Photoacoustic Classification of Tumor Model Morphology Based on Support Vector Machine: A Simulation and Phantom Study

Precise classification to distinguish malignancy of tumors is of great significance to accomplish high specificity diagnosis and avert unnecessary biopsy. Photoacoustic tomography (PAT) is a burgeoning new imaging modality, which combines optical contrast and ultrasound penetrating in a deep medium. However, it has not been fully exploited on the capability of PAT to discriminate tumors malignancy. In this paper, a multistatic classification approach in photoacoustic is proposed, which could discriminate malignant/benign tumors based on a feature in clinical diagnosis that tumors show different shape irregularity. The multistatic photoacoustic waves extract two different features to differentiate the two types of tumor models with high accuracy in three different scenarios using Support Vector Machines. In addition, two conventional PAT image reconstructing algorithm are also performed to reconstruct images as a comparative study, which unfortunately cannot differentiate their malignancy precisely because of limited detector bandwidth and strong acoustic distortion. We performed the feasibility study in this paper with both simulation and experimental results, which shows that the proposed multistatic photoacoustic classification to distinguish between malignant and benign tumors models works well, and could be easily applied for state-of-art array-based PAT system to ameliorate the diagnostic accuracy.