Research And Application Of Auxiliary Diagnosis And Treatment System For Cervical Vertebrae Based On Infrared Thermography

On the one hand,medical infrared thermal imaging technology has a high research and application value as a convenient,rapid and non-invasive medical detection method for early detection of lesions.On the other hand,cervical spondylosis has become a common type of disease in people with high incidence,which seriously affects people's health and life treatment.In recent years,the infonnationization of the medical industry has developed rapidly,and the Internet has been widely used by computers and mobile phones.In this context,this paper studies the algorithm of infrared cervical vertebra intelligent analysis,and realizes the auxiliary diagnosis and treatment system based on infrared cervical vertebra image,which provides the auxiliary diagnosis for the initial diagnosis of cervical vertebrae lesions and the function of testing the rehabilitation effect after the treatnent.In this paper,an auxiliary diagnosis related algorithm based on infrared cervical vertebra image analysis is studied.Firstly,the recognition model based on Faster-RCNN deep network is designed.The extracted region of the human body is extracted from the infrared image of the back of the human body,and the cervical vertebrae is automatically extracted,and a good segmentation effect is achieved.Based on the segmented infrared image of cervical vertebrae,an algorithm based on CNN convolutional neural network for automatic extraction and classification of temperature features was designed.The performance of the algorithm was further improved by introducing attention mechanism,and the effective differentiation of cervical spondylosis was realized.In addition,this paper also introduces the concept of chaotic analysis of nonlinear dynamic systems,and also based on the region of interest of the cervical vertebrae extracted by Faster-RCNN for segmentation of high-heat regions.The Markov random field algorithm is adopted,which can extract the required area better than other clustering algorithms.Based on the segmented high heat parts,the boundary contour values are extracted,and the distance between the contour and the centroid is calculated to form a one-dimensional time series.Using the Lyapunov exponent to analyze the chaotic time series,the Lyapunov exponent spectrum is calculated,and the lesion condition is estimated.The basic law between the index value and the lesion condition is obtained.This paper designs a cervical infrared image assisted diagnosis system based on the above research algorithm.The front end is iIplemented by Vuejs framework,the back end is built with Spiing boot framework,the algorithm service interface is provided by python web,and MySQL persistent data is used.Redis implements cache.And build a cluster service with Docker container orchestration.The system can provide a reference for the diagnosis of the doctor.