Two Stage Segmentation of Cervical Tumors using PocketNet

Awj Twam,Megan Jacobsen,Rachel Glenn,Ann Klopp,Aradhana M. Venkatesan,David Fuentes

2024-09-18

Abstract:Cervical cancer remains the fourth most common malignancy amongst women worldwide.1 Concurrent chemoradiotherapy (CRT) serves as the mainstay definitive treatment regimen for locally advanced cervical cancers and includes external beam radiation followed by brachytherapy.2 Integral to radiotherapy treatment planning is the routine contouring of both the target tumor at the level of the cervix, associated gynecologic anatomy and the adjacent organs at risk (OARs). However, manual contouring of these structures is both time and labor intensive and associated with known interobserver variability that can impact treatment outcomes. While multiple tools have been developed to automatically segment OARs and the high-risk clinical tumor volume (HR-CTV) using computed tomography (CT) images,3,4,5,6 the development of deep learning-based tumor segmentation tools using routine T2-weighted (T2w) magnetic resonance imaging (MRI) addresses an unmet clinical need to improve the routine contouring of both anatomical structures and cervical cancers, thereby increasing quality and consistency of radiotherapy planning. This work applied a novel deep-learning model (PocketNet) to segment the cervix, vagina, uterus, and tumor(s) on T2w MRI. The performance of the PocketNet architecture was evaluated, when trained on data via 5-fold cross validation. PocketNet achieved a mean Dice-Sorensen similarity coefficient (DSC) exceeding 70% for tumor segmentation and 80% for organ segmentation. These results suggest that PocketNet is robust to variations in contrast protocols, providing reliable segmentation of the ROIs.

Computer Vision and Pattern Recognition,Artificial Intelligence,Machine Learning

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: **Improve the accuracy and efficiency of automatic segmentation of tumors and related anatomical structures in cervical cancer radiotherapy planning**. Specifically, manual segmentation of structures such as the cervix, uterus, vagina, and tumors is time - consuming and is easily affected by differences among different observers, which may affect the treatment effect. To solve these problems, researchers have developed a new deep - learning - based model - PocketNet, a two - stage segmentation method on T2 - weighted magnetic resonance imaging (T2w MRI), in order to improve the accuracy and consistency of segmentation and thus optimize the radiotherapy plan for cervical cancer. ### Core of the problem: 1. **Limitations of manual segmentation**: Manual segmentation of tumors and their surrounding anatomical structures in cervical cancer patients is a cumbersome and time - consuming task, and there are significant inter - observer differences at the same time. 2. **Automation requirements**: In order to improve efficiency and reduce human error, a method that can automatically generate high - quality segmentation results is required. 3. **Deficiencies of existing tools**: Although there are already tools that can automatically segment organs at risk (OARs) and high - risk clinical target volumes (HR - CTV), there are still unmet clinical needs in tumor segmentation using conventional T2 - weighted MRI. ### Solutions: - **PocketNet model**: This is a compact deep - learning architecture, which aims to reduce the demand for computing resources by reducing the number of parameters while maintaining performance similar to that of a full - size network. - **Two - stage segmentation process**: - Stage 1: Segment organs (such as the uterus, cervix, vagina) including tumors. - Stage 2: On the basis of the first stage, further accurately segment the tumor to ensure that the tumor segmentation result is located inside the organ. ### Expected results: - **Improve segmentation accuracy**: Verified by experiments, the PocketNet model has achieved a Dice similarity coefficient (DSC) of more than 80% in organ segmentation, and also exceeded 70% DSC in tumor segmentation. - **Enhance robustness**: The model has a strong adaptability to different contrast protocols and can provide reliable ROI (region of interest) segmentation in various situations. - **Promote clinical applications**: This efficient and accurate automatic segmentation method helps to accelerate the radiotherapy planning process and may improve the treatment results of patients. In summary, the main goal of this research is to solve the problems of low efficiency and insufficient accuracy in manual segmentation in current cervical cancer radiotherapy planning by introducing the PocketNet model, thereby improving the treatment quality and consistency.

Two Stage Segmentation of Cervical Tumors using PocketNet

Deep Learning-Based Automatic Segmentation for Size and Volumetric Measurement of Breast Cancer on Magnetic Resonance Imaging.

Automated Clinical Target Volume Delineation Using Deep 3D Neural Networks in Radiation Therapy of Non-small Cell Lung Cancer

Automatic gross tumor volume segmentation with failure detection for safe implementation in locally advanced cervical cancer

Segmentation of Clinical Target Volume From CT Images for Cervical Cancer Using Deep Learning

Dual convolution-transformer UNet (DCT-UNet) for organs at risk and clinical target volume segmentation in MRI for cervical cancer brachytherapy

A deep learning-based framework for segmenting invisible clinical target volumes with estimated uncertainties for post-operative prostate cancer radiotherapy

Technical Note: A deep learning based auto segmentation of rectal tumors in MR images.

Transfer Learning from Tumors to Organs at Risk for Cervical Cancer Image Segmentation.

Clinical evaluation on automatic segmentation results of convolutional neural networks in rectal cancer radiotherapy

Clinical evaluation of deep learning-based automatic clinical target volume segmentation: a single-institution multi-site tumor experience

2022-RA-452-ESGO Automatic semantic segmentation of cervical cancer based on dynamic contrast-enhanced MRI and fully convolutional networks

Automatic delineation of cervical cancer target volumes in small samples based on multi-decoder and semi-supervised learning and clinical application

AugMS-Net:Augmented Multiscale Network for Small Cervical Tumor Segmentation from MRI Volumes

Image segmentation of treated and untreated tumor spheroids by Fully Convolutional Networks

Fully Semantic Segmentation for Rectal Cancer Based on Post-Ncrt MRl Modality and Deep Learning Framework

Comprehensive Benchmarking of CNN-Based Tumor Segmentation Methods Using Multimodal MRI Data

Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy

Segmentation of Planning Target Volume in CT Series for Total Marrow Irradiation Using U-Net

Automatic clinical target volume delineation for cervical cancer in CT images using deep learning

MRI-based Head and Neck Tumor Segmentation Using nnU-Net with 15-fold Cross-Validation Ensemble