MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

A Deep Learning Based Auto Segmentation Method for Rectal Tumors On T2 Weighted MR Images

J Lu*, H Ying , j wang , G Qing , L Shen , Y Sun , Z Zhang , W Hu , L Huang , Fudan University Shanghai Cancer Center, Shanghai, Shanghai

Presentations

(Wednesday, 8/1/2018) 7:30 AM - 9:30 AM

Room: Room 202

Purpose: Manually contouring gross tumor volume (GTV) is a crucial and time-consuming process in rectum cancer radiotherapy. This study aims to develop a simple deep learning based auto segmentation algorithm to segment rectal tumors on T2 weighted MR images.

Methods: MRI scans (3T, T2-weighted) of 93 patients with locally advanced (cT3-4 and/or cN1-2) rectal cancer treated with neoadjuvant chemoradiotherapy followed by surgery were enrolled in this study. A 2D U-net similar network was established as a training model. Considering the 3D structure of the MRI image, 5 image slices were simultaneously input into the network. The model was trained in two phases to increase efficiency. These phases were tumor recognition and tumor segmentation. An opening (erosion and dilation) process was implemented to smooth contours after segmentation. Data were randomly separated into training (90%) and validation (10%) datasets for a 10-folder cross-validation. Additionally, 20 patients were double contoured for performance evaluation. Four indices were calculated to evaluate the similarity of automated and manual segmentation, including Hausdorff distance (HD), average surface distance (ASD), Dice index (DSC), and Jaccard index (JSC).

Results: The DSC, JSC, HD, ASD (mean±SD) were 0.74±0.14, 0.60±0.16, 20.44±13.35 and 3.25±1.69mm for validation dataset; and these indices were 0.71±0.13, 0.57±0.15, 14.91±7.62 and 2.67±1.46mm between two human radiation oncologists, respectively. No significant difference has been observed between automated segmentation and manual segmentation considering DSC (p=0.42), JSC (p=0.35), HD (p=0.079) and ASD (p=0.16). However, significant difference was found for HD (p=0.0027) without opening process.

Conclusion: This study showed that a simple deep learning neural network can perform segmentation for rectum cancer based on MRI T2 images with results comparable to a human.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email