Background: Pregnancy-induced hypertension (PIH) remains a leading cause of maternal and perinatal morbidity and mortality worldwide, with increasing incidence noted in developing countries. Early identification of at-risk pregnancies is crucial for timely intervention. Uterine artery Doppler (UAD) velocimetry offers a non-invasive method to assess uteroplacental perfusion and predict hypertensive complications. Objective: To evaluate the predictive value of second-trimester uterine artery Doppler parameters—specifically pulsatility index (PI), resistance index (RI), and diastolic notching—for the development of pregnancy-induced hypertension. Methods: A prospective observational study was conducted on 48 primigravidae at 19–22 weeks gestation. UAD parameters (PI, RI, and presence of notching) were measured, and participants were followed up until delivery for the development of PIH. Results: Abnormal UAD parameters (mean PI > 1.58, RI > 0.58, or presence of bilateral notching) were significantly associated with PIH development. The sensitivity and specificity of combined abnormal parameters for predicting PIH were 81.25% and 87.50%, respectively. Conclusion: Second-trimester uterine artery Doppler is a valuable, non-invasive predictor of pregnancy-induced hypertension and should be incorporated into routine antenatal screening protocols, especially in resource-limited settings. Keywords: Pregnancy-induced hypertension, uterine artery Doppler, pulsatility index, resistance index, diastolic notching, preeclampsia prediction.
This article presents a comprehensive framework for assessing the effectiveness of Business Intelligence (BI) systems in information technology companies. The aim of the study is to develop a methodological and analytical approach that allows assessing the impact of BI systems on the operational efficiency, decision-making processes and financial results of organizations. For this purpose, a combination of quantitative and qualitative methods is used, as well as mathematical modeling, which allows for a systematic analysis of the impact of BI systems implementation on various areas of an organization’s activity. One of the important results of the research is the formation of a combined Business Intelligence Efficiency Index (BIE index), which allows assessing the overall efficiency of the system with a single and interpretable indicator. This indicator is built on the basis of a weighted approach and emphasizes the important role of financial performance in shaping the overall results of the organization. In addition, the proposed model is complemented by return on investment (ROI) analysis and strategic planning tools, which allows for a comparative assessment of the performance of BI systems. The results of the research can be used both in academic studies and in the field of practical management, serving as a basis for auditing the efficiency of Business Intelligence systems, developing data-driven management, and increasing the competitiveness of information technology companies.
Early diagnosis of fetal health complications is crucial to ensure maternal and infant safety. However, conventional tabular machine learning approaches often suffer from severe majority-class bias and a lack of clinical interpretability. In this research, an optimized Python-based framework was developed using a 2,126-record dataset containing 21 predictive features to classify fetal health status. To resolve structural class imbalance without data leakage, the synthetic minority over-sampling technique (SMOTE) was dynamically applied strictly within stratified 5-fold cross-validation loops. Six configurations—naive Bayes, logistic regression, support vector machines, a multi-tiered deep multi-layer perceptron (deep MLP), random forest, and extreme gradient boosting (XGBoost)—were systematically tuned using randomized grid searches. Among the benchmarked models, the upgraded XGBoost ensemble demonstrated the highest performance, leading with a diagnostic accuracy of 94.87%, a macro F1-score of 91.37%, and an area under the curve (AUC) of 98.77%. While the deep MLP required a high training latency of 13.8179 seconds, the optimal XGBoost layout achieved superior power with a training runtime of only 1.8325 seconds and an inference speed of 0.0046 seconds. Game-theoretic Shapley additive explanations (SHAP) values confirmed abnormal short-term variability as the primary risk driver, ensuring a transparent, high-throughput decision-support tool.
This paper presents a novel approach for automatically extracting and structuring bridge-related data from engineering drawings using multimodal Large Language Models (mmLLMs). Requests are issued via API calls, with the output schema defined using Pydantic classes to guide the LLM and ensure consistent, machine-readable data exchange. We investigate structured prompting strategies across zero-shot, one-shot, and few-shot scenarios. Results show that zero-shot prompting alone achieved only 55.3
Recently, the criticism of the emotional material in speech became a topic of extensive interest because it is used in contact with computers and mental health supervision and automatized customer service. The given paper introduces an exhaustive work of the article Echoes of Emotion: A Machine Learning Approach to Speech Sentiment Analysis, which discusses the power of machine learning based methods in identifying and categorizing emotions by using speech signals. We illustrate some of the feature extraction schemes like deep learning-based feature extraction like prosodic, spectral and representations and compare the performances on known speech emotions databases. This has been achieved through the use of state of art machine learning models, specifically support vector machines, random forests and deep neural networks, to correctly determine emotional states such as happiness, sadness, anger and neutrality. The findings indicate that machine learning models, especially those ones that employ deep learning network, attain encouraging accuracy in classification. The results lead to the possibility of speech emotion recognition systems to improve user inter- action with interactive technologies and bring insight into emotional expression using speech.
