CleanCTG: A Deep Learning Model for Multi-Artefact Detection and Reconstruction in Cardiotocography

2025-08-11

AI Summary

Fetal heart rate (FHR) traces from cardiotocography (CTG) are highly susceptible to multiple types of artifacts, obscuring true physiological patterns and increasing misdiagnosis risk. Existing approaches either rely solely on traditional filtering for missing-value imputation or employ deep learning models that optimize downstream classification without explicitly modeling artifacts—both limiting accurate signal restoration. This paper proposes an end-to-end two-stage deep learning framework: Stage I performs fine-grained, multi-class artifact detection using multi-scale convolution and context-aware cross-attention; Stage II enables artifact-specific signal reconstruction via dedicated parallel branches. Training leverages synthetically generated physiological data augmented with real clinical recordings. Experiments show perfect detection performance (AUROC = 1.00) on synthetic data, >60% reduction in reconstruction error, and AUROC = 0.95 on clinical validation. Integrated into the Dawes–Redman decision system, our method reduces clinical decision time by 33% and improves specificity to 82.70%.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Machine Learning: Calibration & Uncertainty Quantification
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Cardiotocography (CTG) is essential for fetal monitoring but is frequently compromised by diverse artefacts which obscure true fetal heart rate (FHR) patterns and can lead to misdiagnosis or delayed intervention. Current deep-learning approaches typically bypass comprehensive noise handling, applying minimal preprocessing or focusing solely on downstream classification, while traditional methods rely on simple interpolation or rule-based filtering that addresses only missing samples and fail to correct complex artefact types. We present CleanCTG, an end-to-end dual-stage model that first identifies multiple artefact types via multi-scale convolution and context-aware cross-attention, then reconstructs corrupted segments through artefact-specific correction branches. Training utilised over 800,000 minutes of physiologically realistic, synthetically corrupted CTGs derived from expert-verified "clean" recordings. On synthetic data, CleanCTG achieved perfect artefact detection (AU-ROC = 1.00) and reduced mean squared error (MSE) on corrupted segments to 2.74 x 10^-4 (clean-segment MSE = 2.40 x 10^-6), outperforming the next best method by more than 60%. External validation on 10,190 minutes of clinician-annotated segments yielded AU-ROC = 0.95 (sensitivity = 83.44%, specificity 94.22%), surpassing six comparator classifiers. Finally, when integrated with the Dawes-Redman system on 933 clinical CTG recordings, denoised traces increased specificity (from 80.70% to 82.70%) and shortened median time to decision by 33%. These findings suggest that explicit artefact removal and signal reconstruction can both maintain diagnostic accuracy and enable shorter monitoring sessions, offering a practical route to more reliable CTG interpretation.

Problem

• Detects and corrects multiple artefacts in fetal heart rate monitoring
• Improves accuracy and reliability of cardiotocography (CTG) interpretation
• Reduces misdiagnosis and delays in clinical decision-making

Innovation

• Dual-stage model for artefact detection and reconstruction
• Multi-scale convolution and context-aware cross-attention
• Artefact-specific correction branches for signal reconstruction