Mosbah Aouad | Searcharxiv

arXiv2025

Early Detection of Pancreatic Cancer Using Multimodal Learning on Electronic Health Records

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancers, and early detection remains a major clinical challenge due to the absence of specific symptoms and reliable biomarkers. In this work, we propose a new multimodal approach that integrates longitudinal diagnosis code histories and routinely collected laboratory measurements from electronic health records to detect PDAC up to one year prior to clinical diagnosis. Our method combines neural controlled differential equations to model irregular lab time series, pretrained language models and recurrent networks to learn diagnosis code trajectory representations, and cross-attention mechanisms to capture interactions between the two modalities. We develop and evaluate our approach on a real-world dataset of nearly 4,700 patients and achieve significant improvements in AUC ranging from 6.5% to 15.5% over state-of-the-art methods. Furthermore, our model identifies diagnosis codes and laboratory panels associated with elevated PDAC risk, including both established and new biomarkers. Our code is available at https://github.com/MosbahAouad/EarlyPDAC-MML.

cs.LG↗

arXiv2023

RACR-MIL: Rank-aware contextual reasoning for weakly supervised grading of squamous cell carcinoma using whole slide images

Squamous cell carcinoma (SCC) is one of the most common cancer subtype, with an increasing incidence and a significant impact on cancer-related mortality. SCC grading using whole slide images is inherently challenging due to the lack of a standardized grading protocol and substantial tissue heterogeneity. We propose RACR-MIL, a weakly-supervised SCC grading approach that achieves robust generalization across multiple anatomies (skin, head & neck, lung). RACR-MIL is an attention-based multiple-instance learning framework that introduces two key innovations for learning grade-specific contextual representations: (1) a hybrid WSI graph that captures both local tissue context and non-local phenotypic dependencies between tumor regions, and (2) rank-ordering constraints on the attention mechanism that encourage consistent prioritization of higher-grade tumor regions and improve region-level grade confidence, aligning with pathologist's diagnostic process. Our model achieves state-of-the-art performance across multiple SCC datasets, achieving 3-9% improvements over existing methods and up to 10% improvement in tumor localization. In a pilot study, pathologists reported that RACR-MIL improved grading efficiency in 60% of cases, underscoring its potential as a clinically viable cancer diagnosis and grading assistant.

cs.CV↗