Independent component analysis for unmixing multi-wavelength photoacoustic images

Lu An; Ben Cox

doi:10.1117/12.2208137

18 March 2016 Independent component analysis for unmixing multi-wavelength photoacoustic images

Lu An, Ben Cox

Proceedings Volume 9708, Photons Plus Ultrasound: Imaging and Sensing 2016; 970851 (2016) https://doi.org/10.1117/12.2208137
Event: SPIE BiOS, 2016, San Francisco, California, United States

Abstract

Independent component analysis (ICA) is a blind source unmixing method that may be used under certain circumstances to decompose multi-wavelength photoacoustic (PA) images into separate components representing individual chromophores. It has the advantages of being fast, easy to implement and computationally inexpensive. This study uses simulated multi-wavelength PA images to investigate the conditions required for ICA to be an accurate unmixing method and compares its performance to linear inversion. An approximate fluence adjustment based on spatially homogeneous optical properties equal to that of the background region was applied to the PA images before unmixing with ICA or LI. ICA is shown to provide accurate separation of the chromophores in cases where the absorption coefficients are lower than certain thresholds, some of which are comparable to physiologically relevant values. However, the results also show that the performance of ICA abruptly deteriorates when the absorption is increased beyond these thresholds. In addition, the accuracy of ICA decreases in the presence of spatially inhomogeneous absorption in the background.

Citation Download Citation

Lu An and Ben Cox "Independent component analysis for unmixing multi-wavelength photoacoustic images", Proc. SPIE 9708, Photons Plus Ultrasound: Imaging and Sensing 2016, 970851 (18 March 2016); https://doi.org/10.1117/12.2208137

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available