Imaging Algorithm Of Stomach Tumor Detection Based On Ultra-Widebang Capsule Endoscopy

Gastrointestinal diseases are a sort of cancerous diseases which threatens people’s health.Capsule endoscopy is a non-invasive technique that can detect digestive tract diseases,including colon cancers.Due to the complex structure of human gastrointestinal tract,slow transmission of detection signals and slow reading speed of doctors,it restricts the application of this technique in the detection of gastric tumors.A new kind of capsule endoscopy with Ultra-Wideband(UWB)radar is utilized for the first time in the second chapter.Finite difference time domain(FDTD)method is used to establish a electromagnetic simulation model of stomach.After swallowed by a patient,the capsule which consists of the radar moves through esophagus,stomach.Finally it drains out of the body relying on the self peristalsis of the human GI tract.During this time,the radar is used to transmit accurate radar data of human stomach.Then we will carry out electromagnetic inverse scattering imaging by back projection(BP)or frequency wave number migration(F-K)algorithms with the radar data.In order to solve the problems of recognizing the shapes of the tumour in the stomach quickly,several achievements are made as follows:Firstly,the BP algorithm must consider the influence of various tissues in the human body: the attenuation of the signal strength of electromagnetic waves,the decrease in speed and the refraction due to the different permittivity between the different organs of the body.These factors will eventually lead to image offset,and even generate a virtual image.It is effective to refrain the displacement of image with modifying the time element of the imaging algorithm by iterat ion.Secondly we propose a support vector machine(SVM)method.The SVM-based solution successfully deals with the nonlinearity and ill-posedness inherent in thisproblem.Simulation results show the feasibility and effectiveness of the proposed method.The method can effectively locate the tumor target of the stomach regardless of the presence of noise.The positioning effect of the method improves as SNR increases.When the SNR is higher than 50 dB,noise minimally affects the results.Finally,the SVM prediction model is utilized to study the effect of the number of sampling locations on the prediction results.The results show that the more sampling locations,the better the prediction results.Thirdly,a technique based on the combination of improved back-projection(BP)algorithm and support vector machine(SVM)is proposed to solve the problems of rapidly recognizing tumor shapes in the stomach.In this technique,imaging data can be obtained using the improved BP algorithm and are classified by the SVM.Simulation results based on data from the model verify its feasibility and validity.Results further demonstrate that the resolution is extremely high.Tumor shapes,which have different sizes,positions,and quantities,can be reconstructed using this approach.When the data are contaminated by noises,the tumor shape in the stomach can still be suitably predicted,which demonstrates the robustness of the method.Finally,classification accuracy analysis for different sampling distances and sampling inte rvals show that the effects of changing the distance and intervals on shape recognition are limited.The classification accuracy can also be improved by decreasing the sampling intervals or increasing the sampling distance.Lastly,in stomach tumor imaging,traditional time domain algorithm,i.e.,back projection(BP)algorithm,and traditional frequency domain algorithm,i.e.,frequency wave number migration(F-K)algorithm,can locate tumor target accurately.However,BP and F-K algorithms perform poorly in identifying tumor sizes and shapes.In the fifth chapter we propose a method based on combination of Least squares support vector machine(LS-SVM)with BP and F-K algorithms to solve problems in recognizing tumor shape.The method uses field strength ob tained by BP and F-K algorithms as input in SVM to establish the SVM model.Based on BP algorithm,recognition method for LS=SVM includes the following characteristics: short prediction time of SVM and good virtual elimination effect.However,the algorithm requires long periods and possibly misses tumor targets.The recognition method for LS-SVM based on F-K algorithm exhibits the following characteristics: short prediction time of SVM,good virtual elimination effect;the algorithm also works more efficiently,does not miss any tumor targets,and conforms more with requirements of real-time imaging.When the data are contaminated by noises,the tumor shape in the stomach can still be suitably predicted,which demonstrates the robustness of the method.