F21

NETWORKING SESSION
AI will make our lives better

NETWORKING SESSION
AI will make our lives better

Chairpersons: James BOWNESS (Consultant Anaesthetist) (Chairperson, London, United Kingdom), Sandy KOPP (Professor of Anesthesiology and Perioperative Medicine) (Chairperson, Rochester, USA)

10:30 - 10:30 Introduction.

10:30 - 10:52 Improved trainings. Ammar SALTI (Anesthesiologist and Pain Physician) (Keynote Speaker, abu Dhabi, United Arab Emirates)

10:52 - 11:14 #48490 - FT37 Improved presenatations.

Improved presenatations.

Artificial intelligence (AI) is one of the most disruptive innovations of the contemporary era, with a growing impact across numerous fields of knowledge, including medical education. Its integration into teaching and learning processes has brought about a paradigm shift, wherein intelligent digital tools complement and enhance the training of future healthcare professionals. From high-fidelity clinical simulation environments to adaptive learning platforms and predictive analytics systems, AI offers unprecedented opportunities to enrich education, personalize learning, and strengthen pedagogical decision-making. Here’s a clear and straightforward look of the current role of artificial intelligence in medical education, exploring its main applications, advantages, and the ethical and operational challenges it poses within academic and clinical contexts. As artificial intelligence becomes increasingly embedded in daily life, public concern has grown in parallel. These fears (often fuelled by a lack of understanding or by dystopian narratives portrayed in the media) reflect legitimate anxieties that warrant critical examination. While it is true that AI-driven automation is progressively replacing routine tasks in sectors such as manufacturing, logistics, and customer service, concerns about widespread job loss tend to be overstated. Rather than eliminating professions entirely, artificial intelligence often redefines job roles, shifting value toward cognitive, creative, and interpersonal skills. At the same time, new professional opportunities are emerging in fields related to the development, governance, and oversight of these technologies, such as data science, AI ethics, and the management of automated systems. The notion of an artificial intelligence system operating autonomously and beyond human control belongs largely to the realm of theoretical speculation and science fiction. While the concept of a "superintelligence", a cognitive entity capable of vastly surpassing human intellectual capacities, has gained attention in media discourse and certain philosophical circles, it remains far removed from current technological realities. Some experts advocate for proactive reflection on such long-term risks, emphasizing the importance of establishing ethical and regulatory frameworks in advance. However, from a technical standpoint, today's AI systems possess neither independent agency nor consciousness. They function based on algorithms trained with human-provided data and are directed toward objectives defined by human designers. Moreover, their operation is subject to continuous human oversight, validation, and adjustment. In this context, the fear of an uncontrollable AI is better understood as a hypothetical concern rather than an immediate, evidence-based threat. When it comes to privacy, the use of artificial intelligence technologies in surveillance systems, facial recognition, and large-scale personal data analysis presents tangible and legitimate risks. Intelligent platforms, with their ability to collect, process, and interpret vast amounts of sensitive information, have raised growing public concern about the potential erosion of individual rights and the misuse of data. Furthermore, various AI-based tools may be vulnerable to malicious or fraudulent use if not implemented with appropriate safeguards. Nevertheless, it is important to emphasize that not all AI applications are inherently intrusive. Evolving regulatory frameworks, such as the General Data Protection Regulation (GDPR) in the European Union, aim to ensure the responsible deployment of these technologies. In parallel, there is a growing movement toward the development of ethical, transparent, and human-centered AI systems that prioritize the protection of fundamental rights. While it’s natural to have concerns about AI, it is essential to recognize that the potential for malicious use does not lie inherently within the technology itself, but rather in the ethical frameworks, regulatory structures, and intentions of the human agents who design, implement, and govern it. When applied ethically, transparently, and in service of the public good, artificial intelligence can serve as a powerful tool to optimize processes, reduce operational burdens, and redirect professional efforts toward irreplaceable functions that require human judgment, interpersonal sensitivity, and the cultivation of meaningful relationships. AI is changing how regional anaesthesia concepts are taught, enabling the development of more interactive, clear, and effective presentations. Its application in education not only enhances the quality of teaching materials but also expands the possibilities for content personalization and adaptation to different student profiles (FIG 1). One big advantage of AI is the generation of advanced visual content. With intelligent tools, it is possible to create customized anatomical illustrations, simulated ultrasound images, or three-dimensional animations that accurately depict needle trajectories and anaesthetic spread. In addition, clinical images can be enhanced or edited to highlight key anatomical structures, making complex areas easier to understand. These tools enable the creation of original visual content, which is essential for avoiding legal issues related to copyright infringement. There are hundreds of applications that can be used for this purpose, each offering its own advantages and features. The choice among them typically depends on personal preference or the balance they provide between cost and functionality. The author relies on platforms such as Clipdrop (https://clipdrop.co), Leonardo AI (https://leonardo.ai/), and Sora (https://openai.com/es-ES/sora/) to achieve this purpose. AI also plays a significant role in simplifying and clarifying language. It can rewrite technical texts in more accessible language for learners at various levels, from undergraduate students to specialists. High-quality automatic translation is also available, which supports multilingual teaching and provides access to materials in foreign languages, thereby broadening the range of available resources. Among the available tools for translation and text enhancement in multiple languages, Deep-L (https://www.deepl.com/es/translator) stands out for its proven track record and the remarkable quality of its results. Text-to-speech technologies should also not be overlooked, as they enable natural-sounding reading of any content in various languages and even allow for voice cloning, adding a striking level of realism. In this field, I recommend Eleven Labs (https://elevenlabs.io), a platform that excels in voice quality and the versatility of its features. Another big plus is the use of virtual presentation assistants. These tools can generate presentation slides from basic texts or outlines and automatically insert summaries, glossaries, and key points to reinforce learning during lectures. The most well-known application for this purpose is Gamma (https://gamma.app), a platform that stands out for its ability to automatically generate visually engaging presentations using artificial intelligence. In terms of assessment, AI facilitates the automated creation of exams and interactive quizzes, including instant feedback. It can also analyze student performance, allowing instructors to tailor their content to individual needs and learning progress. This type of task can be easily carried out using well-known applications such as ChatGPT (https://chatgpt.com/) or Copilot (https://copilot.microsoft.com) And last but not least, AI helps a lot in research support and continuous content updating. Through the automated analysis of recent scientific publications, it can summarize and highlight the most relevant evidence, as well as suggest emerging techniques that should be incorporated into the teaching of regional anesthesia. For this purpose, I would suggest leveraging two particularly useful platforms: Perplexity (https://www.perplexity.ai/) and Open Evidence (https://www.openevidence.com/)

Vicente ROQUES (Murcia. Spain, Spain)

11:14 - 11:36 Improved image quality. Balavenkat SUBRAMANIAN (Faculty) (Keynote Speaker, Coimbatore, India)

11:36 - 11:58 #48667 - FT38 Artificial Intelligence and Communication in Regional Anesthesia: From Procedural Clarity to Educational Innovation.

Artificial Intelligence and Communication in Regional Anesthesia: From Procedural Clarity to Educational Innovation.

Artificial Intelligence and Communication in Regional Anesthesia: From Procedural Clarity to Educational Innovation Margarita Borislavova MD1, Liesbeth Brullot MD2, Steve Coppens MD, PhD2,3 1 Centre Hospitalier Simone Veil, Department of Anesthesiology, France 2 University Hospitals of Leuven, Department of Anesthesiology, Herestraat 49, B-3000, Leuven, Belgium 3 University of Leuven, Biomedical Sciences Group, Department of Cardiovascular Sciences, KU Leuven, B-3000, Leuven, Belgium Abstract Background: Regional anesthesia (RA) demands effective, real-time communication between clinicians, assistants, and trainees to ensure procedural accuracy, patient safety, and high-quality education. Yet communication breakdowns remain a significant challenge, particularly in high-pressure clinical settings and during hands-on workshops. With the rapid evolution of artificial intelligence (AI), new tools are emerging that can support, augment, and streamline communication in RA. Objectives: This narrative review explores how AI technologies—ranging from ultrasound-integrated anatomical labeling to natural language processing and AI-powered dashboards—can improve communication in regional anesthesia practice and education. The article emphasizes real-world use cases, with particular attention to ultrasound-guided procedures and RA workshops. Key Content and Findings: AI-enhanced ultrasound platforms allow for automatic real-time labeling of nerves, fascial planes, and vascular structures, improving both procedural clarity and educational consistency. Speech recognition and AI-based checklists can transcribe procedural narration into structured documentation, reducing ambiguity and cognitive load. Smart dashboards and alert systems facilitate team coordination and communication across perioperative workflows. In workshop environments, AI reduces the need for physical pointing or repeated verbal correction by visually reinforcing key anatomical landmarks. Conclusions: AI tools have the potential to transform communication in regional anesthesia by reducing verbal ambiguity, improving visualization, and enhancing procedural documentation and workflow coordination. While these systems cannot replace clinical judgment, they represent a powerful adjunct for promoting clarity, safety, and instructional effectiveness in both clinical and teaching contexts. Keywords: Regional anesthesia; artificial intelligence; ultrasound guidance; medical education; communication. Introduction Regional anesthesia (RA) has evolved significantly in the last two decades, particularly with the widespread adoption of ultrasound guidance and the shift toward enhanced recovery after surgery (ERAS) protocols. Despite these advances, communication remains a critical—and often underappreciated—element of safe and effective RA practice. Whether in the block room, operating theatre, or educational setting, communication failures can contribute to adverse events, delays, inefficiencies, or suboptimal learning experiences.1 Artificial intelligence (AI) has recently emerged as a powerful tool in healthcare, with applications ranging from predictive analytics to clinical decision support. In the context of regional anesthesia, AI offers a particularly promising avenue for improving how information is conveyed—between clinicians, between humans and machines, and between educators and trainees. This article explores the ways in which AI can enhance communication in RA, with a particular focus on ultrasound-integrated platforms and workshop facilitation. The Communication Gap in Regional Anesthesia Regional anesthesia, unlike general anesthesia, is a highly interactive and stepwise process that depends on shared understanding between team members. Communication must occur at several levels: verifying the block type and laterality, confirming drug and dose, describing visual findings during ultrasound guidance, and handing over key information about catheter placement or local anesthetic infusion plans. In practice, this communication is often hampered by competing noise in the clinical environment, lack of standardization in language and terminology, and overreliance on verbal instructions without visual reinforcement. In educational settings such as workshops or live demonstrations, these challenges are amplified. Instructors may describe anatomical relationships verbally while simultaneously pointing to structures on an ultrasound screen, often without confirmation that learners have correctly understood what was seen or said. Learner misinterpretation is common, especially in large groups or when instructors are operating remotely. While checklists, time-outs, and standard procedural templates can help mitigate communication breakdowns, these interventions are often inconsistently applied. There remains a critical need for more integrated, real-time communication tools that support both safety and teaching. AI-Enhanced Ultrasound: Visualizing What We Say One of the most promising developments in AI is its ability to enhance visual communication during ultrasound-guided RA.2,3 Deep learning algorithms trained on thousands of labeled scans can now accurately identify and outline key anatomical structures—nerves, blood vessels, muscles, and fascial planes—directly on the ultrasound screen. This functionality is already being incorporated into commercial ultrasound platforms, offering significant benefits both clinically and educationally. In the clinical setting, AI-generated overlays improve shared understanding between the proceduralist and the assistant. Instead of needing to explain anatomy verbally or rely on gesturing toward the screen, the operator can focus on the task while the AI labels structures in real time. This can reduce ambiguity, speed up block placement, and help onboard less experienced staff with clearer visual references. In educational workshops, AI-supported ultrasound offers even greater value. It enables instructors to communicate anatomical relationships without needing to physically point, preserving sterility and reducing cognitive overload. Trainees benefit from seeing consistent, labeled imagery that reinforces the verbal explanation. This can be particularly helpful in cadaveric or phantom model courses, where real-time anatomic variability often complicates traditional instruction. Some platforms are also exploring interactive features that allow the instructor to highlight or annotate directly on the ultrasound image, either via voice command or touch interface, further supporting precise communication. Natural Language Processing and Procedural Documentation Another emerging area is the use of AI for real-time, voice-driven procedural documentation. Natural Language Processing (NLP), a subset of AI, enables systems to interpret spoken language and convert it into structured, actionable text. In regional anesthesia, this means that an anesthesiologist could narrate a procedure while performing it—for example, "Performing a left-sided transversus abdominis plane block with 20 mL of 0.25% bupivacaine"—and have this transcribed and automatically inserted into the patient’s electronic anesthesia record. This approach streamlines documentation, improves standardization, and reduces the risk of omitted details.4 It also frees the clinician from needing to manually record information during or after the procedure, enhancing workflow efficiency. Some experimental systems even use NLP to cross-check spoken data with clinical protocols or drug libraries, providing alerts if inconsistencies or safety risks are detected. From a communication standpoint, this technology formalizes and preserves what would otherwise be fleeting verbal statements, turning them into accessible, traceable data that can be reviewed by colleagues, supervisors, or researchers. AI could also be used to identify regional anesthesia procedures in historical unstructured notes to update registries or do retrospective analysis. Although this is a promising area, it still faces significant implementation problems.5 AI-Guided Checklists and Safety Prompts While checklists are already a part of good anesthetic practice, they are often static and reliant on manual input. AI systems offer the potential to deliver dynamic, context-sensitive checklists that evolve with the case. These systems can track the progress of a procedure through voice or data input and deliver prompts accordingly. For instance, if the AI system detects that the proceduralist is preparing a local anesthetic, it may prompt for confirmation of drug type, concentration, and volume. It might also remind the user to aspirate before injection or warn of potential contraindications based on patient data. Some prototypes respond to speech using a closed-loop communication model: Proceduralist: “Injecting 20 mL of 0.5% ropivacaine.” System: “Confirming: ropivacaine 0.5%, 20 mL. Proceed?” Such systems not only enhance safety but also reinforce shared situational awareness between team members, reducing the chance of misheard or misunderstood instructions. Communication Across the Block Room: Workflow and Coordination In busy institutions where multiple RA procedures are performed daily, communication challenges extend beyond individual blocks. AI-powered dashboards that aggregate scheduling, patient status, consent completion, and block progress can help coordinate efforts between anesthesia teams, surgeons, and operating room staff. These systems can, for example, flag delays in block completion, notify teams when a patient is ready for transfer, or alert anesthetists about potential allergy conflicts or missing documentation. By serving as a centralized communication hub, they reduce reliance on fragmented verbal updates or paper tracking sheets. This type of AI-supported communication promotes better alignment across the perioperative team and allows the regional anesthesia service to function more efficiently and transparently. AI in Regional Anesthesia Workshops and Education Workshops remain essential to the dissemination of RA techniques, particularly for newer fascial plane blocks or advanced catheter techniques. However, teaching these skills—especially with large groups or remote instruction—can be challenging. Communication must be clear, concise, and anatomically precise, yet is often undermined by poor visibility, variable skill levels among learners, and time constraints. AI-enhanced ultrasound systems improve workshop communication by standardizing anatomical labeling across all learner stations. This ensures that even novice participants can identify key structures without constant verbal correction. Instructors can focus on advanced teaching points without needing to repeatedly identify nerve bundles or fascial layers. Some platforms also offer automated scanning guidance, suggesting probe adjustments in real time to optimize image quality. These features act as a “silent assistant,” reinforcing instruction while reducing the instructor’s cognitive burden. Finally, AI systems that store and annotate scans allow trainees to revisit images post-workshop, extending the learning experience beyond the session itself. This is particularly valuable for remote learners or those in low-resource settings. Limitations and Considerations Despite these promising applications, AI systems for communication in regional anesthesia face important limitations. Integration into existing clinical infrastructure remains uneven, and many commercial ultrasound platforms have yet to adopt open AI interfaces. The cost of upgrading equipment may be prohibitive, especially for community hospitals or educational institutions in lower-income regions. There is also a risk of overreliance. If clinicians begin to trust AI labeling or voice prompts too heavily, they may overlook unexpected anatomical variants or clinical nuances. AI should be viewed as an adjunct, not a replacement, for clinical judgment. Furthermore, AI models are only as good as their training data. In patients with atypical anatomy—such as the obese, elderly, or those with prior surgery—AI overlays may be less accurate. Continuous validation and updating of AI systems are essential to maintaining safety and reliability. Privacy and data security present another layer of complexity. AI tools that record voice, video, or scan data must comply with GDPR, HIPAA, or similar regulations, particularly in teaching environments where patient anonymity must be preserved. Future Directions As the field matures, AI-driven communication tools will likely become more integrated and multimodal. We may soon see systems that combine speech recognition, ultrasound image interpretation, eye tracking, and gesture recognition to create fully interactive and intelligent block room environments. Remote collaboration platforms may also benefit. An expert anesthesiologist could supervise multiple procedures at satellite centers, providing real-time feedback with the support of AI-enhanced ultrasound streams and structured communication protocols. In education, AI may support procedural scoring, skill assessment, and feedback delivery, making RA workshops more objective and scalable. Conclusion Artificial intelligence offers a compelling opportunity to improve communication in regional anesthesia—whether between team members during clinical care or between instructors and learners in workshops. Through applications in ultrasound labeling, voice-driven documentation, dynamic checklists, and workflow dashboards, AI can make communication more efficient, precise, and resilient to distraction. However, these technologies must be integrated thoughtfully, with attention to clinical context, training, and human oversight. By enhancing clarity without adding complexity, AI can help overcome long-standing communication barriers in regional anesthesia and support the next generation of safe, effective, and well-taught anesthetic practice. References 1. Hashimoto, D. A., Witkowski, E., Gao, L., Meireles, O. & Rosman, G. Artificial Intelligence in Anesthesiology: Current Techniques, Clinical Applications, and Limitations. Anesthesiology 132, 379–394 (2020). 2. Bowness, J. S. et al. Artificial intelligence for ultrasound scanning in regional anaesthesia: a scoping review of the evidence from multiple disciplines. Br J Anaesth 132, 1049–1062 (2024). 3. Mika, S., Gola, W., Gil-Mika, M., Wilk, M. & Misiołek, H. Artificial Intelligence-Supported Ultrasonography in Anesthesiology: Evaluation of a Patient in the Operating Theatre. J Pers Med 14, (2024). 4. Huber, S. et al. Evaluating an AI Documentation Assistant for Anesthesiology Teams. in Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems 1–8 (ACM, New York, NY, USA, 2025). doi:10.1145/3706599.3706658. 5. Graham, L. A., Illarmo, S. S., Wren, S. M., Odden, M. C. & Mudumbai, S. C. Use of natural language processing method to identify regional anesthesia from clinical notes. Reg Anesth Pain Med 50, 271–275 (2025).

Borislavova MARGARITA, Liesbeth BRULLOT, Steve COPPENS (Leuven, Belgium)

11:58 - 12:20 Q&A.