A three-dimensional automatic positioning system for epileptogenic focus based on deep learning

Abstract

The invention discloses a three-dimensional automatic positioning system for epileptogenic focus based on deep learning. The system includes: a PET image acquisition and marking module; a registration module for PET images and standard symmetrical brain templates; a PET image data preprocessing module to generate left and right Mirrored pair of brain image blocks; SiameseNet training module of Siamese network, including two deep residual convolutional neural networks with shared weight parameters, the output layer is connected to a multi-layer perceptron and a softmax layer, using images with epileptogenic foci and normal images The training set trains the network to obtain a network model; the classification module and the epileptogenic focus positioning module use the trained network model to generate a probability heat map for the newly input PET image, and first judge the image as normal or carry epileptogenicity through a classifier. focus samples, and then predict the location of the epileptogenic focus area. The system automatically locates the epileptogenic focus of PET images by introducing the mirror image pair of image blocks and the twin network SiameseNet, which can effectively improve the accuracy and efficiency of epileptogenic focus positioning and has high robustness.

Description

technical field [0001] The invention relates to the technical field of medical imaging engineering, in particular to a three-dimensional automatic positioning system for epileptogenic focus based on deep learning. Background technique [0002] With the development of medical imaging technology and artificial intelligence technology, automatic and semi-automatic computer-aided diagnosis systems are widely used in precise diagnosis and treatment to improve diagnostic accuracy and prognosis. Currently, detection systems for epilepsy include positron emission tomography (PET), magnetic resonance imaging (MRI), single photon emission computed tomography (SPECT) and electroencephalography (EEG). and prognosis with higher sensitivity. For the determination of the type of epilepsy and the surgical treatment of refractory epilepsy, it is necessary to use a diagnostic system to accurately locate the location of the epileptogenic focus. However, conventional routine clinical diagnosi...

AI Technical Summary

Problems solved by technology

Because the area is often much larger than the lesion area, the method will ignore subtle changes, resulting in a decrease in its detection sensitivity

Voxelwise statistical methods typically use Statistical Parametric Mapping (SPM) software to compare data from individual cases and control groups, however, voxelwise statistical methods are highly sensitive to registration errors and are prone to false positives in misaligned regions

2. Most of the existing algorithms are only suitable for 2D natural image processing, and since PET imaging is a 3D structure composed of parallel scanning image frames, 2D localization algorithms will ignore important inter-frame information

3. Due to the small amount of medical image data, the lack of high-quality labeled data and training samples, and the large difference in the number of positive and negative samples resulting in sample imbalance, the trained model may be overfitting or the model generalization ability is not high.

Images