Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Constrained regularization: hybrid method for multivariate calibration.

Wei-Chuan Shih1, Kate L Bechtel, Michael S Feld

  • 1G. R. Harrison Spectroscopy Laboratory, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Room 6-014, Cambridge, Massachusetts 02139, USA. wshih@mit.edu

Analytical Chemistry
|December 30, 2006
PubMed
Summary

Constrained regularization (CR) offers a superior multivariate calibration method by balancing model complexity and noise using spectral constraints. This approach is more robust than existing techniques, especially for complex samples like biological tissues.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Isolation and Purification-Free Digital Single-Small Extracellular Vesicle Biosensing with Scalable Plasmonic Arrays.

bioRxiv : the preprint server for biology·2026
Same author

Microfluidic nano-plasmonic imaging platform for purification- and label-free single small extracellular vesicle characterization.

Npj biosensing·2025
Same author

Microfluidic Nano-Plasmonic Imaging Platform for Purification- and Label-Free Single Small Extracellular Vesicle Counting.

bioRxiv : the preprint server for biology·2025
Same author

Author Correction: Plasmonic nano-aperture label-free imaging of single small extracellular vesicles for cancer detection.

Communications medicine·2024
Same author

Plasmonic nano-aperture label-free imaging of single small extracellular vesicles for cancer detection.

Communications medicine·2024
Same author

Functional Plasmonic Microscope: Characterizing the Metabolic Activity of Single Cells via Sub-nm Membrane Fluctuations.

Analytical chemistry·2024

Area of Science:

  • Spectroscopy
  • Chemometrics
  • Data Analysis

Background:

  • Multivariate calibration is essential for quantitative analysis in spectroscopy.
  • Existing methods like Partial Least-Squares (PLS) can be susceptible to noise and spurious correlations.
  • Rigidly incorporating prior information can limit model adaptability.

Purpose of the Study:

  • To introduce and validate a novel hybrid multivariate calibration method: Constrained Regularization (CR).
  • To demonstrate the advantages of CR over existing methods, particularly in challenging sample matrices.
  • To highlight the flexibility of CR in handling variations in spectral data.

Main Methods:

  • Developed a hybrid multivariate calibration method, Constrained Regularization (CR).

Related Experiment Videos

  • Treated multivariate calibration as an inverse problem incorporating spectral constraints.
  • Validated CR using numerical simulations and experimental Raman spectra.

    • Main Results:

    • CR achieves an optimal balance between model complexity and noise rejection.
    • CR outperforms methods lacking prior information (e.g., PLS) and is less prone to spurious correlations.
    • CR demonstrates superior robustness compared to rigidly constrained methods (e.g., Hybrid Linear Analysis) when analyte spectra vary.

    Conclusions:

    • Constrained Regularization (CR) provides a flexible and robust approach to multivariate calibration.
    • CR is particularly advantageous for analyzing complex or turbid samples where spectral variations are common.
    • The method's ability to incorporate prior information adaptively enhances analytical accuracy and reliability.