Suppr超能文献

基于γ相机图像的计算机辅助肝脏肿块估计

Computer-assisted liver-mass estimation from gamma-camera images.

作者信息

Eikman E A, Mack G A, Jain V K, Madden J A

出版信息

J Nucl Med. 1979 Feb;20(2):144-8.

PMID:372502
Abstract

We have devised a computer-assisted method for objective estimation of liver mass from the right lateral projection of radiocolloid images of the liver. Gamma-camera images were digitized, preprocessed, and stored in computer memory. The definition of liver for area measurement was adaptively determined by means of a Laplacian operator that measures change in radioactivity slope associated with the liver margin. Individual thresholds were calculated for each of 16 subregions. A liver-mass index was derived from a linear regression model correlating the area of the right lateral projection with liver weight at autopsy in 50 patients whose livers weighed between 0.8 to 3.0 Kg. The correlation coefficient found for this method was 0.83 using the equation: Liver Mass [kg] = Area [cm2]/275 [kg/cm2]--0.215 [kg]. Liver-mass estimates using an alternative computer-assisted method or representative manual methods adapted for gamma-camera images showed lower correlation with liver weight at autopsy.

摘要

我们设计了一种计算机辅助方法,用于从肝脏放射性胶体图像的右侧位投影客观估计肝脏质量。γ相机图像被数字化、预处理并存储在计算机内存中。通过拉普拉斯算子自适应确定用于面积测量的肝脏定义,该算子测量与肝脏边缘相关的放射性斜率变化。为16个亚区域分别计算个体阈值。通过线性回归模型得出肝脏质量指数,该模型将右侧位投影面积与50例肝脏重量在0.8至3.0千克之间的患者尸检时的肝脏重量相关联。使用公式“肝脏质量[千克]=面积[平方厘米]/275[千克/平方厘米] - 0.215[千克]”,此方法得到的相关系数为0.83。使用另一种计算机辅助方法或适用于γ相机图像的代表性手动方法进行的肝脏质量估计显示,与尸检时的肝脏重量相关性较低。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验