A Multi-Stage Deep Learning Framework with PKCP-MixUp Augmentation for Pediatric Liver Tumor Diagnosis Using Multi-Phase Contrast-Enhanced CT

2025-11-22

AI Summary

Accurate differentiation between benign and malignant pediatric hepatic tumors—and their histopathological subtyping—currently relies on invasive biopsy, which poses significant risks in children due to rich vascularity, tissue fragility, and poor procedural compliance, leading to elevated hemorrhage risk and substantial anesthetic burden. Addressing the scarcity of pediatric imaging data, severe class imbalance, and limited AI adoption in this domain, we propose a novel multi-stage deep learning framework. First, automatic tumor detection is performed on multiphase contrast-enhanced CT scans (mAP = 0.871); second, a dual-stage, triple-backbone classification model achieves benign–malignant discrimination (AUC = 0.989) and fine-grained subtyping (benign AUC = 0.915; malignant AUC = 0.979). We introduce PKCP-MixUp—a pathology-knowledge-constrained data augmentation strategy—and integrate SHAP-based feature attribution with CAM-based activation visualization to enhance model interpretability. This framework enables non-invasive, accurate, and interpretable intelligent diagnosis of pediatric liver tumors, establishing the first end-to-end AI solution for this clinical challenge.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Interpretability, Explainability, and Transparency
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Pediatric liver tumors are one of the most common solid tumors in pediatrics, with differentiation of benign or malignant status and pathological classification critical for clinical treatment. While pathological examination is the gold standard, the invasive biopsy has notable limitations: the highly vascular pediatric liver and fragile tumor tissue raise complication risks such as bleeding; additionally, young children with poor compliance require anesthesia for biopsy, increasing medical costs or psychological trauma. Although many efforts have been made to utilize AI in clinical settings, most researchers have overlooked its importance in pediatric liver tumors. To establish a non-invasive examination procedure, we developed a multi-stage deep learning (DL) framework for automated pediatric liver tumor diagnosis using multi-phase contrast-enhanced CT. Two retrospective and prospective cohorts were enrolled. We established a novel PKCP-MixUp data augmentation method to address data scarcity and class imbalance. We also trained a tumor detection model to extract ROIs, and then set a two-stage diagnosis pipeline with three backbones with ROI-masked images. Our tumor detection model has achieved high performance (mAP=0.871), and the first stage classification model between benign and malignant tumors reached an excellent performance (AUC=0.989). Final diagnosis models also exhibited robustness, including benign subtype classification (AUC=0.915) and malignant subtype classification (AUC=0.979). We also conducted multi-level comparative analyses, such as ablation studies on data and training pipelines, as well as Shapley-Value and CAM interpretability analyses. This framework fills the pediatric-specific DL diagnostic gap, provides actionable insights for CT phase selection and model design, and paves the way for precise, accessible pediatric liver tumor diagnosis.

Problem

• Developing non-invasive AI diagnosis for pediatric liver tumors using CT
• Addressing data scarcity and class imbalance with PKCP-MixUp augmentation
• Differentiating benign vs malignant tumors and their pathological subtypes

Innovation

• Multi-stage deep learning framework for tumor diagnosis
• PKCP-MixUp augmentation addresses data scarcity
• Two-stage diagnosis pipeline with ROI-masked images