Image quality improvement methods for optical coherence tomography
Ophthalmological images generated by coherent imaging modalities have multiple types of noise, including random noise caused by the imaging system and speckle noise caused by turbid objects such as living tissues. These noises can occur at different levels in different locations. A noise-reduction method and system of the present disclosure thus relates to applying different filters for different types of noise and/or different locations of images, sequentially or in parallel and combined, to produce a final noise-reduced image.
1 . An image processing method comprising:
inputting an input image to a machine learning system; and
generating a first noise-reduced image based on an output of the machine learning system,
wherein the machine learning system is trained to suppress from the input image either a first type of noise or a second type of noise, but not both of the first and second types of noise,
wherein the machine learning system is trained with images taken from the same or substantially the same location as the input image to suppress the first type of noise, or is trained with images taken from locations that are adjacent to or nearby a location of the input image to suppress the second type of noise,
wherein training images for the machine learning system comprise at least one image taken from a different subject than the input image,
wherein the input image is obtained by an optical coherence tomography imaging system configured to operate at least at a 400 kHz A-line rate,
wherein the first type of noise is random noise or noise caused by an imaging system that captured the input image,
wherein the second type of noise is speckle noise.
2 . The image processing method of claim 1 , further comprising:
combining the first noise-reduced image with a second noise-reduced image, thereby generating a final noise-reduced image,
wherein the first type of noise and the second type of noise are suppressed in the final noise-reduced image.
3 . The method of claim 2 , wherein the first noise-reduced image and the second noise-reduced images are combined by weighted averaging, the first noise-reduced image being weighted according to a level of the first type of noise in the input image and the second noise-reduced image being weighted according to a level of the second type of noise in the input image.
4 . The method of claim 2 , further comprising:
displaying the final noise-reduced image in real-time with a capturing of the input image.
5 . The method of claim 1 , wherein an intensity of at least one pixel of the first noise-reduced image is set by the machine learning system to correspond to a maximum intensity of a probability distribution of intensities of pixels at a corresponding location of an object in the image.
6 . The method of claim 1 , wherein the input image is an optical coherence tomography (OCT) or OCT-angiography B-scan image.
7 . The method of claim 1 , wherein the input image is an en face optical coherence tomography (OCT) or OCT-angiography image.
8 . The method of claim 1 , wherein the input image is an image of a retina.
9 . The method of claim 1 , further comprising:
segmenting the input image based on the output of the machine learning system.
10 . The method of claim 1 , further comprising:
displaying the first noise-reduced image in real-time with a capturing of the input image.
11 . An image processing method comprising:
filtering a first input image;
filtering a second input image; and
combining the filtered first input image and the filtered second input image, thereby generating a final noise-reduced volume,
wherein the first input image and the second input image are different 2D images of a 3D volume,
wherein the first input image and the second input image are from different planes of, and/or are en face images of different reference layers of the 3D volume,
wherein the first input image and second input image are obtained by an optical coherence tomography imaging system configured to operate at least at a 400 kHz A-line rate,
wherein filtering the first input image comprises:
applying a first filter to the first input image, thereby generating a first noise-reduced image;
applying a second filter to the first input image, thereby generating a second noise-reduced image; and
combining the first noise-reduced image and the second noise-reduced image, thereby generating a final noise-reduced image, and
wherein the first filter is configured to suppress a first type of noise from the first input image and the second filter is configured to suppress a second type of noise from the first input image, the first and second types of noise being different.