文献速递心脏核医学骨显像与常规 SPECT

卷积神经网络用于二维闪烁显像图像中甲状腺素运载蛋白淀粉样变性的检测

Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images.
PubMed 文献 · EJNMMI Res · 2022-05-07

研究纳入1334例接受[99mTc]标记羟亚甲基二膦酸盐(HMDP)骨闪烁显像并按Perugini分级(0–3级)视觉分级的患者,其中47例视觉分级≥2级被视为ATTR阳性。训练两种自定义卷积神经网络(CNN)区分四个Perugini心脏摄取等级,并与四种先进CNN模型比较,自定义模型在四分类中AUC≥0.85,检测阴性vs阳性ATTR(分级<2 vs ≥2)AUC>0.88,高等级心脏摄取vs其他(分级<3 vs 3)AUC=0.94。

要点

  • 自定义CNN在心脏摄取检测与分级中优于或等同于先进CNN模型。
  • 四分类Perugini分级AUC≥0.85,阴性vs阳性ATTR的AUC>0.88。
  • 高等级心脏摄取vs其他患者AUC=0.94,最大激活图显示模型聚焦心肌。

核心结论

  • 自动卷积神经网络可准确检测并分类骨闪烁显像中的不同心脏摄取等级。
  • CNN模型聚焦于临床相关的图像特征。
  • 使用CNN自动筛查骨闪烁显像图像可能改善ATTR的早期诊断。

研究设计与样本

  • 研究人群为1334例接受[99mTc]标记HMDP闪烁显像并按Perugini分级(0–3级)视觉分级的患者。
  • 47例患者视觉分级≥2级,被视为ATTR阳性。
  • 训练两种自定义CNN架构区分四个Perugini心脏摄取等级,并与四种先进CNN模型比较分类性能。

主要发现

  • 自定义CNN模型在心脏摄取的检测与分类中优于或等同于先进模型。
  • 两种模型在四分类Perugini分级中AUC均≥0.85。
  • 检测阴性vs阳性ATTR患者(分级<2 vs 分级≥2)AUC>0.88。
  • 检测高等级心脏摄取vs其他患者(分级<3 vs 分级3)AUC=0.94。
  • 最大激活图显示自动深度学习模型聚焦于检测心肌而非心外特征。

关键数值与推荐

  • 视觉分级≥2级被视为ATTR阳性。
  • 四分类Perugini分级AUC≥0.85。
  • 阴性vs阳性ATTR(分级<2 vs ≥2)AUC>0.88。
  • 高等级心脏摄取vs其他(分级<3 vs 3)AUC=0.94。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Transthyretin amyloidosis (ATTR) is a progressive disease which can be diagnosed non-invasively using bone avid [99mTc]-labeled radiotracers. Thus, ATTR is also an occasional incidental finding on bone scintigraphy. In this study, we trained convolutional neural networks (CNN) to automatically detect and classify ATTR from scintigraphy images. The study population consisted of 1334 patients who underwent [99mTc]-labeled hydroxymethylene diphosphonate (HMDP) scintigraphy and were visually graded using Perugini grades (grades 0-3). A total of 47 patients had visual grade ≥ 2 which was considered positive for ATTR. Two custom-made CNN architectures were trained to discriminate between the four Perugini grades of cardiac uptake. The classification performance was compared to four state-of-the-art CNN models. RESULTS: Our CNN models performed better than, or equally well as, the state-of-the-art models in detection and classification of cardiac uptake. Both models achieved area under the curve (AUC) ≥ 0.85 in the four-class Perugini grade classification. Accuracy was good in detection of negative vs. positive ATTR patients (grade < 2 vs grade ≥ 2, AUC > 0.88) and high-grade cardiac uptake vs. other patients (grade < 3 vs. grade 3, AUC = 0.94). Maximum activation maps demonstrated that the automated deep learning models were focused on detecting the myocardium and not extracardiac features. CONCLUSION: Automated convolutional neural networks can accurately detect and classify different grades of cardiac uptake on bone scintigraphy. The CNN models are focused on clinically relevant image features. Automated screening of bone scintigraphy images using CNN could improve the early diagnosis of ATTR.

来源:EJNMMI Res,查看原文。版权归原作者所有。

引用

  1. Halme HL, Ihalainen T, Suomalainen O, et al. Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images. EJNMMI Res. 2022. PMID: 35524861. doi:10.1186/s13550-022-00897-9

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。