Lamyaa Nkhali

Segmentation of heterogeneous or small FDG PET positive tissue based on a 3D-locally adaptive random walk algorithm

A segmentation algorithm based on the random walk (RW) method, called 3D-LARW, has been developed to delineate small tumors or tumors with a heterogeneous distribution of FDG on PET images. Based on the original algorithm of RW [1], we propose an improved approach using new parameters depending on the Euclidean distance between two adjacent voxels instead of a fixed one and integrating probability densities of labels into the system of linear equations used in the RW. These improvements were evaluated and compared with the original RW method, a thresholding with a fixed value (40% of the maximum in the lesion), an adaptive thresholding algorithm on uniform spheres filled with FDG and FLAB method, on simulated heterogeneous spheres and on clinical data (14 patients). On these three different data, 3D-LARW has shown better segmentation results than the original RW algorithm and the three other methods. As expected, these improvements are more pronounced for the segmentation of small or tumors having heterogeneous FDG uptake.

FDG-PET/CT during concomitant chemo radiotherapy for esophageal cancer: Reducing target volumes to deliver higher radiotherapy doses

Abstract Background. A planning study investigated whether reduced target volumes defined on FDG-PET/CT during radiotherapy allow total dose escalation without compromising normal tissue tolerance in patients with esophageal cancer. Material and methods. Ten patients with esophageal squamous cell carcinoma (SCC), candidate to curative-intent concomitant chemo-radiotherapy (CRT), had FDG-PET/CT performed in treatment position, before and during (Day 21) radiotherapy (RT). Four planning scenarios were investigated: 1) 50 Gy total dose with target volumes defined on pre-RT FDG-PET/CT; 2) 50 Gy with boost target volume defined on FDG-PET/CT during RT; 3) 66 Gy with target volumes from pre-RT FDG-PET/CT; and 4) 66 Gy with boost target volume from during-RT FDG-PET/CT. Results. The median metabolic target volume decreased from 12.9 cm3 (minimum 3.7–maximum 44.8) to 5.0 cm3 (1.7–13.5) (p = 0.01) between pre- and during-RCT FDG-PET/CT. The median PTV66 was smaller on during-RT than on baseline FDG-PET/CT [108 cm3 (62.5–194) vs. 156 cm3 (68.8–251), p = 0.02]. When total dose was set to 50 Gy, planning on during-RT FDG-PET/CT was associated with a marginal reduction in normal tissues irradiation. When total dose was increased to 66 Gy, planning on during-RT PET yielded significantly lower doses to the spinal cord [Dmax = 44.1Gy (40.8–44.9) vs. 44.7Gy (41.5–45.0), p = 0.007] and reduced lung exposure [V20Gy = 23.2% (17.3–27) vs. 26.8% (19.7–30.2), p = 0.006]. Conclusion. This planning study suggests that adaptive RT based on target volume reduction assessed on FDG-PET/CT during treatment could facilitate dose escalation up to 66 Gy in patients with esophageal SCC.

Segmentation of pelvic organs at risk using superpixels and graph diffusion in prostate radiotherapy

Segmentation of organs at risk (OAR) in male pelvis is critical for planning prostate cancer radiotherapy. We are interested in femoral heads, rectum and bladder segmentation in magnetic resonance imaging (MRI) and computed tomography (CT) images in order to protect OARs during radiotherapy planning. The proposed methodology is based on superpixel algorithm in order to over-segment patient image by solving a local Eikonal function from initial seeds. Afterwards, the segmentation is obtained by computing a graph diffusion on a region adjacency graph (RAG) extracted from the over-segmentation thanks to some nodes labeled by the user. Superpixel segmentation is carried out slice-by-slice in 2D. Then, a RAG is constructed in 3D to obtain 3D OAR segmentation. The influence of the initial number of seeds on the segmentation is studied. The performances of the algorithm is evaluated and compared to 4 other methods.