Producción científica y docente
Producción científica reciente
SPE-ECG: An Ontology-Guided Knowledge Representation Framework for Enhancing LLM-Based Code Generation
Large Language Models (LLMs) can translate natural-language instructions into code, but when the input is unstructured clinical protocol text, the generated code may omit doses, monitoring intervals, or safety thresholds that the text states. Such omissions matter because they can propagate into software that supports clinical monitoring. We propose SPE-ECG (Structured Prompt Engineering for Enhanced Code Generation), a transparent, rule-based pipeline that converts unstructured domain text into a JSON prompt using text cleaning, part-of-speech tagging, bigram matching, and hand-built domain dictionaries, without fine-tuning any model. In a case study of 43 rheumatoid arthritis DMARD monitoring protocols, we compared raw and structured prompts on three LLMs (GPT-3.5, LLaMA 2, and Gemma), scoring the generated Python code using version 3.15.1 with BLEU against the raw-prompt code of the other two models. Mean BLEU rose for all three models (relative gains of 55.1%, 32.5%, and 19.9%); the paired differences were significant for GPT-3.5 and LLaMA 2 but not, after correction for multiple comparisons, for Gemma. These results suggest that lightweight input structuring can change how closely different models’ outputs agree, and that it could serve as a low-cost, auditable preparation step for protocol-to-code tools built on free or open models. BLEU does not measure correctness, no ablation or state-of-the-art baseline was run, and clinical validation is required before any use.
Cost-sensitive random forest with adaptive threshold optimization for imbalanced financial fraud detection
Background: Financial transaction fraud detection is a challenging binary classification problem characterized by severe class imbalance, rare fraudulent cases, and asymmetric misclassification costs. In real-world banking systems, failing to detect fraudulent transactions can lead to substantial financial losses, whereas false alarms mainly increase operational review costs. Therefore, fraud detection models should not be optimized only for overall accuracy, as accuracy-oriented classifiers may favor the majority non-fraud class and fail to identify minority fraud cases effectively. To address this issue, this study proposes a cost-sensitive ML framework for financial transaction fraud detection under class imbalance and asymmetric financial risk.Methods: In this study, a cost-sensitive RF framework with adaptive probability threshold optimization is proposed for binary fraud classification. Misclassification costs are incorporated during model training through class-weighted learning, allowing the classifier to assign greater importance to minority fraud cases. In addition, instead of using the default probability threshold of 0.5, the decision threshold is optimized using validation data to minimize the expected misclassification cost and improve minority-class detection. The proposed framework was evaluated on two publicly available financial fraud datasets with different transaction characteristics: the Credit Card Fraud Detection dataset and the Financial Fraud Prediction dataset. Its performance was compared with conventional ML baselines using fraud-oriented evaluation metrics, including precision, recall, F1-score, false-negative rate, false-positive rate, and expected misclassification cost.Results: The proposed framework achieved strong and consistent performance across both datasets. On the Credit Card Fraud Detection dataset, it achieved a precision of 94.4%, recall of 85.7%, and F1-score of 89.8%. On the Financial Fraud Prediction dataset, it achieved a precision of 93.1%, a recall of 83.8%, and an F1-score of 88.2%. Compared with conventional fixed-threshold ML models, the adaptive threshold strategy reduced false-negative predictions while maintaining high precision. These results indicate that the proposed cost-sensitive framework provides a more balanced decision strategy for fraud detection under highly imbalanced transaction distributions.Conclusion: The findings demonstrate that combining class-weighted learning with adaptive threshold optimization can improve fraud detection performance under asymmetric misclassification costs. The proposed framework offers a practical and interpretable decision-support approach for banking fraud detection systems, particularly in settings where reducing undetected fraudulent transactions is more important than maximizing overall accuracy alone.
Fine-Grained Human Activity Recognition in Domestic Settings Using Ultra-Wideband Radar: A Comparative Study of Deep Learning Models
Human activity recognition with ultra-wideband (UWB) radar offers contactless, privacy-preserving monitoring of daily living, but distinguishing fine-grained activities performed in the same posture remains difficult. We recorded nine daily activities from 30 participants with a single PulsON 410 radar (3.1–5.3 GHz); seven are performed seated in the same chair and differ only by subtle hand and arm motions. Each 15 s recording is a 120×1440 channel–impulse–response matrix, preprocessed by background subtraction, two-dimensional wavelet denoising and standardization. Six deep learning models (Transformer, CNN-LSTM, 1D-CNN, LSTM, GRU and MLP) and a soft-voting ensemble were compared with participant-independent splits (22 training, 3 validation and 5 test participants) and five-fold cross-validation. The Transformer reached the highest cross-validation accuracy (97.74±0.30% ). On the held-out participants, the Transformer (97.16%), 1D-CNN (96.93%) and GRU (96.45%) performed similarly, ahead of the CNN-LSTM, LSTM and MLP (87.47%), and a three-model ensemble reached 98.35%. Two additional participant splits and a retrained re-run confirmed that the MLP is clearly the weakest model, but the order of the other models changed by several points between runs; in the re-run, an ensemble of all six models did not exceed the Transformer (97.74%), and a range–Doppler baseline reached only 50.63%. Eating versus drinking tea and mobile phone use versus reading or sitting idle were the most confused pairs. Given the small test cohort and the controlled setting, the results are indicative rather than definitive.
Dietary microRNAs from dairy products and their potential relevance to breast cancer: A systematic review
Background The potential role of dairy products in breast cancer development remains controversial, with studies reporting both protective and detrimental effects. MicroRNAs, small non-coding RNA molecules involved in post-transcriptional gene regulation, are increasingly recognized as important mediators in cancer biology, including breast cancer. Milk and other dairy products contain extracellular vescicle-associated microRNAs, some of which are conserved, bioactive, and capable of surviving digestion. Additionally, dairy intake may modulate the expression of endogenous microRNAs in humans. Scope and approach This review explores the landscape of dairy-related microRNAs—both exogenous microRNAs naturally present in dairy products and endogenous microRNAs influenced by dairy consumption—and examines their potential overlap with those implicated in breast cancer. A structured PubMed search identified studies addressing: (1) the most relevant microRNAs detected in dairy products, (2) human-based studies investigating microRNA modulation following dairy intake, and (3) reported associations with breast cancer across clinical and molecular contexts. Key findings and conclusions We identified a group of dairy-associated microRNAs—particularly, miR-148a-3p-, miR-21-5p, miR-30a-5p-5p, miR-125b-5p, miR-26a-5p, miR-155-5p, and miR-146a-5p—that have been linked to key processes involved in breast cancer development, including cell proliferation, apoptosis, epithelial-to-mesenchymal transition, and chemoresistance. Some of these microRNAs are also differentially expressed in mastitic milk or modulated by farming systems and stress-related conditions, highlighting the need for interdisciplinary approaches bridging food systems and human health. While causal inference is beyond the scope of this review, the widespread and sustained consumption of dairy products supports the need for further research to clarify how dairy intake may influence microRNA-mediated regulatory mechanisms and breast cancer risk in humans. The available evidence also suggests that dairy subtype, origin, and processing may be relevant when considering potential dietary implications.
Framework to Detect Adulterated Honey Using Hyperspectral Imaging and Machine Learning
Honey is a natural product renowned for its nutritional and medicinal properties. However, the increasing market demand in Pakistan has led to frequent instances of adulterated and fraudulent honey. Conventional detection methods are widely used but are often ineffective in accurately identifying adulteration. This study proposes a novel approach to detecting adulteration in honey, specifically the mixing of sugar, using machine learning (ML) techniques in conjunction with hyperspectral imaging (HSI). By leveraging HSI, we capture distinct spectral features of both pure and adulterated honey. These spectral features are processed and analyzed using ML algorithms trained on a dataset comprising pure and mixed honey samples. Experiments involve binary, as well as, multiclass (5 classes) honey samples with different levels of adulteration. In addition, data balancing is also carried out using synthetic minority oversampling technique (SMOTE). The results prove the proposed approach to be efficient in detecting adulteration across various honey varieties, with linear regression achieved 99.9% while support vector machine and multilayer perceptron obtaining a 98.7% accuracy for binary class. For the multiple classes of honey, the multilayer perceptron shows a 96.67% accuracy outperforming the remaining ML and deep learning models. Among the deep learning models, the convolutional neural network (CNN) achieved the best performance with an accuracy of 94.67% after applying the SMOTE × 3 (three times of over samples of original samples of each class) data balancing strategy, whereas recurrent neural network (RNN), long short-term memory (LSTM), and gated recurrent unit (GRU) models showed comparatively lower performance.