Key Findings
The DS-CAT framework demonstrated high accuracy in glucose monitoring, explaining 99% of the variation in true glucose concentration. The model achieved a root mean square error of 1.09 g/L and a mean absolute error of 0.89 g/L, figures that the researchers state are consistent and sufficient for effective feed control. These results remained stable throughout five-fold validation.
Technical Details
The system integrates two distinct spectroscopic modalities to overcome individual limitations: Raman spectroscopy provides specific chemical fingerprints, while near-infrared (NIR) spectroscopy offers broad spectral coverage. The development team, including Xiwei Tian of East China University of Science and Technology and Yinlan Ruan of Guilin University of Electronic Technology, encoded these modalities separately before merging them via the DS-CAT architecture.
The model was tested using 142 preprocessed, synchronized Raman-NIR records. Sensitivity analysis identified the 1122−1335 cm-1 interval, associated with carbohydrate vibrational modes, as the area of strongest prediction sensitivity. Furthermore, the DS-CAT framework outperformed four industry-standard modeling approaches under the same evaluation protocol:
- Partial least squares regression
- Gradient boosting regression trees
- Convolutional neural networks
- Temporal convolutional networks
Background & Context
Accurate glucose monitoring has long been a bottleneck in fed-batch fermentation. This is particularly critical as biopharmaceutical manufacturers move toward hybrid and continuous bioprocessing for advanced therapeutics. Existing commercial off-line sensors typically cannot provide the high-frequency, low-latency measurements required for real-time processing controls.
Significance & Outlook
The ability of the DS-CAT approach to reduce noise and drift while enhancing sensitivity allows for more reliable insights into critical parameters in closed-loop mammalian and microbial fermentation. By providing online-compatible soft sensing, this technology may serve as a key element in the rational transition from traditional fed-batch fermentation to continuous manufacturing.
Source: https://www.genengnews.com/topics/bioprocessing/ramannir-spectra-improves-glucose-soft-sensing/
Get our weekly technology intelligence — free
Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.
Subscribe Free — Weekly Tech Intelligence
By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.
- Your email and selected fields are used only to deliver the newsletter.
- We never share your information with third parties.
- You can unsubscribe anytime via the link in each email.
See our Privacy Policy for details.
Takes about a minute · Unsubscribe anytime
