Intracranial hemorrhage detection model with optimized and enhanced window adjustment and construction method of intracranial hemorrhage detection model

Abstract

The invention relates to an intracranial hemorrhage detection model with optimized and enhanced window adjustment and a construction method of the intracranial hemorrhage detection model. The invention provides an intracranial hemorrhage detection model on one hand. The intracranial hemorrhage detection model comprises a window adjustment optimization enhancement module and a RetinaNet network. The window adjustment optimization enhancement module is constructed by a 1 * 1 convolution layer and a window activation function layer. The network comprises a basic feature extraction network, an FPNfeature pyramid and a classification and regression sub-network. On the other hand, the invention also provides a construction method of the intracranial hemorrhage detection model with optimized andenhanced window adjustment. The construction method comprises the following steps: step 1, preparing a craniocerebral CT examination data set and carrying out data preprocessing; step 2, constructingan intracranial hemorrhage detection model; step 3, training an intracranial hemorrhage detection model; and step 4, verifying the intracranial hemorrhage detection model. According to the invention,the contrast between a bleeding area and a normal tissue is enhanced through the window adjustment optimization module, and the accuracy of model detection is greatly improved by combining the feature extraction of ResNet and the setting of a network.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a window-tuning optimized and enhanced intracranial hemorrhage detection model and a construction method thereof. Background technique [0002] Intracranial Hemorrhage (ICH) is one of the symptoms of acute stroke, which may lead to disability or death if not diagnosed and treated in time. Intracranial hemorrhage can be divided into five subtypes based on anatomical location and underlying cause: intraparenchymal hemorrhage (IPH), intraventricular hemorrhage (IVH), epidural hemorrhage (EDH), subdural hemorrhage (SDH), and subarachnoid hemorrhage (SAH). [0003] In the current clinical diagnosis of intracranial hemorrhage, radiologists generally read brain CT scans manually to determine whether there is ICH and the type of hemorrhage in the scan and locate the hemorrhage area. However, this process depends largely on the clinical experience of radiologists; if it...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem of low detection accuracy of the traditional deep learning method, to provide a window optimization enhanced intracranial hemorrhage detection model and its construction method, which can quickly and accurately locate the bleeding area

Images