TACL: Threshold-Adaptive Curriculum Learning Strategy for Enhancing Medical Text Understanding

2025-10-16

AI Summary

Medical texts—particularly electronic medical records (EMRs)—are highly unstructured, domain-specific, and exhibit substantial contextual variability, leading to poor model generalization on complex or rare clinical cases. To address this, we propose a dynamic curriculum learning framework grounded in sample complexity assessment: first, we design a multi-dimensional complexity metric to quantify the difficulty of clinical texts; second, we introduce a threshold-adaptive mechanism to dynamically partition samples into difficulty levels and optimize training sequencing; third, we integrate a multilingual medical text encoder to jointly support diverse tasks—including ICD coding, readmission prediction, and Traditional Chinese Medicine (TCM) syndrome identification. This work is the first to systematically apply progressive curriculum learning to multilingual EMR analysis. Experiments on both Chinese and English datasets demonstrate significant improvements in accuracy on complex samples and enhanced model robustness, establishing a scalable paradigm for cross-lingual, multi-task medical NLP.

Technology Category

• Machine Learning: Multimodal Learning
• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Multiagent Systems: Multiagent Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Medical texts, particularly electronic medical records (EMRs), are a cornerstone of modern healthcare, capturing critical information about patient care, diagnoses, and treatments. These texts hold immense potential for advancing clinical decision-making and healthcare analytics. However, their unstructured nature, domain-specific language, and variability across contexts make automated understanding an intricate challenge. Despite the advancements in natural language processing, existing methods often treat all data as equally challenging, ignoring the inherent differences in complexity across clinical records. This oversight limits the ability of models to effectively generalize and perform well on rare or complex cases. In this paper, we present TACL (Threshold-Adaptive Curriculum Learning), a novel framework designed to address these challenges by rethinking how models interact with medical texts during training. Inspired by the principle of progressive learning, TACL dynamically adjusts the training process based on the complexity of individual samples. By categorizing data into difficulty levels and prioritizing simpler cases early in training, the model builds a strong foundation before tackling more complex records. By applying TACL to multilingual medical data, including English and Chinese clinical records, we observe significant improvements across diverse clinical tasks, including automatic ICD coding, readmission prediction and TCM syndrome differentiation. TACL not only enhances the performance of automated systems but also demonstrates the potential to unify approaches across disparate medical domains, paving the way for more accurate, scalable, and globally applicable medical text understanding solutions.

Problem

• Addresses unstructured medical text complexity for automated understanding
• Improves model generalization on rare and complex clinical cases
• Enhances multilingual medical text analysis across diverse clinical tasks

Innovation

• Dynamically adjusts training based on sample complexity
• Categorizes data into difficulty levels for progressive learning
• Prioritizes simpler cases before tackling complex records