DeSeg: auto detector-based segmentation for brain metastases

Hui Yu, Zhongzhou Zhang, Wenjun Xia, Yan Liu, Lunxin Liu, Wuman Luo, Jiliu Zhou, Yi Zhang

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)

Abstract

Delineation of brain metastases (BMs) is a paramount step in stereotactic radiosurgery treatment. Clinical practice has specific expectation on BM auto-delineation that the method is supposed to avoid missing of small lesions and yield accurate contours for large lesions. In this study, we propose a novel coarse-to-fine framework, named detector-based segmentation (DeSeg), to incorporate object-level detection into pixel-wise segmentation so as to meet the clinical demand. DeSeg consists of three components: a center-point-guided single-shot detector to localize the potential lesion regions, a multi-head U-Net segmentation model to refine contours, and a data cascade unit to connect both tasks smoothly. Performance on tiny lesions is measured by the object-based sensitivity and positive predictive value (PPV), while that on large lesions is quantified by dice similarity coefficient (DSC), average symmetric surface distance (ASSD) and 95% Hausdorff distance (HD95). Besides, computational complexity is also considered to study the potential of method in real-time processing. This study retrospectively collected 240 BM patients with Gadolinium injected contrast-enhanced T1-weighted magnetic resonance imaging (T1c-MRI), which were randomly split into training, validating and testing datasets (192, 24 and 24 scans, respectively). The lesions in the testing dataset were further divided into two groups based on the volume size (small S: ≤1.5 cc, N = 88; large L: > 1.5 cc, N = 15). On average, DeSeg yielded a sensitivity of 0.91 and a PPV of 0.77 on S group, and a DSC of 0.86, an ASSD 0f 0.76 mm and a HD95 of 2.31 mm on L group. The results indicated that DeSeg achieved leading sensitivity and PPV for tiny lesions as well as segmentation metrics for large ones. After our clinical validation, DeSeg showed competitive segmentation performance while kept faster processing speed comparing with existing 3D models.

Original languageEnglish
Article number025002
JournalPhysics in Medicine and Biology
Volume68
Issue number2
DOIs
Publication statusPublished - 21 Jan 2023

Keywords

  • brain metastasis auto-delineation
  • coarse-to-fine framework
  • fine segmentation
  • small object detection
  • stereotactic radiosurgery

Fingerprint

Dive into the research topics of 'DeSeg: auto detector-based segmentation for brain metastases'. Together they form a unique fingerprint.

Cite this