Apr 16, 2015
Professor Louis Bouchard
UCLA researchers develop novel image filtering algorithm. 
 
The algorithm, developed by Professor Louis Bouchard and UCLA Biomedical Engineering graduate student and Bouchard lab member Khalid Youssef, removes image noise while preserving image features with unprecedented fidelity.
 
Invention: Image Filtering Algorithm for Enhanced Noise Removal and Feature Preservation
 
Background
Traditional image processing techniques use filtering algorithms based either on spatial smoothing (averaging) of pixel intensities or on distinguishing the true image signal from noise in a mathematically transformed domain (e.g. sorting image data by frequency content rather than by location).  However, both of these approaches rely on imperfect assumptions about the statistical distribution of noise. Consequently, these methods often blur the image by suppressing certain spatial variations in intensity or by mistakenly discarding certain shapes as noise. The end result of these filtering techniques is noise reduction at the expense of diminished image quality. 
 
 
As shown in the photos above, single copies of a noisy image (left) can be acquired multiple times and averaged to yield a signal-averaged image (center-left). Signal averaging of the noisy images results in some improvements of the signal-to-noise ratio (SNR), but not enough to provide a faithful representation of the noiseless image (right). MC-MLP denoising (center-right) operates on the seven copies of the noisy images with a suitably designed nonlinear filter to provide high fidelity removal of the noise components while preserving the relevant anatomical features and image contrast. (Photos: Louis Bouchard, UCLA Department of Chemistry and Biochemistry)
 
Innovation
Prof. Louis Bouchard and Khalid Youssef, UCLA Biomedical Engineering graduate student and Bouchard lab member, have developed an advanced image filtering algorithm that effectively removes image noise while preserving image features with unprecedented fidelity. This efficient de-noising algorithm employs a nonlinear filter based on multilayer perceptrons (MLPs) to groups of similar-looking image patches across multiple copies of the original image. This filtering technique outperforms current state-of-the-art noise removal algorithms including those based on collaborative filtering and total variation.
 
Applications
•Medical image filtering, including the following modalities:
•Magnetic resonance imaging (MRI), especially noisy images such as 23Na (sodium) MRI and 31P (phosphorous) MRI as well as time-course analysis such as functional MRI (fMRI).
•Computed tomography (CT)
•Positron emission tomography (PET)
•X-ray imaging
•Medical ultrasound imaging
•Optical (e.g. endoscopic & laparoscopic) imaging
•Electron microscopy
•Image filtering for digital photography (commercial and consumer), such as photography in dark environments or short exposure times
•Image restoration for video editing
•Image filtering for machine vision & artificial intelligence
•Automated face & object recognition algorithms
•Security and defense-related camera systems
•Self-driving vehicles
•Efficient image compression & de-compression
 
Advantages
•Superior noise removal & feature preservation vs. current state-of-the-art image filtering methods
•Applicable to any imaging modality: can be used to reduce scan time or dose to the patient (or sample)
•Does not require knowledge or modeling of noise distribution
•Works well for both low- and high-noise images
•Handles both additive and multiplicative noise
•Computationally efficient
•Accounts for spatial correlations in pixel intensity within the image
 
Details about the invention can be found in the related paper (in press):
Youssef K, Jarenwattananon NN, Bouchard LS, Feature-preserving noise removal, IEEE Trans. Med. Imag. (in press)
 
Invention information
Tech ID: 24924 / UC Case 2015-240-0
 
Patent Information
US Patent Application No. 62/092762 (Filed: Dec. 16, 2014).  Title: “Feature-Preserving Image Noise Removal for Arbitrary Noise Distribution”.  Inventors: Youssef K, Bouchard LS.