文献类型：期刊文献

英文题名：Harnessing near-infrared and Raman spectral sensing and artificial intelligence for real-time monitoring and precision control of bioprocess

作者：Xu, Feng[1,2,3];Su, Lihuan[1,2];Gao, Hao[1,2];Wang, Yuan[1,2];Ben, Rong[1,2];Hu, Kaihao[1,2];Mohsin, Ali[1,2];Li, Chao[1,2,3];Chu, Ju[1,2,3];Tian, Xiwei[1,2,3]

机构：[1]East China Univ Sci & Technol, Qingdao Innovat Inst, State Key Lab Bioreactor Engn, Shanghai 200237, Peoples R China;[2]East China Univ Sci & Technol, Natl Ctr Bioengn & Technol Shanghai, Shanghai 200237, Peoples R China;[3]East China Univ Sci & Technol, Shanghai Collaborat Innovat Ctr Biomfg Technol, Shanghai 200237, Peoples R China

年份：2025

卷号：421

外文期刊名：BIORESOURCE TECHNOLOGY

收录：;EI(收录号：20250617833757);WOS:【SCI-EXPANDED(收录号:WOS:001426519200001)】；

基金：This work was financially supported by the National Key Research and Development Program of China (2024YFA0917900) , the Taishan Scholars Program of Shandong Province (NO. tsqn202312316) , the Shanghai Pilot Program for Basic Research (22TQ1400100-14) , the Natural Science Foundation of Shanghai (23ZR1416500) , the Frontiers Science Center for Materiobiology and Dynamic Chemistry (JKVJ1231036) . Thanks for the financial support from the Arawana Charity Foundation. Thanks for the technical support from the engineers of Tian Jin RuiPu Analytical Instruments Co. Ltd., Mettler Toledo Technology (CHINA) CO., and Suzhou Womei Biology Co. Ltd.

语种：英文

外文关键词：Artificial intelligence; Near-infrared spectroscopy; Raman spectroscopy; Bioprocess optimization; Gentamicin C1a

摘要：Effective monitoring and control of bioprocesses are critical for industrial biomanufacturing. This study demonstrates the integration of near-infrared and Raman spectroscopy for real-time monitoring and precise control of gentamicin fermentation. The orthogonal method reduced redundant features and improved spectral model performance by 9.2-100.4 % in terms of the coefficient of determination (R-2). The combinatorial spectral model outperformed single-source models in external validation (R-2 > 0.99). An AI-based platform, combining dual-sensors data collection, ML-based prediction, and automated feeding control, was developed for fully automated fed-batch fermentation. This platform dynamically adjusted feeding rates, maintained low glucose concentrations (5 g/L) with accuracy and coefficient of variation below 2 %, and increased gentamicin C1a concentration (346.5 mg/L) by 33.0 % compared to traditional intermittent feeding. These findings underscore the transformative potential of combinatorial spectroscopy and machine learning for real-time bioprocess monitoring, offering a scalable solution for enhancing industrial fermentation efficiency and product titer.