Improving Automatic Segmentation of lymphoma with Additional Medical Knowledge Priors

Lymphoma is a type of malignant tumor that is often fatal to people of all ages. Positron Emission Tomography (PET) / Computed Tomography (CT) is the primary imaging method to assess lymphoma and monitor treatment response. As PET is sensitive to identify lymphoma regions while CT provides detailed anatomic information, the two imaging modalities can complement each other to enable improved diagnosis. However, automatic lymphoma segmentation is still a challenging task due to its substantial size and shape variability and limited datasets for training. To that end, we designed an automatic lymphoma segmentation model with nnU-net as the backbone. Our pipeline incorporates pre- and post- processing mechanisms to remove regions of normal increased radiotracer uptake, and augment training with non-lymphoma samples. The proposed method was validated on the PET/CT scans from forty-six patients. Experimental results revealed that the Dice coefficient improved from 0.263 to 0.477 in comparison with the baseline nnU-Net. The results and analysis demonstrate the efficacy of the proposed method.

1. INTRODUCTION

Lymphoma, a hematopoietic malignancy with numerous subtypes, can affect people of all ages [1]. Precise and accurate identification of lymphomatous lesions, and measurement of tumor burden, are pivotal for lymphoma prognosis and treatment response. Positron Emission Tomography (PET) / Computed Tomography (CT) [2], [3] is the primary imaging method to assess lymphoma and monitor treatment response. In PET imaging [2], Standardized Uptake Value (SUV) is used to measure 18F-fluorodeoxyglucose (FDG) uptake or glucose metabolism of tumor regions in the PET scan. Lymphoma manifests visually as metabolically hyperactive regions in PET scans. The advantage of combined PET/CT imaging method is that PET is sensitive to identify lymphoma regions whilst CT preserves anatomical structures. Combined PET/CT is therefore useful for lymphoma diagnosis as it helps to identify overall tumour burden and treatment effects. However, segmentation of lymphoma on PET/CT image is still a challenging task because it has various patterns, for example; (1) dispersed throughout the body, (2) outside organs (3) inside organs (4) small delocalized patches. Furthermore, the brain, heart, bowel, kidneys and bladder may also have high SUV responses, which mimics the distinct appearance of lymphoma. These challenges hinder lymphoma segmentation from practical application.