Implication: Opportunistic, scalable, and cost-effective screening—especially useful in resource-limited settings.

Results: Strong generalization and superior AUC vs single modality and foundation models.

Key innovations:. Cross-modal co-attention for CXR-ECG alignment. Causal inference to handle confounders. Dual back-propagation for de-biasing.

We developed MOSCARD: a novel multimodal framework aligning CXR with ECG via causal reasoning.

Existing risk models often rely on single modalities or clinical scores—limited by bias and incomplete data.

Codebase

MACE remains the leading global cause of death. Our MICCAI 2025 work, MOSCARD, fuses CXR+ECG via multimodal causal reasoning for bias-aware risk prediction. Outperforms SOTA on ED & MIMIC (AUC: 0.75–0.83). #MICCAI2025 #AIHealth.

🔓 The full pipeline and model are publicly available under an academic license—ready to support clinical and research applications. (.

🧠 Evaluation used both:. Manual expert annotations.LLM-based metrics: G-eval and Prometheus.

The fine-tuned model was evaluated on:. Internal test set (Mayo Clinic, n=80).External test set (MIMIC-III, n=123).Large-scale validation (MIMIC-IV, n=5000).

🧪 Our two-phase approach fine-tunes LLMs on 15,000 unlabeled Mayo Clinic reports, using heuristics and domain knowledge to guide learning.

We leverage weak supervision + LLMs to identify critical terms without relying on large-scale labeled datasets.

📢 New paper: We propose a framework that leverages weak supervision to train LMs for structured information extraction—minimizing annotation cost while maintaining clinical relevance. #NLP #RadiologyAI #WeakSupervision #ClinicalNLP #LLMs

RT @NIH: “It is absolutely vital that NIH investments are geographically dispersed. The way we combat scientific groupthink is by empowerin….

Population drift is breaking your clinical #AI. We built a causal reasoning model that actually generalizes - across hospital, ethnicities, and risk levels - for 1 year MACE prediction from chest X-rays. Better fairness, Real-world readiness.#AI #Medtech #CausalInference.

We propose a categorization framework for #VLM studies, accompanied by tailored reporting standards that address key aspects including performance evaluation, data reporting protocols, and manuscript composition guidelines.

Traditional ML reporting fall short for multiphase #VLM studies. We argue for a restructuring that balances intuitive clarity for developers with rigorous reproducibility. #AI #MachineLearning #MedAI #Reproducibility #MLStandards

@MayoClinic @MayoRadiologyAZ.

Current methodologies in disease predictions overlook the significant impact of chronic comorbidities. We propose a causal reasoning framework to address selection bias in opportunistic screening for 1-year composite MACE risk using chest X-ray.

causal+confounder model demonstrated more consistent performance across age groups, with AUC values of 0.701, 0.694, and 0.710 (within +/- 0.01), indicating reduced age-related bias.