Portrait de Dan Poenaru

Dan Poenaru

Membre académique associé
Professeur, McGill University
Sujets de recherche
Apprentissage automatique médical
IA en santé
IA et santé

Biographie

Dan Poenaru est professeur de chirurgie pédiatrique à l’Université McGill et chercheur principal à l’Institut de recherche du Centre universitaire de santé McGill, à Montréal. Il est titulaire d’une maîtrise en éducation aux professions de la santé et en développement international, et d’un doctorat en stratégie et gestion de la santé.

Chercheur financé par le Fonds de recherche du Québec - Santé (FRQS) et les Instituts de recherche en santé du Canada (IRSC) dans le domaine des soins chirurgicaux centrés sur le patient, il est également chef du laboratoire CommiSur de l’Université McGill, directeur de la bourse Jean-Martin-Laberge en chirurgie pédiatrique mondiale et membre fondateur de l’Initiative mondiale pour la chirurgie infantile (GICS).

Ses domaines d’intérêt actuels sont la communication chirurgicale et l’enseignement médical assistés par la technologie, y compris l’IA, la réalité virtuelle et les dispositifs de santé numériques, les soins chirurgicaux centrés sur le patient et le développement de la capacité de recherche chirurgicale mondiale.

Étudiants actuels

Publications

When blood is needed but not available: A proposal for measuring patient‐level blood unavailability
Ayla Gerk
Meghan Delaney
DIGOM: Disentangled Interaction Graphs for Outcome prediction in Multimodal healthcare
Multimodal healthcare data contain relational information that extends beyond individual patient observations. Effective clinical outcome pr… (voir plus)ediction therefore requires not only integrating information from multiple modalities, but also modeling the relationships between them. Standard fusion approaches do not explicitly capture these interactions, while existing graph-based approaches typically model only a subset of them. We propose DIGOM (Disentangled Interaction Graphs for Outcome prediction in Multimodal healthcare), a graph-based framework that jointly models three complementary interaction types: Intra-patient cross-modal, Cross-patient same-modal, and Cross-patient cross-modal interactions. DIGOM learns a sparse multiplex graph with disentangled views, each encoding interactions between pairs of modalities across patients. Discrete adjacency matrices with patient-adaptive connectivity are sampled via a differentiable mechanism, enabling end-to-end learning and allowing the model to select informative relationships without imposing a fixed number of neighbors. A regularization objective maintains overall graph sparsity, while message passing integrates this relational context into patient representations for outcome prediction. We evaluate DIGOM on two multimodal healthcare datasets, MIMIC-IV and NIH-TB, across five clinical prediction tasks. DIGOM achieves the highest AUROC on all five tasks while using fewer connections than existing multi-interaction approaches. Ablation studies reveal that each interaction type's contribution varies across tasks, highlighting the need to adapt relational structures per task. DIGOM also demonstrates greater robustness to missing modalities, suggesting learned relationships provide complementary information when modalities are unavailable. Post-hoc analyses further reveal task- and outcome-specific connectivity patterns in the learned graphs.
Complementary Vulnerabilities: Auditing and Governing Clinician-LLM Error Overlap
Kathleen Fraser
Health systems are deploying large language models (LLMs) into clinical workflows, often on the assumption that a second automated check can… (voir plus) only reduce error. But a second reasoner reduces error only insofar as it fails differently from the first. We say that a clinician and a model have complementary vulnerabilities to the extent that they break down through different mechanisms and on different cases. The two reasoners' failure substrates differ: they are embodied and cognitive in the clinician, and statistical and generative in the model - so their weak points need not line up. That divergence is plausible rather than guaranteed, and the independence it creates is fragile, eroding when clinicians defer to the model or the model conforms to the clinician. We therefore propose a predeployment audit in which each site measures error overlap on its own cases. We also propose a governance model with which the site keeps those errors independent over time. Health systems should then judge a clinical artificial intelligence (AI) tool by whether its failures overlap with their clinicians' failures, as well as by its accuracy.
Diagnosing Low-Confidence Patient Profiles via Localized Harmful Covariate Shift Detection
Before deploying clinical prediction models, their reliability must be assessed to ensure safe use. Errors may arise from distribution shift… (voir plus)s between training and deployment or from heterogeneous performance where the model systematically underperforms for certain patient subgroups. Yet most post-hoc diagnostic methods evaluate reliability only at the population level, checking whether the deployment data resemble the training data and whether the model performs well on average, without accounting for heterogeneity that may cause some patients to be disproportionately affected. We propose a unified diagnostic pipeline that combines both perspectives. MED3pa first discovers interpretable low-confidence patient profiles, and Detectron is then applied within each profile to test whether its degradation is associated with a detectable harmful covariate shift. Across three clinical prediction tasks and three injected subgroup shifts in MIMIC-IV, we find that shift detectability itself is informative. An injected age shift, undetectable at the population level, was nonetheless localized to specific degraded profiles, whereas sex- and race-related shifts were detected broadly and therefore offered little guidance for identifying the most affected patients.
Synthetic Content Validation of Pediatric Trust Instruments Using Persona-Driven Large Language Models: Development and Evaluation Study
Katya Loban
Suganthi Rajasegaran
Elena Guadagno
Jeannie L Haggerty
Sylvie Lambert
Background: Large language models (LLMs) could assist in streamlining health care instrument validation. This is particularly relevant as tr… (voir plus)aditional instrument development is time-intensive and resource-intensive. Currently, there is a significant gap in specifically developed and validated trust instruments for pediatric emergency and surgical contexts. Such tools are key because trust is foundational to the patient family-physician relationship and is associated with improved care-seeking and treatment adherence. Objective: This study aimed (1) to develop new trust instruments and conduct their preliminary content validation confirming item relevance through a Synthetic Instrument Validation approach integrating human and LLM capabilities and (2) to evaluate appropriate use cases for LLMs in psychometric assessment. Methods: Two trust instruments were developed (patient family and physician versions). In phase 1, instruments underwent a 2-stage content validation process using parallel synthetic and human expert panels (5 patient family members, 3 physician synthetic experts, and 7 unique human experts). Synthetic panels consisted of 3 persona-prompted LLMs (Claude Sonnet 4, GPT-5, and Grok 4), with human panels serving as comparators. The scale-content validity index (S-CVI) threshold was set at ≥0.80, percent agreement on item relevancy was used to compare panels. In phase 2, LLM performance in quantitative research tasks was evaluated. The patient family instrument underwent Flesch-Kincaid assessment; both instruments underwent cosine similarity analyses using algorithmic and LLM-based methods. Results: In phase 1, percent agreement between panels for the patient family and physician instruments was 90.9% and 88.9% for dimensional validation and 90.9% and 100% for contextual validation. Fleiss κ for within-panel agreement ranged from poor to almost perfect; 0.20 (95% CI -0.23 to 0.63) to 1.00 (95% CI 1.00-1.00). All instrument subsections met or exceeded S-CVI thresholds across both rounds and panels. Phase 2 Flesch-Kincaid metrics converged across all 3 methods to target benchmark (mean Grade Level 8.1, SD 1.1; mean Reading Ease 60.1, SD 5.6), while reference readability was near target, showing convergence across implementations of the same formula. In contrast, cosine similarity analyses revealed significant LLM performance limitations; cosine similarity analyses were, therefore, abandoned in favor of the algorithmic method alone, which achieved a maximum cosine similarity of 0.83, indicating acceptable item distinctiveness overall. Conclusions: Persona-prompted LLMs effectively assisted preliminary content validation and may help reduce timelines from months to weeks, but showed limitations in quantitative computations. This suggests that, in this study, persona-prompted LLMs performed better in subjective rating tasks than in deterministic cosine similarity computations. These findings help illustrate task-dependent boundaries for LLM integration in psychometric research, necessitating selective human-LLM collaboration. While these instruments achieved preliminary content validity, they require full psychometric testing before they can be considered validated for clinical use. Synthetic Instrument Validation offers a feasible approach to preliminary health care instrument development while preserving key elements of content validation.
Need for ethical use of artificial intelligence in humanitarian data collection to address aid shortfalls
Ayla Gerk
Duha Shellah
Dunya Moghul
Justina Seyi-Olajide
The global humanitarian crisis is escalating due to conflicts, climate-related disasters, and infection outbreaks.1 Recent cuts by the USA a… (voir plus)nd other high-income donors to international aid, including considerable reductions in foreign assistance, have abruptly diminished crucial support for global health programmes, food security, and disaster relief.2 With overburdened humanitarian systems facing worsening resource constraints, data-driven aid allocation has become increasingly important for survival worldwide.
Evaluative Stance Toward Artificial Intelligence in High-Quartile Medical Journals (2021– 2026): Large-Scale LLM-Assisted Computational Content Analysis
Lixing Wang
Abstract Background Medical-AI publications do more than report technical performance; they also frame AI as beneficial, uncertain, or risky… (voir plus). How this evaluative stance has changed across the medical literature is not well characterized. Objective To characterize evaluative stance in published medical-AI discourse abstracts from January 2021 through April 2026 and examine variation over time, concern themes, failure mechanisms, specialties, first-author geography, and publication format. Methods We conducted an LLM-assisted computational content analysis of medical-AI abstracts from first- and second-quartile medical journals. Of 97,492 post-cutoff Q1/Q2 records entering the prefilter, 16,759 were retained as discourse or evaluative. Claude Sonnet 4.6 assigned 16,749 valid stance classifications using Alarm, Caution, Neutral, Cautious Optimism, and Advocacy. Annual analyses used 16,747 records dated 2021–2026. Critical stance was Alarm plus Caution and indexed evaluative scrutiny, rather than opposition or author psychology. Each LLM step was validated against blinded human coding by one author: prefilter Cohen κ=0.51, stance quadratic-weighted κ=0.79 (95% CI 0.72–0.84) for codable, in-scope records, specialty κ=0.75, and mechanism axes κ=0.84 (model type) and κ=0.57 (failure mode). Results Advocacy declined from 2.9% in 2021 to 0.6% in partial 2026, while Cautious Optimism remained the majority stance. Among 16,749 valid classifications, 30.8% were critical. Critical share increased from 25.4% to 32.6%, a 7.25-percentage-point increase based on unrounded estimates. Among critical records, patient safety remained the most prevalent concern. Hallucination/errors increased by 30.9 percentage points. Regulation declined by 22.0 percentage points and ethics/bias by 8.1 percentage points in prevalence share; these declines do not necessarily indicate lower publication counts. Within the hallucination/error theme, factual error was more common than fabrication. Fabrication estimates should be treated as an upper bound, because failure-mode agreement was moderate. Specialty patterns were heterogeneous. Critical rate was inversely associated with FDA cleared-device availability (Spearman ρ=−0.65, two-sided p=0.004), which does not measure adoption, deployment, maturity, or clinical use. First-author geography described publication metadata and discourse, not national attitudes or research quality. Reviews were the least critical and most favourable format. In exploratory forward-validation, 2 of 78 early Advocacy predictions were fully borne out, although the analysis was single-rater and retrieval-dependent. Conclusions Published medical-AI abstracts became modestly less promotional and more focused on specific errors and safety concerns. Unqualified promotion declined, but qualified favourable framing remained dominant, and the rise in critical stance was modest. Concern moved toward errors and patient safety, with factual error discussed more often than fabrication. These findings describe published discourse, not AI capability or whether the evaluations were correct.
Artificial intelligence-assisted ganglion cell detection in Hirschsprung's disease: A comparative evaluation of two deep learning approaches
E Wang
Karl Grenier
Peter Savadjiev
Background. Definitive diagnosis of Hirschsprung's disease (HD) requires pathological identification of enteric ganglion cells. This process… (voir plus) is time-consuming and subject to inter-observer variability. Artificial intelligence (AI) tools have the potential to standardize and accelerate this workflow, but no study has determined which AI approach best serves intraoperative HD pathology diagnostics. Method. This study compared the U-Net and You Only Look Once version 26 (YOLO26) frameworks for ganglion cell detection using a single-centre retrospective dataset of 54 whole-slide images (WSIs) from rectal biopsies. WSIs were tiled into 397,731 image patches (128x128 pixels), further partitioned into training (70%), validation (15%), and testing (15%) sets. Models were evaluated on tile- and patient-level diagnostic metrics and processing latency. Results. The U-Net achieved a tile-level sensitivity of 82.9%, showing no statistically significant difference compared to YOLO26 (79.1%; p = 0.097). However, YOLO26 demonstrated a statistically significant advantage in tile-level specificity (96.1% vs. 93.9%; p < 0.001) and reduced mean inference latency (7.64 ms vs. 11.57 ms/tile). At the patient level, both models achieved 100% diagnostic sensitivity. Despite low patient-level specificity (0.0% U-Net; 11.8% YOLO26), the tissue-level diagnostic burden of false positives was 6.00% for U-Net and 3.50% for YOLO26. Conclusion. The U-Net is preferred when nominal gains in sensitivity are prioritized, while the YOLO26 is an alternative that optimizes efficiency and false positive suppression. Both models serve as robust screening filters to augment the pathologist's workflow and should be selected based on workflow requirements. Prospective validation on larger, multi-centre datasets is required before clinical implementation.
Coconstructing CHAMP, an Artificial Intelligence Chatbot for Pediatric Infectious Symptoms Management: Protocol for a Multiphase Participatory Study
Jia Lin
Nikhil Jaiswal
Yuanchao Ma
Bertrand Lebouché
Sebastian Villanueva
Sofiane Achiche
David Lessard
Kim Engler
Simon Berthelot
David Llewellyn Buckeridge
Leo Anthony Celi
Isabelle Gagnon
Jocelyn Gravel
Laurie Hayes Plotnick
Marie‐Pascale Pomey
Zoua M. Vang
Esli Osmanlliu
Aida Fouad Margaux Elisabete JoAnne Annie-Danielle Maria Brenna Lisa Paul Sonia Amy Kofi Arleth Marie-José Berberovic
Background: Acute infectious symptoms are a leading cause of pediatric emergency department visits in Canada, many of which are low acuity a… (voir plus)nd could be safely managed at home. Artificial intelligence (AI) chatbots offer a promising avenue for delivering accessible, evidence-based guidance to support families in managing these symptoms. Objective: This study aims to adapt and coconstruct CHAMP (CHatbot to Assist the Management of Pediatric patients), an AI chatbot to support patients and families with acute pediatric infectious symptoms. CHAMP aims to deliver timely, tailored, and validated health information to support safe at-home self-management and informed care-seeking. Methods: This multiphase, mixed methods participatory study will be conducted at the Montreal Children's Hospital in Montreal, Quebec, Canada. A coconstruction committee comprised of youth, parents, caregivers, and partners will be engaged as coresearchers. Eligible participants will include (1) youth aged 14-17 years and (2) parents and caregivers of children aged 0-17 years. The study comprises 5 phases. Phase 1 involves a qualitative needs assessment using focus groups with 20 participants to explore their informational needs, preferences, and concerns regarding pediatric infections and the use of AI chatbots. Phase 2 focuses on coconstructing and validating CHAMP's knowledge database through 3-5 workshops. Coresearchers will review pediatric clinical guidelines, map care questions and decision-making processes, and shape CHAMP's conversational framework. Phase 3 consists of iterative prototyping and testing through 3-5 workshops. Coresearchers will engage in prototyping and scenario testing, alongside preliminary usability and acceptability assessments. Phase 4 examines equity and accessibility through focus groups with 20 participants at risk of digital exclusion, as well as multilingual evaluation of an automated large language model-based translation layer. Phase 5 uses collaborative ethnography to explore the process of participatory coconstruction and its impact on CHAMP's design. Results: Funding was secured in 2024, and Research Ethics Board approval was obtained in December 2024. As of December 2025, the coconstruction committee is being assembled, and Phase 1 recruitment is underway. Conclusions: This study will produce a functioning CHAMP prototype grounded in participatory, equitable, and responsible pediatric AI development. Findings will inform usability testing and an implementation-effectiveness evaluation, contributing to best practices for pediatric-centered AI health tools. By providing timely, tailored, and validated health information on acute infections, CHAMP may support safe at-home self-management, reduce preventable emergency department visits, ensure at-risk children are directed to appropriate care, and improve patient and family health care experiences.
Assessing Language Bias in Pediatric Surgical Systematic Reviews: A Meta-epidemiological Study.
Dunya Moghul
Elena Guadagno
Robert Baird
Patient safety culture in the operating room of African hospitals: a systematic review
Jacques Fadhili Bake
Naïcen Ghanmi
Elena Guadagno
K. M. Claude
Tsongo Kibendelwa Zacharie
Patient safety in operating rooms has globally improved through interventions such as the World Health Organization (WHO) Surgical Safety Ch… (voir plus)ecklist and multidisciplinary team training. However, while evidence from high-income countries is well documented, there remains limited consolidated knowledge on the understanding, application, and effectiveness of safety culture interventions in African surgical settings, which this review seeks to address. This systematic review examined factors and protocols affecting surgical safety in African operating rooms. We hypothesized that persistent systemic barriers undermine safety culture despite adoption of global measures. Following PRISMA 2020, we searched eight databases (Medline, Embase, Cochrane, Africa-Wide, CINAHL, Global Health, Global Index Medicus, Web of Science) from inception to 5 December 2024, using variations of text words present in the title, abstract, or keyword fields, alongside relevant subject headings, to identify articles addressing surgical safety and culture throughout Africa. Included studies involved operating room professionals in African countries and used quantitative, qualitative, or mixed-methods designs. We excluded non-operating room settings, patient-only studies, inaccessible full texts, reviews, editorials, letters, conference abstracts, and duplicates. Two reviewers independently screened and appraised studies using the Mixed Methods Appraisal Tool. Findings were synthesized narratively with subgroup analysis by study type and theme. Out of 9,875 identified records, 22 studies from 12 African countries (2014–2024) met inclusion criteria, with Ethiopia contributing the highest number (n = 4). Various assessment tools, including the Hospital Survey on Patient Safety Culture, the Safety Attitudes Questionnaire, and the National Surgical, Obstetric, and Anaesthesia Plans interview manual, revealed recurring challenges: inadequate non-punitive responses to errors, communication barriers, hierarchical structures, and resource constraints. Four interventions showed promise: implementation and training on the WHO Surgical Safety Checklist, Safe Surgery 2020 initiatives, Non-Technical Skills for Surgeons training, and multidisciplinary training. The heterogeneity of study designs, sample sizes, and outcome measures limited direct comparisons and precluded meta-analysis. Nonetheless, the review highlights persistent barriers and emerging opportunities to strengthen patient safety culture in African operating rooms. While the WHO Surgical Safety Checklist remains valuable, sustainable progress requires multi-level strategies that address systemic constraints and incorporate context-sensitive adaptations. PROSPERO, CRD42024627076.
Large language models for electronic health records in pediatric and surgical care: a systematic review.
Waseem Abu-Ashour
Elena Guadagno