Scholarlink

<p>Bloodstream infections (BSIs) of high morbidity and mortality are across all age groups, and urgent for accurate intervention. Gram stain interpretation of positive blood cultures (PBCs) is crucial for early diagnosing BSIs, yet this manual process is labor-intensive, time-consuming, and highly operator-dependent. Artificial intelligence&#xa0;(AI)-assisted microscopic interpretation of stained smears presents beneficial to microbiology diagnostics. Addressing the auto-identification of blood-culture Gram stains, this study introduces a dataset of Gram-stain smears collected in clinical practice. The dataset includes 505 microscopic images, covering up to 57 species associated with BSIs, with a total of 7528 annotations. These annotations categorized by staining&#xa0;characteristics and morphological&#xa0;features into cocci, bacilli, and fungi. We trained and validated an object detection model based on the YOLOv10 architecture on this dataset to automatically localize and classify these morphological categories in microscopic images. The publicly released&#xa0;dataset will help developments that utilize artificial intelligence to auto-interpretate the Gram stains from PBCs for routine clinical application.</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

An annotated dataset of Gram stains from positive blood cultures

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月23日

查看全文
  • Qiaolian Yi,
  • Xiaoyan Gou,
  • Renyuan Zhu,
  • Xiuli Xie,
  • Mengting Hu,
  • Xing Wang,
  • Tai’e Wang,
  • Kaiwen Xu,
  • Ying-Chun Xu

摘要

Bloodstream infections (BSIs) of high morbidity and mortality are across all age groups, and urgent for accurate intervention. Gram stain interpretation of positive blood cultures (PBCs) is crucial for early diagnosing BSIs, yet this manual process is labor-intensive, time-consuming, and highly operator-dependent. Artificial intelligence (AI)-assisted microscopic interpretation of stained smears presents beneficial to microbiology diagnostics. Addressing the auto-identification of blood-culture Gram stains, this study introduces a dataset of Gram-stain smears collected in clinical practice. The dataset includes 505 microscopic images, covering up to 57 species associated with BSIs, with a total of 7528 annotations. These annotations categorized by staining characteristics and morphological features into cocci, bacilli, and fungi. We trained and validated an object detection model based on the YOLOv10 architecture on this dataset to automatically localize and classify these morphological categories in microscopic images. The publicly released dataset will help developments that utilize artificial intelligence to auto-interpretate the Gram stains from PBCs for routine clinical application.