Automated mitochondrial oxygen consumption (mitoVO2) analysis via a bi-directional long short-term memory neural network.

IF 2 3区 医学 Q2 ANESTHESIOLOGY
C J de Wijs, J R Behr, L W J M Streng, M E van der Graaf, F A Harms, E G Mik
{"title":"Automated mitochondrial oxygen consumption (mitoVO<sub>2</sub>) analysis via a bi-directional long short-term memory neural network.","authors":"C J de Wijs, J R Behr, L W J M Streng, M E van der Graaf, F A Harms, E G Mik","doi":"10.1007/s10877-025-01291-1","DOIUrl":null,"url":null,"abstract":"<p><p>Monitoring in vivo mitochondrial oxygen tension (mitoPO<sub>2</sub>) enables the measurement of mitochondrial oxygen consumption (mitoVO<sub>2</sub>), providing deeper insights into the skin's mitochondrial environment. However, current mitoVO<sub>2</sub> analysis often relies on manual identification of start and end points, which introduces substantial inter-user variability. Addressing this limitation is crucial for broader adoption, comparability, and reproducibility across research groups. Therefore, the aim of this study was to develop a neural network-based software that automatically analyzes mitoVO<sub>2</sub>. A Bi-directional Long Short-Term Memory neural network was trained on 125 mitoPO<sub>2</sub> measurement sequences and optimized through Bayesian optimization. It identifies start points and measurement periods, then applies a modified Michaelis-Menten fit to calculate mitoVO<sub>2</sub>. This framework, embedded in automated software, was validated against the consensus of 3 raters. Bayesian optimization yielded an overall network performance of 94.2% on the test set. The neural network identified 91% of mitoVO<sub>2</sub> start points within a ± 5-sample range of the manual consensus. Mean mitoVO<sub>2</sub> values for the consensus and software were 6.56 and 6.63 mmHg s<sup>- 1</sup>, respectively, corresponding to a bias of -0.057 mmHg s<sup>- 1</sup>. Multiple runs of the network on the same dataset produced identical results, confirming consistency and eliminating inter-user variability. The developed neural network-based software automatically and consistently analyzes mitoVO<sub>2</sub> measurements, substantially reducing reliance on subjective judgments. By enabling a standardized approach to mitoVO<sub>2</sub> analysis, this tool improves data comparability and reproducibility across research settings. Future work will focus on further refining precision and extending functionality through multi-center collaborations.</p>","PeriodicalId":15513,"journal":{"name":"Journal of Clinical Monitoring and Computing","volume":" ","pages":""},"PeriodicalIF":2.0000,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Clinical Monitoring and Computing","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s10877-025-01291-1","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ANESTHESIOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Monitoring in vivo mitochondrial oxygen tension (mitoPO2) enables the measurement of mitochondrial oxygen consumption (mitoVO2), providing deeper insights into the skin's mitochondrial environment. However, current mitoVO2 analysis often relies on manual identification of start and end points, which introduces substantial inter-user variability. Addressing this limitation is crucial for broader adoption, comparability, and reproducibility across research groups. Therefore, the aim of this study was to develop a neural network-based software that automatically analyzes mitoVO2. A Bi-directional Long Short-Term Memory neural network was trained on 125 mitoPO2 measurement sequences and optimized through Bayesian optimization. It identifies start points and measurement periods, then applies a modified Michaelis-Menten fit to calculate mitoVO2. This framework, embedded in automated software, was validated against the consensus of 3 raters. Bayesian optimization yielded an overall network performance of 94.2% on the test set. The neural network identified 91% of mitoVO2 start points within a ± 5-sample range of the manual consensus. Mean mitoVO2 values for the consensus and software were 6.56 and 6.63 mmHg s- 1, respectively, corresponding to a bias of -0.057 mmHg s- 1. Multiple runs of the network on the same dataset produced identical results, confirming consistency and eliminating inter-user variability. The developed neural network-based software automatically and consistently analyzes mitoVO2 measurements, substantially reducing reliance on subjective judgments. By enabling a standardized approach to mitoVO2 analysis, this tool improves data comparability and reproducibility across research settings. Future work will focus on further refining precision and extending functionality through multi-center collaborations.

通过双向长短期记忆神经网络自动分析线粒体耗氧量(mitoVO2)。
监测体内线粒体氧张力(mitoPO2)可以测量线粒体耗氧量(mitoVO2),从而更深入地了解皮肤的线粒体环境。然而,当前的mitoVO2分析通常依赖于手动识别起点和终点,这引入了大量的用户间可变性。解决这一限制对于更广泛的采用、可比性和跨研究小组的可重复性至关重要。因此,本研究的目的是开发一种基于神经网络的软件,自动分析mitoVO2。对125个mitoPO2测量序列进行双向长短期记忆神经网络训练,并进行贝叶斯优化。它识别起始点和测量周期,然后应用改进的Michaelis-Menten拟合来计算mitoVO2。该框架嵌入到自动化软件中,并与3个评分者的一致意见进行了验证。在测试集上,贝叶斯优化产生了94.2%的整体网络性能。神经网络在人工共识的±5个样本范围内识别出91%的mitoVO2起点。共识和软件的平均mitoVO2值分别为6.56和6.63 mmHg s- 1,对应于-0.057 mmHg s- 1的偏差。在同一数据集上多次运行网络产生了相同的结果,确认了一致性并消除了用户间的可变性。开发的基于神经网络的软件自动且一致地分析mitoVO2测量值,大大减少了对主观判断的依赖。通过采用标准化的方法进行mitoVO2分析,该工具提高了研究设置中的数据可比性和可重复性。未来的工作将集中在通过多中心协作进一步提高精度和扩展功能上。
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 求助全文
来源期刊
CiteScore
4.30
自引率
13.60%
发文量
144
审稿时长
6-12 weeks
期刊介绍: The Journal of Clinical Monitoring and Computing is a clinical journal publishing papers related to technology in the fields of anaesthesia, intensive care medicine, emergency medicine, and peri-operative medicine. The journal has links with numerous specialist societies, including editorial board representatives from the European Society for Computing and Technology in Anaesthesia and Intensive Care (ESCTAIC), the Society for Technology in Anesthesia (STA), the Society for Complex Acute Illness (SCAI) and the NAVAt (NAVigating towards your Anaestheisa Targets) group. The journal publishes original papers, narrative and systematic reviews, technological notes, letters to the editor, editorial or commentary papers, and policy statements or guidelines from national or international societies. The journal encourages debate on published papers and technology, including letters commenting on previous publications or technological concerns. The journal occasionally publishes special issues with technological or clinical themes, or reports and abstracts from scientificmeetings. Special issues proposals should be sent to the Editor-in-Chief. Specific details of types of papers, and the clinical and technological content of papers considered within scope can be found in instructions for authors.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信