A high-speed microscopy system based on deep learning to detect yeast-like fungi cells in blood.

IF 1.9 4区医学 Q3 BIOCHEMICAL RESEARCH METHODS Bioanalysis Pub Date : 2024-03-01 Epub Date: 2024-02-09 DOI:10.4155/bio-2023-0193

Ruiqi Liu, Xiaojie Li, Yingyi Liu, Lijun Du, Yingzhu Zhu, Lichuan Wu, Bo Hu

引用次数: 0

Abstract

Background: Blood-invasive fungal infections can cause the death of patients, while diagnosis of fungal infections is challenging. Methods: A high-speed microscopy detection system was constructed that included a microfluidic system, a microscope connected to a high-speed camera and a deep learning analysis section. Results: For training data, the sensitivity and specificity of the convolutional neural network model were 93.5% (92.7-94.2%) and 99.5% (99.1-99.5%), respectively. For validating data, the sensitivity and specificity were 81.3% (80.0-82.5%) and 99.4% (99.2-99.6%), respectively. Cryptococcal cells were found in 22.07% of blood samples. Conclusion: This high-speed microscopy system can analyze fungal pathogens in blood samples rapidly with high sensitivity and specificity and can help dramatically accelerate the diagnosis of fungal infectious diseases.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

基于深度学习的高速显微镜系统，用于检测血液中的酵母样真菌细胞。

背景：血源性真菌感染可导致患者死亡，而真菌感染的诊断却很困难。方法：构建一个高速显微检测系统：构建了一个高速显微镜检测系统，其中包括一个微流控系统、一个连接高速相机的显微镜和一个深度学习分析部分。结果对于训练数据，卷积神经网络模型的灵敏度和特异性分别为 93.5%（92.7%-94.2%）和 99.5%（99.1%-99.5%）。验证数据的灵敏度和特异性分别为 81.3%（80.0-82.5%）和 99.4%（99.2-99.6%）。在 22.07% 的血液样本中发现了隐球菌细胞。结论该高速显微系统可快速分析血液样本中的真菌病原体，灵敏度和特异性都很高，有助于大大加快真菌感染性疾病的诊断速度。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Bioanalysis BIOCHEMICAL RESEARCH METHODS-CHEMISTRY, ANALYTICAL

CiteScore

3.30

自引率

16.70%

发文量

审稿时长

2 months

期刊介绍： Reliable data obtained from selective, sensitive and reproducible analysis of xenobiotics and biotics in biological samples is a fundamental and crucial part of every successful drug development program. The same principles can also apply to many other areas of research such as forensic science, toxicology and sports doping testing. The bioanalytical field incorporates sophisticated techniques linking sample preparation and advanced separations with MS and NMR detection systems, automation and robotics. Standards set by regulatory bodies regarding method development and validation increasingly define the boundaries between speed and quality. Bioanalysis is a progressive discipline for which the future holds many exciting opportunities to further reduce sample volumes, analysis cost and environmental impact, as well as to improve sensitivity, specificity, accuracy, efficiency, assay throughput, data quality, data handling and processing. The journal Bioanalysis focuses on the techniques and methods used for the detection or quantitative study of analytes in human or animal biological samples. Bioanalysis encourages the submission of articles describing forward-looking applications, including biosensors, microfluidics, miniaturized analytical devices, and new hyphenated and multi-dimensional techniques. Bioanalysis delivers essential information in concise, at-a-glance article formats. Key advances in the field are reported and analyzed by international experts, providing an authoritative but accessible forum for the modern bioanalyst.