Adaptive Multicore Dual-Path Fusion Multimodel Extraction of Heterogeneous Features for FAIMS Spectral Analysis

IF 1.8 3区 化学 Q4 BIOCHEMICAL RESEARCH METHODS
Ruilong Zhang, Xiaoxia Du, Wenxiang Xiao, Hua Li
{"title":"Adaptive Multicore Dual-Path Fusion Multimodel Extraction of Heterogeneous Features for FAIMS Spectral Analysis","authors":"Ruilong Zhang,&nbsp;Xiaoxia Du,&nbsp;Wenxiang Xiao,&nbsp;Hua Li","doi":"10.1002/rcm.9967","DOIUrl":null,"url":null,"abstract":"<div>\n \n <p>With the increasing application scenarios and detection needs of high-field asymmetric waveform ion mobility spectrometry (FAIMS) analysis, deep learning–assisted spectral analysis has become an important method to improve the analytical effect and work efficiency. However, a single model has limitations in generalizing to different types of tasks, and a model trained from one batch of spectral data is difficult to achieve good results on another task with large differences. To address this problem, this study proposes an adaptive multicore dual-path fusion multimodel extraction of heterogeneous features for FAIMS spectral analysis model in conjunction with FAIMS small-sample data analysis scenarios. Multinetwork complementarity is achieved through multimodel feature extraction, adaptive feature fusion module adjusts feature size and dimension fusion to heterogeneous features, and multicore dual-path fusion can capture and integrate information at all scales and levels. The model's performance improves dramatically when performing complex mixture multiclassification tasks: accuracy, precision, recall, f1-score, and micro-AUC reach 98.11%, 98.66%, 98.33%, 98.30%, and 98.98%. The metrics for the generalization test using the untrained xylene isomer data were 96.42%, 96.66%, 96.96%, 96.65%, and 97.60%. The model not only exhibits excellent analytical results on preexisting data but also demonstrates good generalization ability on untrained data.</p>\n </div>","PeriodicalId":225,"journal":{"name":"Rapid Communications in Mass Spectrometry","volume":"39 5","pages":""},"PeriodicalIF":1.8000,"publicationDate":"2024-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Rapid Communications in Mass Spectrometry","FirstCategoryId":"92","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/rcm.9967","RegionNum":3,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}
引用次数: 0

Abstract

With the increasing application scenarios and detection needs of high-field asymmetric waveform ion mobility spectrometry (FAIMS) analysis, deep learning–assisted spectral analysis has become an important method to improve the analytical effect and work efficiency. However, a single model has limitations in generalizing to different types of tasks, and a model trained from one batch of spectral data is difficult to achieve good results on another task with large differences. To address this problem, this study proposes an adaptive multicore dual-path fusion multimodel extraction of heterogeneous features for FAIMS spectral analysis model in conjunction with FAIMS small-sample data analysis scenarios. Multinetwork complementarity is achieved through multimodel feature extraction, adaptive feature fusion module adjusts feature size and dimension fusion to heterogeneous features, and multicore dual-path fusion can capture and integrate information at all scales and levels. The model's performance improves dramatically when performing complex mixture multiclassification tasks: accuracy, precision, recall, f1-score, and micro-AUC reach 98.11%, 98.66%, 98.33%, 98.30%, and 98.98%. The metrics for the generalization test using the untrained xylene isomer data were 96.42%, 96.66%, 96.96%, 96.65%, and 97.60%. The model not only exhibits excellent analytical results on preexisting data but also demonstrates good generalization ability on untrained data.

求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
4.10
自引率
5.00%
发文量
219
审稿时长
2.6 months
期刊介绍: Rapid Communications in Mass Spectrometry is a journal whose aim is the rapid publication of original research results and ideas on all aspects of the science of gas-phase ions; it covers all the associated scientific disciplines. There is no formal limit on paper length ("rapid" is not synonymous with "brief"), but papers should be of a length that is commensurate with the importance and complexity of the results being reported. Contributions may be theoretical or practical in nature; they may deal with methods, techniques and applications, or with the interpretation of results; they may cover any area in science that depends directly on measurements made upon gaseous ions or that is associated with such measurements.
×
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