Near-infrared spectrum tomography reconstruction method based on convolutional neural network
A convolutional neural network and near-infrared spectroscopy technology, applied in the field of medical image processing, can solve the problems of morbidity of tomography, limited number of measurements, and mixed noise.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0039] The present invention will be described below based on specific implementation examples and accompanying drawings.
[0040] First, a circular phantom with a diameter of 80mm is established through the toolbox nirfast of Matlab and the finite element meshing of the phantom is completed. The result of the finite element meshing is as follows figure 2 shown. In the experiment, there are 16 light sources and detectors, such as image 3 As shown, it is evenly placed on the surface of the phantom, and 240 boundary measurements can be measured for each phantom sample, and the imaging pixels are 2001 uniform finite element nodes on the circle. In order to train the network, the absorption coefficient values on 2001 nodes are mapped to a 98*98 matrix to form an absorption coefficient distribution image, so the measured value is used as the input of the convolutional neural network, and the absorption coefficient distribution image is used as the input of the convolutional neur...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


