Abstract This study examines the influence of two data preprocessing techniques, Robust Scaler (RS) and Principal Component Analysis (PCA), on the predictive behaviour of Machine Learning (ML) models used for triage assessment in Internet of Medical Things (IoMT)-based telemedicine environments to enhance the decision support for healthcare, where heterogeneous and high-dimensional clinical data often introduce challenges related to variability and analytical stability. The objective is to evaluate how RS and PCA contribute to the consistency and reliability of triage predictions when integrated with established ML algorithms. A dataset of 55,680 outpatient records was used to assess the integration of RS and PCA with five supervised ML models: Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), Logistic Regression (LR), and Naïve Bayes (NB). Performance was evaluated using accuracy, precision, recall, and F-score metrics. The evaluation reveals that each preprocessing technique affects model behaviour in distinct ways, and the RS Decision Tree combination exhibits stable performance in generating triage outcomes. PCA-based models exhibit characteristic patterns associated with dimensionality reduction, affecting interpretability and model response. The findings emphasize the role of preprocessing techniques in shaping ML-driven telemedicine workflows. Applying RS within ML pipelines supports consistent triage prediction, contributing to timely identification of patient conditions and strengthening data-driven remote healthcare services.

Abstract Stroke remains a leading cause of morbidity and mortality worldwide, necessitating rapid and accurate diagnosis to improve patient outcomes. The gold standard in stroke diagnosis relies on a costly and time-consuming combination of clinical evaluation and neuroimaging techniques, while also being associated with misdiagnoses due to mimicking conditions. This highlights the need for improving the current healthcare system towards faster, more accurate, and low-resource diagnostics. In this direction, this study presents the development of a novel artificial intelligence-driven decision support system for stroke diagnosis using machine learning techniques to fuse standard EHR variables (electronic health records) with information derived from extracellular vesicles (EVs) in blood plasma, which hold promise as diagnostic markers due to their association with stroke pathophysiology. As such, flow cytometry was employed to measure blood samples from 140 patients suffering from stroke or stroke-mimicking conditions. The derived EV-features, including statistical characteristics of their diameters’ distribution and concentration, were integrated with demographical, clinical, and biochemical patient data to form a comprehensive feature set. By utilizing a decision tree learning model, the proposed system achieved a classification weighted accuracy of 86%. This achievement indicates the approach’s potential for integrating cutting-edge photonics-based signals with standard HER information to improve clinical decision-making in emergency settings and for aiding treatment strategies in stroke care.