Self-Paced Learning for Images of Antinuclear Antibodies

2025-11-26

AI Summary

ANA fluorescence image analysis poses challenges in multi-instance, multi-label (MIML) modeling due to clinical annotation uncertainty and complex pattern co-occurrence. This paper proposes an end-to-end adaptive MIML framework that eliminates manual image preprocessing and emulates expert visual reasoning by automatically localizing consistent fluorescent regions and aggregating instance-level predictions into image-level classifications. A novel self-paced learning mechanism is introduced, jointly comprising an instance sampler and a probabilistic pseudo-label allocator, which dynamically models pattern confidence, adaptively adjusts sample weights, and enhances instance discriminability—thereby mitigating annotation noise and interference from antibody coexistence. On the ANA dataset, our method achieves new state-of-the-art performance, improving F1-Macro by 7.0% and mAP by 12.6%. Across multiple medical MIML benchmarks, it reduces Hamming Loss by 18.2% and One-Error by 26.9%, consistently ranking among the top two methods overall.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Medical and Biological Imaging
• Multiagent Systems: Multiagent Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Antinuclear antibody (ANA) testing is a crucial method for diagnosing autoimmune disorders, including lupus, Sjögren's syndrome, and scleroderma. Despite its importance, manual ANA detection is slow, labor-intensive, and demands years of training. ANA detection is complicated by over 100 coexisting antibody types, resulting in vast fluorescent pattern combinations. Although machine learning and deep learning have enabled automation, ANA detection in real-world clinical settings presents unique challenges as it involves multi-instance, multi-label (MIML) learning. In this paper, a novel framework for ANA detection is proposed that handles the complexities of MIML tasks using unaltered microscope images without manual preprocessing. Inspired by human labeling logic, it identifies consistent ANA sub-regions and assigns aggregated labels accordingly. These steps are implemented using three task-specific components: an instance sampler, a probabilistic pseudo-label dispatcher, and self-paced weight learning rate coefficients. The instance sampler suppresses low-confidence instances by modeling pattern confidence, while the dispatcher adaptively assigns labels based on instance distinguishability. Self-paced learning adjusts training according to empirical label observations. Our framework overcomes limitations of traditional MIML methods and supports end-to-end optimization. Extensive experiments on one ANA dataset and three public medical MIML benchmarks demonstrate the superiority of our framework. On the ANA dataset, our model achieves up to +7.0% F1-Macro and +12.6% mAP gains over the best prior method, setting new state-of-the-art results. It also ranks top-2 across all key metrics on public datasets, reducing Hamming loss and one-error by up to 18.2% and 26.9%, respectively. The source code can be accessed at https://github.com/fletcherjiang/ANA-SelfPacedLearning.

Problem

• Automating antinuclear antibody detection from unprocessed microscope images
• Addressing multi-instance multi-label learning challenges in clinical settings
• Overcoming limitations of manual analysis through self-paced learning framework

Innovation

• Uses instance sampler to suppress low-confidence patterns
• Employs probabilistic dispatcher for adaptive label assignment
• Applies self-paced learning with empirical weight adjustment