Scholarlink

<p>Brain metastases (BM) are a severe complication of multiple primary malignancies and constitute the most frequent tumors of the central nervous system. Clinical imaging plays a crucial role in the management of BMs, serving as an essential tool for accurate diagnosis, effective treatment planning, and systematic follow-up care. However, progress in improving BM diagnosis and treatment is hindered by the limited availability of high-quality, annotated imaging datasets that reflect clinical variability in morphology, treatment planning, and patient characteristics. This study presents a comprehensive longitudinal dataset of 40 patients, including 744 MRI scans (baseline and follow-up), as well as 45 radiotherapy plans and 45 computed tomography scans. Detailed segmentations of 65 BMs are provided, defining three key tumor regions: enhancing tumor, edema, and necrotic core, offering a precise representation of tumor morphology for advanced quantitative analysis. This data-sharing initiative aims to support research efforts in automated BM detection, lesion segmentation, disease assessment, treatment planning, while contributing to the development of pertinent predictive and prognostic tools to enhance clinical decision-making.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A longitudinal MRI dataset of brain metastases with tumor segmentations, clinical & radiomic data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月19日

查看全文
  • Dimitra Flouri,
  • Christos-Panagiotis Papanikas,
  • Georgios C. Manikis,
  • Eleftherios Ioannou,
  • Georgia Karaoli,
  • Elisavet Papageorgiou,
  • Anastasia Constantinidou,
  • Loizos Siakallis,
  • Marilena Theodorou,
  • Vasileios Vavourakis

摘要

Brain metastases (BM) are a severe complication of multiple primary malignancies and constitute the most frequent tumors of the central nervous system. Clinical imaging plays a crucial role in the management of BMs, serving as an essential tool for accurate diagnosis, effective treatment planning, and systematic follow-up care. However, progress in improving BM diagnosis and treatment is hindered by the limited availability of high-quality, annotated imaging datasets that reflect clinical variability in morphology, treatment planning, and patient characteristics. This study presents a comprehensive longitudinal dataset of 40 patients, including 744 MRI scans (baseline and follow-up), as well as 45 radiotherapy plans and 45 computed tomography scans. Detailed segmentations of 65 BMs are provided, defining three key tumor regions: enhancing tumor, edema, and necrotic core, offering a precise representation of tumor morphology for advanced quantitative analysis. This data-sharing initiative aims to support research efforts in automated BM detection, lesion segmentation, disease assessment, treatment planning, while contributing to the development of pertinent predictive and prognostic tools to enhance clinical decision-making.