Research On Image Formation For Combined Intravascular Ultrasound/Photoacoustic/OCT

Intravascular ultrasound(IVUS),intravascular optical coherence tomography(IV-OCT)and intravascular photoacoustic(IVPA)are complementary imaging modalities for the invasive diagnosis and treatment of coronary artery diseases.IVUS can provide overall structures of vessel wall tissues and atheromatous plaques.IV-OCT has very high resolution near to histological resolution and can provide accurate morphology of the superficial wall tissue and vulnerable plaques.IVPA has the ability to assess the functional component of tissues,especially soft plaques.Multimodal imaging technique can image the target with higher precision and contrast than a single modality by combined use of complementary imaging techniques.It can accurately identify vulnerable plaques and analyze functional components qualitatively and quantitatively.The purpose of this thesis is to model and numerically simulate IVUS-PA-OCT combined imaging.The ultrasonic signals reflected,scattered or generated by the tissues and collected by ultrasonic transducers are fused with OCT interference signals reflected by the tissues and then measured by a photoelectric detector from inside the vascular lumen.Two-modal or three-modal combined images of vascular cross-sections are finally formed to display the morphology and functional composition of the wall tissues and vulnerable plaques.First,OCT interference signals and PA signals produced by near infrared light and short pulse laser illuminating the vascular wall tissues are simulated.Also,ultrasonic echoes reflected or scattered by the multi-layered wall tissues scanned by ultrasonic pulses are simulated.The propagation and time-sharing acquisition of OCT signals,PA and ultrasonic echoes by an photodetector/ ultrasonic transducer mounted on the tip of the imaging catheter are simulated.Second,the three types of signals are fused with a weighted fusion algorithm based on variance contribution rate and a a weighted fusion algorithm based on self-adaption to obtain combined images.The validity of the methods was demonstrated with computer-simulated coronary vessel mimicking phantoms with different morphologies and different types of atherosclerotic plaques.The effectiveness of different fusion algorithms was quantitatively evaluate according to experimental results.Results show that the combined imaging signals obtained with the presented methods can retain more information on the morphology,structure and composition of the wall and plaques than a single imaging modality.The combined images have high spatial resolution,contrast and sensitivity,and can clearly display the morphological structure and location of the wall and plaque tissue.