Collaço, Bernardo Gabriele

Background: During postoperative recovery, patients and their caregivers often lack crucial information, leading to numerous repetitive inquiries that burden healthcare providers. Traditional discharge materials, including paper handouts and patient portals, are often static, overwhelming, or underutilized, leading to patient overwhelm and contributing to unnecessary ER visits and overall healthcare overutilization. Conversational chatbots offer a solution, but Natural Language Processing (NLP) systems are often inflexible and limited in understanding, while powerful Large Language Models (LLMs) are prone to generating “hallucinations”. Objective: To combine the deterministic framework of traditional NLP with the probabilistic capabilities of LLMs, we developed the AI Virtual Assistant (AIVA) Platform. This system utilizes a retrieval-augmented generation (RAG) architecture, integrating Gemini 2.0 Flash with a medically verified knowledge base via Google Vertex AI, to safely deliver dynamic, patient-facing postoperative guidance grounded in validated clinical content. Methods: The AIVA Platform was evaluated through 750 simulated patient interactions derived from 250 unique postoperative queries across 20 high-frequency recovery domains. Three blinded physician reviewers assessed formal system performance, evaluating classification metrics (accuracy, precision, recall, F1-score), relevance (SSI Index), completeness, and consistency (5-point Likert scale). Safety guardrails were tested with 120 out-of-scope queries and 30 emergency escalation scenarios. Additionally, groundedness, fluency, and readability were assessed using automated LLM metrics. Results: The system achieved 98.4% classification accuracy (precision 1.0, recall 0.98, F1-score 0.9899). Physician reviews showed high completeness (4.83/5), consistency (4.49/5), and relevance (SSI Index 2.68/3). Safety guardrails successfully identified 100% of out-of-scope and escalation scenarios. Groundedness evaluations demonstrated strong context precision (0.951), recall (0.910), and faithfulness (0.956), with 95.6% verification agreement. While fluency and semantic alignment were high (BERTScore F1 0.9013, ROUGE-1 0.8377), readability was 11th-grade level (Flesch–Kincaid 46.34). Conclusion: The simulated testing demonstrated strong technical accuracy, safety, and clinical relevance in simulated postoperative care. Its architecture effectively balances flexibility and safety, addressing key limitations of standalone NLP and LLMs. While readability remains a challenge, these findings establish a solid foundation, demonstrating readiness for clinical trials and real-world testing within surgical care pathways.

Retrieval-augmented generation (RAG) quality depends on how source documents are segmented before indexing; fixed-length chunks can split concepts or add noise, reducing precision. We evaluated whether proposition, semantic, and adaptive chunking improve accuracy and relevance for safer clinical decision support. Using a curated domain knowledge base with Gemini 1.0 Pro, we built four otherwise identical RAG pipelines that differed only in the chunking strategy: adaptive length, proposition, semantic, and a fixed token-dependent baseline. Thirty common postoperative rhinoplasty questions were submitted to each pipeline. Outcomes included medical accuracy and clinical relevance (3-point Likert scale) and retrieval precision, recall, and F1; group differences were tested with ANOVA and Tukey post hoc analyses. Adaptive chunking achieved the highest accuracy—87% (Likert 2.37 ± 0.72) versus baseline 50% (1.63 ± 0.72; p = 0.001)—and the highest relevance (93%, 2.90 ± 0.40). Retrieval metrics were strongest with adaptive (precision 0.50, recall 0.88, F1 0.64) versus baseline (0.17, 0.40, 0.24). Proposition and semantic strategies improved all metrics relative to baseline, though less than adaptive. Aligning chunks to logical topic boundaries yielded more accurate, relevant answers without modifying the language model, offering a model-agnostic, data-source-neutral lever to enhance the safety and utility of LLM-based clinical decision support.

Background: Artificial Intelligence (AI)-driven Decision Support Systems (DSSs) promise improvements in diagnostic accuracy and clinical workflow efficiency, but their adoption is hindered by inadequate confidence calibration, limited transparency, and poor alignment with real-world decision processes, which limit clinician trust and lead to high override rates. Methods: We developed and validated a dynamic scoring framework to enhance trust in AI-generated diagnoses by integrating AI confidence scores, semantic similarity measures, and transparency weighting into the override decision process using 6689 cardiovascular cases from the MIMIC-III dataset. Override thresholds were calibrated and validated across varying transparency and confidence levels, with override rate as the primary acceptance measure. Results: The implementation of this framework reduced the override rate to 33.29%, with high-confidence predictions (90–99%) overridden at a rate of only 1.7%, and low-confidence predictions (70–79%) at a rate of 99.3%. Minimal transparency diagnoses had a 73.9% override rate compared to 49.3% for moderate transparency. Statistical analyses confirmed significant associations between confidence, transparency, and override rates (p < 0.001). Conclusions: These findings suggest that enhanced transparency and confidence calibration can substantially reduce override rates and promote clinician acceptance of AI diagnostics. Future work should focus on clinical validation to optimize patient safety, diagnostic accuracy, and efficiency.

Abstract Background/ Objectives: Agentic AI represents a promising evolution of AI technology applied to healthcare, with systems increasingly capable of operating autonomously to achieve defined clinical goals. However, the literature lacks conceptual clarity between “AI agents” and “agentic AI”, and few studies have rigorously explored their clinical applications. Therefore, this study aims to conduct a novel systematic review addressing this gap by examining agentic AI systems in healthcare settings, characterizing their applications, features, outcomes, and limitations, and clarifying the conceptual distinctions between AI agents and agentic AI systems using predefined and objective criteria. Methods A comprehensive search was conducted across PubMed, Embase, Cochrane, Scopus, and Google Scholar on April 6th, 2025. Studies were included if they involved AI systems in healthcare settings that demonstrated the following agentic features: autonomous operation, goal-directed behavior, and initiating action. Data on the clinical tasks achieved by the agents, key findings, features, and limitations were collected from the included studies. Screening and extraction followed PRISMA guidelines, with Risk of Bias assessed using ROBINS-I and Cochrane's Risk of Bias tools. Results Of 984 retrieved records, seven studies met the inclusion criteria, spanning domains such as emergency medicine, oncology, radiology, and rehabilitation. Multi-agent architecture was frequently used to decompose and coordinate complex workflows. Among the included studies, the AIs showed high accuracy in diagnosing cancer patients, conducting treatment plans, sending alerts, coaching messages, analysing image data, and adapting to challenging experimental scenarios. While demonstrating potential for improved efficiency, task accuracy, and patient engagement, significant limitations were noted: narrow task scope, lack of physical agency, limited clinical validation, and barriers to integration into real-world healthcare systems. Only one system had been deployed in a patient-facing trial setting. Conclusion The current literature suggests an emerging role and application of Agentic AI, holding promise with the potential to revolutionize diagnostics, triage, treatment planning, and patient management. However, real-world implementation and evaluations in the literature are limited. Future research must address critical validation, regulation, ethics, and clinical integration challenges to realize their full potential. Clear operational definitions and frameworks for evaluating agency are essential to support safe and effective deployment of these systems.

Background: Data accessibility remains a significant barrier in healthcare AI due to privacy constraints and logistical challenges. Synthetic data, which mimics real patient information while remaining both realistic and non-identifiable, offers a promising solution. Large Language Models (LLMs) create new opportunities to generate high-fidelity clinical conversations between patients and physicians. However, the value of this synthetic data depends on careful evaluation of its realism, accuracy, and practical relevance. Objective: To assess the performance of four leading LLMs: ChatGPT 4.5, ChatGPT 4o, Claude 3.7 Sonnet, and Gemini Pro 2.5 in generating synthetic transcripts of patient–physician interactions in plastic surgery scenarios. Methods: Each model generated transcripts for ten plastic surgery scenarios. Transcripts were independently evaluated by three clinically trained raters using a seven-criterion rubric: Medical Accuracy, Realism, Persona Consistency, Fidelity, Empathy, Relevancy, and Usability. Raters were blinded to the model identity to reduce bias. Each was rated on a 5-point Likert scale, yielding 840 total evaluations. Descriptive statistics were computed, and a two-way repeated measures ANOVA was used to test for differences across models and metrics. In addition, transcripts were analyzed using automated linguistic and content-based metrics. Results: All models achieved strong performance, with mean ratings exceeding 4.5 across all criteria. Gemini 2.5 Pro received mean scores (5.00 ± 0.00) in Medical Accuracy, Realism, Persona Consistency, Relevancy, and Usability. Claude 3.7 Sonnet matched the scores in Persona Consistency and Relevancy and led in Empathy (4.96 ± 0.18). ChatGPT 4.5 also achieved perfect scores in Relevancy, with high scores in Empathy (4.93 ± 0.25) and Usability (4.96 ± 0.18). ChatGPT 4o demonstrated consistently strong but slightly lower performance across most dimensions. ANOVA revealed no statistically significant differences across models (F(3, 6) = 0.85, p = 0.52). Automated analysis showed substantial variation in transcript length, style, and content richness: Gemini 2.5 Pro generated the longest and most emotionally expressive dialogues, while ChatGPT 4o produced the shortest and most concise outputs. Conclusions: Leading LLMs can generate medically accurate, emotionally appropriate synthetic dialogues suitable for educational and research use. Despite high performance, demographic homogeneity in generated patients highlights the need for improved diversity and bias mitigation in model outputs. These findings support the cautious, context-aware integration of LLM-generated dialogues into medical training, simulation, and research.