PET/CT automatic lung cancer diagnosis classification model training method based on pathological feature assistance

Abstract

The invention discloses a PET / CT automatic lung cancer diagnosis classification model training method based on pathological feature assistance, and belongs to the field of medical images. The method comprises the following steps: training a classification network of pathological images to preferentially obtain a group of better pathological classification network model parameters; and obtaining the feature information of the pathological image through the group of parameters to guide feature extraction of the PET / CT image classification network, so that the precision of the PET / CT image classification network is improved, popularization and application of early lung cancer diagnosis and classification based on PET / CT images are facilitated, and help is provided for diagnosis and subsequent follow-up visit of clinicians. According to the invention, before the subsequent invasive pathological examination is not carried out, a more accurate lung cancer diagnosis classification result close to a pathological diagnosis result can be achieved only through a noninvasive PET / CT image, so that the diagnosis efficiency of a clinician can be effectively improved, and the wound of a patient is reduced.

Description

technical field [0001] The invention relates to the field of medical imaging, in particular to a PET / CT automatic lung cancer diagnosis and classification model training method based on pathological features. Background technique [0002] With the continuous development of medical technology, more and more imaging methods have been applied. A number of studies have confirmed that PET / CT is of great value in the diagnosis of benign and malignant pulmonary nodules, the staging of lung cancer, and the evaluation of lung cancer after treatment. At present, the most widely used tracer in PET / CT scanning is 18F-FDG (18 fluorine-labeled glucose analogue). According to the abnormal proliferation of malignant tumor cells, it is necessary to increase glucose uptake and glycolysis to maintain cellular energy supply, making different Different types of tumors show different levels of glucose uptake on glucose metabolism imaging images. The tracer decays and then annihilates in the pat...

AI Technical Summary

Problems solved by technology

[0004] Although the existing PET / CT-based automatic lung cancer diagnosis and classification model can obtain better diagnostic classification accuracy, there is still a certain distance from clinical use. The existing pathology-based automatic lung cancer diagnosis model can obtain higher Diagnostic classification accuracy, but in early screening and rapid diagnosis, pathology requires invasive examination, and the result is slow, so it is difficult to use in early diagnosis

Images