The NIRS Analysis Package: Noise Reduction and Statistical Inference
Author Information
Author(s): Tomer Fekete, Denis Rubin, Joshua M. Carlson, Lilianne R. Mujica-Parodi
Primary Institution: State University of New York at Stony Brook
Hypothesis
How can NIRS-specific noise be effectively reduced to improve statistical inference in neuroimaging?
Conclusion
The NIRS Analysis Package significantly improves the detection power and reliability of NIRS data by effectively reducing noise.
Supporting Evidence
- The NIRS Analysis Package (NAP) was validated using both simulated and actual data.
- Systemic artifact cancellation resulted in an increase in signal to noise by a factor of 4.33.
- Motion artifact correction is imperative for valid statistical analysis.
Takeaway
This study created a toolbox to help scientists get better brain images by cleaning up the noise that can mess up the pictures.
Methodology
The study involved using a Matlab toolbox to clean NIRS data from motion and systemic artifacts and applying statistical models for analysis.
Potential Biases
There is a risk of false positives due to motion artifacts if not properly corrected.
Limitations
The NIRS sampling rate may not resolve the hemodynamic response function effectively, and the current version of the toolbox is limited to block or event-related designs.
Participant Demographics
12 adults (3 females), ages 23-39, all healthy and free of neurological or cardiovascular illness.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website