLP) encoder with SHapley Additive exPlanations (SHAP)-based feature weighting and a random forest classification head optimized via Optuna. Class imbalance was addressed using weighted sampling, class weighting in BCE(Binary Cross Entropy) With Logits Loss, and decision-threshold optimization. For clinical decision support, a large language model generated structured nursing surveillance recommendations in an action–evidence–rationale JSON format and was aligned through supervised fine-tuning (SFT) using human-evaluated cases. Results: The prediction model achieved an AUROC of 0.810, with an accuracy of 0.811, precision of 0.547, and recall of 0.545. In expert evaluation, the SFT-aligned model improved recommendation quality, reducing incorrect nursing actions from 19.3% to 8.0%. Conclusions: The proposed system demonstrates the feasibility of an end-to-end nursing surveillance decision support framework linking PPC risk prediction with structured clinical recommendations. The findings suggest its potential to support more accurate risk prediction and more actionable nursing surveillance for patients undergoing abdominal surgery. Full article

Background/Objectives: Breast cancer is one of the most common malignancies, but its heterogeneous molecular subtypes make treatment decision-making complex and patient-specific. Both the pathology reports and the electronic medical record (EMR) play a critical role for an appropriate treatment decision. This study aimed to develop an integrated clinical decision support system (CDSS) that combines a large language model (LLM)-based pathology analysis with deep learning-based treatment prediction to support standardized and reliable decision-making. Methods: Real-world data (RWD) obtained from a cohort of 5015 patients diagnosed with breast cancer were analyzed. Meta-Llama-3-8B-Instruct automatically extracted the TNM stage and tumor size from the pathology reports, which were then integrated with EMR variables. A multi-label classification of 16 treatment combinations was performed using six models, including Decision Tree, Random Forest, GBM, XGBoost, DNN, and Transformer. Performance was evaluated using accuracy, macro/micro-averaged precision, recall, F1 score, and AUC. Results: Using combined LLM-extracted pathology and EMR features, GBM and XGBoost achieved the highest and most stable predictive performance across all feature subset configurations (macro-F1 ≈ 0.88–0.89; AUC = 0.867–0.868). Both models demonstrated strong discrimination ability and consistent recall and precision, highlighting their robustness for multi-label classification in real-world settings. Decision Tree and Random Forest showed moderate but reliable performance (macro-F1 = 0.84–0.86; AUC = 0.849–0.821), indicating their applicability despite lower predictive capability. By contrast, the DNN and Transformer models produced comparatively lower scores (macro-F1 = 0.74–0.82; AUC = 0.780–0.757), especially when using the full feature set, suggesting limited suitability for structured clinical data without strong contextual dependencies. These findings indicate that gradient-boosting ensemble approaches are better optimized for tabular medical data and generate more clinically reliable treatment recommendations. Conclusions: The proposed artificial intelligence-based CDSS improves accuracy and consistency in breast cancer treatment decision support by integrating automated pathology interpretation with deep learning, demonstrating its potential utility in real-world cancer care. Full article

Class imbalance is a major challenge in supervised classification, often leading to biased predictions and limited generalization. This issue is particularly pronounced in medical diagnostics, where datasets typically contain far more negative than positive cases. In this study, we compare two oversampling strategies: the Synthetic Minority Oversampling Technique (SMOTE) and the Conditional Tabular Generative Adversarial Network (ctGAN). Using the benchmark Pima Indians Diabetes dataset, we generated balanced datasets through both methods and trained a multilayer perceptron classifier. Performance was evaluated with accuracy, precision, sensitivity, and F1 Score. The results show that both SMOTE and ctGAN improve classification on imbalanced data, with SMOTE consistently achieving superior sensitivity and F1 Score. These findings highlight the importance of selecting appropriate augmentation strategies to enhance the reliability and clinical usefulness of machine learning models in medical diagnostics. Full article

Machine learning (ML) offers significant potential for disease prediction, clinical decision support, and medical data classification, but its reliance on sensitive patient data raises privacy and security concerns, particularly under strict healthcare regulations. Traditional encryption methods require data to be decrypted prior to computation, such as in ML workflows, thereby introducing risks of exposure and undermining data confidentiality. Homomorphic Encryption (HE) addresses this challenge by enabling computations directly on encrypted data, ensuring end-to-end privacy. This paper explores the integration of the Cheon-Kim-Kim-Song (CKKS) HE scheme into the inference phase of medical tabular data classification. We evaluate the performance of Logistic Regression (LR), Support Vector Machine (SVM), and a lightweight multilayer perceptron (MLP) under HE-based inference, and compare their classification accuracy, computational overhead, and latency against plaintext counterparts. Additionally, we propose two hybrid models (LR-MLP and SVM-MLP) to accelerate training convergence and enhance inference performance. Experimental results demonstrate that while HE-based inference introduces moderate computational cost and data transmission overheads, it maintains accuracy comparable to plaintext inference. These outcomes affirm the practical feasibility of HE for privacy-preserving machine learning in healthcare, while also highlighting key implementation trade-offs. Furthermore, the findings support the advancement of secure AI systems and promote the adoption of cryptographic techniques in digital health and other privacy-critical fields. Full article

Oral cancer represents a life-threatening malignancy with profound implications for patient survival and quality of life. Oral squamous cell carcinoma (OSCC), the predominant histological variant of oral cancer, constitutes a substantial healthcare challenge wherein early detection remains critical for therapeutic efficacy and enhanced survival outcomes. Recent advances in deep learning methodologies have demonstrated superior performance in medical imaging applications. However, existing investigations have predominantly employed unimodal image data for oral lesion classification, thereby neglecting the potential advantages of multimodal data integration. To address this limitation, we propose a comprehensive multimodal pipeline for the classification of OSCC versus leukoplakia through the integration of histopathological imagery with tabular data encompassing anatomical characteristics and behavioral risk factors. Our methodology achieved a precision of 0.97, F1-score of 0.97, recall of 0.98, and accuracy of 0.97. These findings demonstrate the enhanced diagnostic precision and efficacy afforded by multimodal approaches in oral cancer classification, suggesting a promising avenue for improved diagnostic accuracy and treatment planning optimization. Full article

In recent years, data augmentation techniques have become the predominant approach for addressing highly imbalanced classification problems in machine learning. Algorithms such as the Synthetic Minority Over-sampling Technique (SMOTE) and Conditional Tabular Generative Adversarial Network (CTGAN) have proven effective in synthesizing minority class samples. However, these methods often introduce distributional bias and noise, potentially leading to model overfitting, reduced predictive performance, increased computational costs, and elevated cybersecurity risks. To overcome these limitations, we propose a novel architecture, FADEL, which integrates feature-type awareness with a supervised discretization strategy. FADEL introduces a unique feature augmentation ensemble framework that preserves the original data distribution by concurrently processing continuous and discretized features. It dynamically routes these feature sets to their most compatible base models, thereby improving minority class recognition without the need for data-level balancing or augmentation techniques. Experimental results demonstrate that FADEL, solely leveraging feature augmentation without any data augmentation, achieves a recall of 90.8% and a G-mean of 94.5% on the internal test set from Kaohsiung Chang Gung Memorial Hospital in Taiwan. On the external validation set from Kaohsiung Medical University Chung-Ho Memorial Hospital, it maintains a recall of 91.9% and a G-mean of 86.7%. These results outperform conventional ensemble methods trained on CTGAN-balanced datasets, confirming the superior stability, computational efficiency, and cross-institutional generalizability of the FADEL architecture. Altogether, FADEL uses feature augmentation to offer a robust and practical solution to extreme class imbalance, outperforming mainstream data augmentation-based approaches. Full article

Background: Cardiotocography (CTG) is widely used in obstetrics to monitor fetal heart rate and uterine contractions. It helps detect early signs of fetal distress. However, manual interpretation of CTG can be time-consuming and may vary between clinicians. Recent advances in machine learning provide more efficient and consistent alternatives for analyzing CTG data. Objectives: This study aims to investigate the classification of fetal health using various machine learning models to facilitate early detection of fetal health conditions. Methods: This study utilized a tabular dataset comprising 2126 patient records and 21 features. To classify fetal health outcomes, various machine learning algorithms were employed, including CatBoost, Decision Tree, ExtraTrees, Gradient Boosting, KNN, LightGBM, Random Forest, SVM, ANN and DNN. To address class imbalance and enhance model performance, the Synthetic Minority Oversampling Technique (SMOTE) was employed. Results: Among the tested models, the LightGBM algorithm achieved the highest performance, boasting a classification accuracy of 90.73% and, more notably, a balanced accuracy of 91.34%. This superior balanced accuracy highlights LightGBM’s effectiveness in handling imbalanced datasets, outperforming other models in ensuring fair classification across all classes. Conclusions: This study highlights the potential of machine learning models as reliable tools for fetal health classification. The findings emphasize the transformative impact of such technologies on medical diagnostics. Additionally, the use of SMOTE effectively addressed dataset imbalance, further enhancing the reliability and applicability of the proposed approach. Full article

Background/Objectives: The COVID-19 pandemic, caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), led to significant global health challenges, including the urgent need for accurate symptom severity prediction aimed at optimizing treatment. While machine learning (ML) and deep learning (DL) models have shown promise in predicting COVID-19 severity using imaging and clinical data, there is limited research utilizing comprehensive tabular symptom datasets. This study aims to address this gap by leveraging a detailed symptom dataset to develop robust models for categorizing COVID-19 symptom severity, thereby enhancing clinical decision making. Methods: A unique tabular dataset was created using questionnaire responses from 5654 individuals, including demographic information, comorbidities, travel history, and medical data. Both unsupervised and supervised ML techniques were employed, including k-means clustering to categorize symptom severity into mild, moderate, and severe clusters. In addition, classification models, namely, Support Vector Machine (SVM), Adaptive Boosting (AdaBoost), eXtreme Gradient Boosting (XGBoost), random forest, and a deep neural network (DNN) were used to predict symptom severity levels. Feature importance was analyzed using the random forest model for its robustness with high-dimensional data and ability to capture complex non-linear relationships, and statistical significance was evaluated through ANOVA and Chi-square tests. Results: Our study showed that fatigue, joint pain, and headache were the most important features in predicting severity. SVM, AdaBoost, and random forest achieved an accuracy of 94%, while XGBoost achieved an accuracy of 96%. DNN showed robust performance in handling complex patterns with 98% accuracy. In terms of precision and recall metrics, both the XGBoost and DNN models demonstrated robust performance, particularly for the moderate class. XGBoost recorded 98% precision and 97% recall, while DNN achieved 99% precision and recall. The clustering approach improved classification accuracy by reducing noise and dimensionality. Statistical tests confirmed the significance of additional features like Body Mass Index (BMI), age, and dominant variant type. Conclusions: Integrating symptom data with advanced ML models offers a promising approach for accurate COVID-19 severity classification. This method provides a reliable tool for healthcare professionals to optimize patient care and resource management, particularly in managing COVID-19 and potential future pandemics. Future work should focus on incorporating imaging and clinical data to further enhance model accuracy and clinical applicability. Full article

Hospital overcrowding, driven by both structural management challenges and widespread medical emergencies, has prompted extensive research into machine learning (ML) solutions for predicting patient length of stay (LOS) to optimize bed allocation. While many existing models simplify the LOS prediction problem to a classification task, predicting broad ranges of hospital days, an exact day-based regression model is often crucial for precise planning. Additionally, available data are typically limited and heterogeneous, often collected from a small patient cohort. To address these challenges, we present a novel multimodal ML framework that combines imaging and clinical data to enhance LOS prediction accuracy. Specifically, our approach uses the following: (i) feature extraction from chest CT scans via a convolutional neural network (CNN), (ii) their integration with clinically relevant tabular data from patient exams, refined through a feature selection system to retain only significant predictors. As a case study, we applied this framework to pneumonia patient data collected during the COVID-19 pandemic at two hospitals in Naples, Italy—one specializing in infectious diseases and the other general-purpose. Under our experimental setup, the proposed system achieved an average prediction error of only three days, demonstrating its potential to improve patient flow management in critical care environments. Full article

Coronavirus disease 2019 (COVID-19), originating in China, has rapidly spread worldwide. Physicians must examine infected patients and make timely decisions to isolate them. However, completing these processes is difficult due to limited time and availability of expert radiologists, as well as limitations of the reverse-transcription polymerase chain reaction (RT-PCR) method. Deep learning, a sophisticated machine learning technique, leverages radiological imaging modalities for disease diagnosis and image classification tasks. Previous research on COVID-19 classification has encountered several limitations, including binary classification methods, single-feature modalities, small public datasets, and reliance on CT diagnostic processes. Additionally, studies have often utilized a flat structure, disregarding the hierarchical structure of pneumonia classification. This study aims to overcome these limitations by identifying pneumonia caused by COVID-19, distinguishing it from other types of pneumonia and healthy lungs using chest X-ray (CXR) images and related tabular medical data, and demonstrate the value of incorporating tabular medical data in achieving more accurate diagnoses. Resnet-based and VGG-based pre-trained convolutional neural network (CNN) models were employed to extract features, which were then combined using early fusion for the classification of eight distinct classes. We leveraged the hierarchal structure of pneumonia classification within our approach to achieve improved classification outcomes. Since an imbalanced dataset is common in this field, a variety of versions of generative adversarial networks (GANs) were used to generate synthetic data. The proposed approach tested in our private datasets of 4523 patients achieved a macro-avg F1-score of 95.9% and an F1-score of 87.5% for COVID-19 identification using a Resnet-based structure. In conclusion, in this study, we were able to create an accurate deep learning multi-modal to diagnose COVID-19 and differentiate it from other kinds of pneumonia and normal lungs, which will enhance the radiological diagnostic process. Full article

Explainable Artificial Intelligence (XAI) evaluation has grown significantly due to its extensive adoption, and the catastrophic consequence of misinterpreting sensitive data, especially in the medical field. However, the multidisciplinary nature of XAI research resulted in diverse scholars possessing significant challenges in designing proper evaluation methods. This paper proposes a novel framework of a three-layered top-down approach on how to arrive at an optimal explainer, accenting the persistent need for consensus in XAI evaluation. This paper also investigates a critical comparative evaluation of explanations in both model agnostic and specific explainers including LIME, SHAP, Anchors, and TabNet, aiming to enhance the adaptability of XAI in a tabular domain. The results demonstrate that TabNet achieved the highest classification recall followed by TabPFN, and XGBoost. Additionally, this paper develops an optimal approach by introducing a novel measure of relative performance loss with emphasis on faithfulness and fidelity of global explanations by quantifying the extent to which a model’s capabilities diminish when eliminating topmost features. This addresses a conspicuous gap in the lack of consensus among researchers regarding how global feature importance impacts classification loss, thereby undermining the trust and correctness of such applications. Finally, a practical use case on medical tabular data is provided to concretely illustrate the findings. Full article

Classification problems due to data imbalance occur in many fields and have long been studied in the machine learning field. Many real-world datasets suffer from the issue of class imbalance, which occurs when the sizes of classes are not uniform; thus, data belonging to the minority class are likely to be misclassified. It is particularly important to overcome this issue when dealing with medical data because class imbalance inevitably arises due to incidence rates within medical datasets. This study adjusted the imbalance ratio (IR) within the National Biobank of Korea dataset “Epidemiologic data of Parkinson’s disease dementia patients” to values of 6.8 (raw data), 9, and 19 and compared four traditional oversampling methods with techniques using the conditional generative adversarial network (CGAN) and conditional tabular generative adversarial network (CTGAN). The results showed that when the classes were balanced with CGAN and CTGAN, they showed a better classification performance than the more traditional oversampling techniques based on the AUC and F1-score. We were able to expand the application scope of GAN, widely used in unstructured data, to structured data. We also offer a better solution for the imbalanced data problem and suggest future research directions. Full article

Data normalization is a data preprocessing task and one of the first to be performed during intellectual analysis, particularly in the case of tabular data. The importance of its implementation is determined by the need to reduce the sensitivity of the artificial intelligence model to the values of the features in the dataset to increase the studied model’s adequacy. This paper focuses on the problem of effectively preprocessing data to improve the accuracy of intellectual analysis in the case of performing medical diagnostic tasks. We developed a new two-step method for data normalization of numerical medical datasets. It is based on the possibility of considering both the interdependencies between the features of each observation from the dataset and their absolute values to improve the accuracy when performing medical data mining tasks. We describe and substantiate each step of the algorithmic implementation of the method. We also visualize the results of the proposed method. The proposed method was modeled using six different machine learning methods based on decision trees when performing binary and multiclass classification tasks. We used six real-world, freely available medical datasets with different numbers of vectors, attributes, and classes to conduct experiments. A comparison between the effectiveness of the developed method and that of five existing data normalization methods was carried out. It was experimentally established that the developed method increases the accuracy of the Decision Tree and Extra Trees Classifier by 1–5% in the case of performing the binary classification task and the accuracy of the Bagging, Decision Tree, and Extra Trees Classifier by 1–6% in the case of performing the multiclass classification task. Increasing the accuracy of these classifiers only by using the new data normalization method satisfies all the prerequisites for its application in practice when performing various medical data mining tasks. Full article