AI-based educational chatbots can expand access to learning, but many remain limited to text-only interfaces and fixed infrastructures, while purely generative responses raise concerns of reliability and consistency. In this context, we present ROboMC, a portable and multimodal system that combines a validated knowledge base with generative responses (OpenAI) and voice–text interaction, designed to enable both text and voice interaction, ensuring reliability and flexibility in diverse educational scenarios. The system, developed in Django, integrates two response pipelines: local search using normalized keywords and fuzzy matching in the LocalQuestion database, and fallback to the generative model GPT-3.5-Turbo (OpenAI, San Francisco, CA, USA) with a prompt adapted exclusively for Romanian and an explicit disclaimer. All interactions are logged in AutomaticQuestion for later analysis, supported by a semantic encoder (SentenceTransformer—paraphrase-multilingual-MiniLM-L12-v2’, Hugging Face Inc., New York, NY, USA) that ensures search tolerance to variations in phrasing. Voice interaction is managed through gTTS (Google LLC, Mountain View, CA, USA) with integrated audio playback, while portability is achieved through deployment on a Raspberry Pi 4B (Raspberry Pi Foundation, Cambridge, UK) with microphone, speaker, and battery power. Voice input is enabled through a cloud-based speech-to-text component (Google Web Speech API accessed via the Python SpeechRecognition library, (Anthony Zhang, open-source project, USA) using the Google Web Speech API (Google LLC, Mountain View, CA, USA; language = “ro-RO”)), allowing users to interact by speaking. Preliminary tests showed average latencies of 120–180 ms for validated responses on laptop and 250–350 ms on Raspberry Pi, respectively, 2.5–3.5 s on laptop and 4–6 s on Raspberry Pi for generative responses, timings considered acceptable for real educational scenarios. A small-scale usability study (N ≈ 35) indicated good acceptability (SUS ~80/100), with participants valuing the balance between validated and generative responses, the voice integration, and the hardware portability. Although system validation was carried out in the eHealth context, its architecture allows extension to any educational field: depending on the content introduced into the validated database, ROboMC can be adapted to medicine, engineering, social sciences, or other disciplines, relying on ChatGPT only when no clear match is found in the local base, making it a scalable and interdisciplinary solution.

This pilot study evaluated the visionMC system, a low-cost artificial intelligence system integrating HOG-based facial recognition and voice notifications, for workflow optimization in a family medicine practice. Implemented on a Raspberry Pi 4, the system was tested over two weeks with 50 patients. It achieved 85% recognition accuracy and an average detection time of 3.4 s. Compared with baseline, patient waiting times showed a substantial reduction in waiting time and administrative workload, and the administrative workload decreased by 5–7 min per patient. A satisfaction survey (N = 35) indicated high acceptance, with all scores above 4.5/5, particularly for usefulness and waiting time reduction. These results suggest that visionMC can improve efficiency and enhance patient experience with minimal financial and technical requirements. Larger multicenter studies are warranted to confirm scalability and generalizability. visionMC demonstrates that effective AI integration in small practices is feasible with minimal resources, supporting scalable digital health transformation. Beyond biometric identification, the system’s primary contribution is streamlining practice management by instantly displaying the arriving patient and enabling rapid chart preparation. Personalized greetings enhance patient experience, while email alerts on motion events provide a secondary security benefit. These combined effects drove the observed reductions in waiting and administrative times.

Background and Objectives: Coronary computed tomography angiography (CCTA) has become a central non-invasive imaging modality for the evaluation of suspected chronic coronary syndromes. Nevertheless, referral patterns and sex-related differences in real-world CCTA utilization may vary according to demographic structure, healthcare accessibility, and regional clinical practice. The present study aimed to evaluate age- and sex-related patterns in CCTA referrals and coronary atherosclerotic burden within a real-world Eastern European single-center CCTA referral cohort. Materials and Methods: We performed a retrospective analysis of 2742 consecutive CCTA examinations at a tertiary center in Sibiu, Romania, between March 2019 and January 2025. The demographic profile of the referral cohort was compared with the adult population structure of Sibiu County. Subsequently, an age- and sex-stratified proportional subsample of 397 cases was selected from the CCTA cohort for detailed analysis of sex-related differences regarding coronary artery calcium score (CACS), significant coronary stenoses, and significant cardiovascular event risk. Results: The age distribution of the CCTA referral cohort differed significantly from that of the adult county population (p < 0.01), with referrals concentrated predominantly between 55 and 79 years of age. Female patients were overall more frequently represented within the referral cohort compared with the underlying county population; however, males demonstrated significantly higher coronary artery calcium scores, a greater prevalence of significant coronary stenoses, and higher revascularization risk compared with females—particularly between 50 and 79 years of age—within the stratified proportional subsample. In contrast, patients younger than 40 years demonstrated very low rates of significant coronary disease regardless of sex. Conclusions: This study provides a real-world perspective on age- and sex-related patterns in CCTA utilization and coronary atherosclerotic burden within an Eastern European imaging center. The findings highlight the interaction between demographic structure, guideline-directed diagnostic strategies, and coronary imaging utilization in routine clinical practice.

The health monitoring of military personnel is critical for sustaining operational readiness and minimizing medical risks. In this study, we present an artificial intelligence-driven application that classifies the health status of service members based on two simple medical parameters: heart rate and body weight. The system integrates two complementary approaches: supervised learning using decision trees and unsupervised learning with the K-Means algorithm. Experimental evaluation demonstrates that the decision tree achieved an accuracy of 90%, as confirmed by the confusion matrix, while the K-Means method identified distinct clusters aligned with clinically relevant health profiles: healthy, cardiac risk, and metabolic risk. These results highlight the potential of artificial intelligence to be effectively integrated into military medical monitoring systems, supporting early diagnosis and enhancing the efficiency of health assessment processes.

Increasing workload, time pressure, and administrative demands in family medicine create conditions in which delays in routine preventive and monitoring activities may gradually become normalized. When such delays are no longer perceived as exceptional, they can evolve into latent safety threats that remain largely invisible in everyday clinical practice. This study explores whether a simple digital monitoring approach can support the visibility and management of normalized risk in a primary care setting. This exploratory, single-site, practice-based quality improvement initiative was conducted in a family medicine setting. A lightweight digital monitoring dashboard was used to track routine preventive and chronic care recommendations over time, based on commonly accepted follow-up intervals. Overdue actions were identified through comparison between expected timing and recorded completion, using routinely available clinical data integrated into everyday workflows. At baseline, a small number of overdue recommendations (including at least one active delay distributed across multiple care domains) were identified, with a temporal pattern suggestive of accumulating latent risk. Following targeted clinical follow-up, a partial reduction in delays was observed, with improvement across care domains. In the final observation phase, all overdue recommendations were resolved (0 active delays), suggesting a transition toward a more stable and lower-risk operational state within the observed practice context. These findings suggest that simple, low-cost digital monitoring tools may help support the visibility and management of normalized risk in family medicine. By making routine delays visible, such approaches may enhance situational awareness and support timely clinical follow-up in resource-constrained primary care environments.