Evaluating image analysis techniques for ultrasound optical tomography in breast tissue

Sammanfattning: Breast cancer is the most commonly diagnosed cancer in the world today and early detection is crucial to minimize the mortality. ”Ultrasound optical tomography” (UOT) is a method under development for deep tissue imaging. The technique combines the resolution of ultrasound with the tissue contrast from light to image optical absorption in tissues. This report presents simulations analyzing how changing the light configuration can improve the imaging performance as well as two reconstruction methods, relative and absolute absorption reconstruction, to process the signal to be able to detect tumors. The simulations were conducted using a simulated breast tissue sample with in vivo optical parameters. The results indicated that a larger light-source can increase the light power and improve the signal-to-noise ratio and contrast-to-noise ratio. Both reconstruction methods improved the readability compared to the raw signal. The ratio reconstruction performed well in localizing the tumor but with poor resolution while the absorption reconstruction revealed the tumor with high resolution and contrast but the method also generated artifacts. In conclusion, indications suggest that a larger light-source will improve the performance of a UOT-system and the two reconstructions method can enhance the detection capability of the signal but neither has a high enough performance to work on its own at this stage and further developments are required.