Sayed Mattin, Saba-Sadiya Sari, Wichtlhuber Benedikt, Dietz Julia, Neitzel Matthias, Keller Leopold, Roig Gemma, Bucher Andreas M
Clinic for Radiology and Nuclear Medicine, University Hospital, Goethe University Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany.
Institut für Informatik, Goethe University Frankfurt, Robert-Mayer-Str 11, 60325 Frankfurt am Main, Germany.
Tomography. 2024 Dec 23;10(12):2128-2143. doi: 10.3390/tomography10120150.
Medical imagesegmentation is an essential step in both clinical and research applications, and automated segmentation models-such as TotalSegmentator-have become ubiquitous. However, robust methods for validating the accuracy of these models remain limited, and manual inspection is often necessary before the segmentation masks produced by these models can be used.
To address this gap, we have developed a novel validation framework for segmentation models, leveraging data augmentation to assess model consistency. We produced segmentation masks for both the original and augmented scans, and we calculated the alignment metrics between these segmentation masks.
Our results demonstrate strong correlation between the segmentation quality of the original scan and the average alignment between the masks of the original and augmented CT scans. These results were further validated by supporting metrics, including the coefficient of variance and the average symmetric surface distance, indicating that agreement with augmented-scan segmentation masks is a valid proxy for segmentation quality.
Overall, our framework offers a pipeline for evaluating segmentation performance without relying on manually labeled ground truth data, establishing a foundation for future advancements in automated medical image analysis.
医学图像分割在临床和研究应用中都是关键步骤，像TotalSegmentator这样的自动分割模型已无处不在。然而，用于验证这些模型准确性的可靠方法仍然有限，在使用这些模型生成的分割掩码之前，通常需要进行人工检查。
为弥补这一差距，我们开发了一种用于分割模型的新型验证框架，利用数据增强来评估模型一致性。我们为原始扫描和增强扫描都生成了分割掩码，并计算了这些分割掩码之间的对齐指标。
我们的结果表明，原始扫描的分割质量与原始CT扫描和增强CT扫描掩码之间的平均对齐度之间存在很强的相关性。这些结果通过包括方差系数和平均对称表面距离在内的支持指标得到进一步验证，表明与增强扫描分割掩码的一致性是分割质量的有效替代指标。
总体而言，我们的框架提供了一个无需依赖人工标记的地面真值数据即可评估分割性能的流程，为自动医学图像分析的未来发展奠定了基础。
