Toward Using Machine Learning as a Shape Quality Metric for Liver Point Cloud Generation

2025-08-04

AI Summary

Existing evaluation methods for 3D liver shape generation models lack individualized, reference-free quality assessment, relying either on expert annotations or ground-truth shapes. Method: We propose a lightweight, interpretable proxy evaluation framework that jointly leverages handcrafted geometric features and PointNet to perform binary classification (high- vs. low-quality) directly on generated liver point clouds—eliminating dependence on clinical ground truth or expert labeling. Contributions/Results: (1) We introduce the first individualized liver shape quality discriminator; (2) we address the limitation of distribution-level metrics (e.g., FID) in single-sample assessment; and (3) we provide clinically aligned, fine-grained quality feedback. Experiments demonstrate strong agreement with expert assessments (Cohen’s κ = 0.82), significantly improving generative model iteration efficiency and evaluation transparency.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Natural Language Processing: Generation
• Computer Vision: 3D Computer Vision

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

While 3D medical shape generative models such as diffusion models have shown promise in synthesizing diverse and anatomically plausible structures, the absence of ground truth makes quality evaluation challenging. Existing evaluation metrics commonly measure distributional distances between training and generated sets, while the medical field requires assessing quality at the individual level for each generated shape, which demands labor-intensive expert review. In this paper, we investigate the use of classical machine learning (ML) methods and PointNet as an alternative, interpretable approach for assessing the quality of generated liver shapes. We sample point clouds from the surfaces of the generated liver shapes, extract handcrafted geometric features, and train a group of supervised ML and PointNet models to classify liver shapes as good or bad. These trained models are then used as proxy discriminators to assess the quality of synthetic liver shapes produced by generative models. Our results show that ML-based shape classifiers provide not only interpretable feedback but also complementary insights compared to expert evaluation. This suggests that ML classifiers can serve as lightweight, task-relevant quality metrics in 3D organ shape generation, supporting more transparent and clinically aligned evaluation protocols in medical shape modeling.

Problem

• Evaluating quality of generated liver shapes without ground truth
• Replacing labor-intensive expert review with ML-based quality assessment
• Providing interpretable feedback for 3D organ shape generation

Innovation

• Uses ML for liver shape quality assessment
• Extracts geometric features from point clouds
• Trains classifiers as proxy discriminators