Deep learning–based fully automated detection and segmentation of lymph nodes on multiparametric-mri for rectal cancer: A multicentre study

Abstract

Background: Accurate lymph nodes (LNs) assessment is important for rectal cancer (RC) staging in multiparametric magnetic resonance imaging (mpMRI). However, it is incredibly time-consumming to identify all the LNs in scan region. This study aims to develop and validate a deep-learning-based, fully-automated lymph node detection and segmentation (auto-LNDS) model based on mpMRI. Methods: In total, 5789 annotated LNs (diameter 3 mm) in mpMRI from 293 patients with RC in a single center were enrolled. Fused T2-weighted images (T2WI) and diffusion-weighted images (DWI) provided input for the deep learning framework Mask R-CNN through transfer learning to generate the auto-LNDS model. The model was then validated both on the internal and external datasets consisting of 935 LNs and 1198 LNs, respectively. The performance for LNs detection was evaluated using sensitivity, positive predictive value (PPV), and false positive rate per case (FP/vol), and segmentation performance was evaluated using the Dice similarity coefficient (DSC). Findings: For LNs detection, auto-LNDS achieved sensitivity, PPV, and FP/vol of 80.0%, 73.5% and 8.6 in internal testing, and 62.6%, 64.5%, and 8.2 in external testing, respectively, significantly better than the performance of junior radiologists. The time taken for model detection and segmentation was 1.3 s/case, compared with 200 s/ case for the radiologists. For LNs segmentation, the DSC of the model was in the range of 0.81À0.82. Interpretation: This deep learningÀbased auto-LNDS model can achieve pelvic LNseffectively based on mpMRI for RC, and holds great potential for facilitating N-staging in clinical practice.

Radiomic signature of the FOWARC trial predicts pathological response to neoadjuvant treatment in rectal cancer

“…The analyzed MRI data included T1W [ 36 ], T2W [ 25 ] and DWI [ 24 ] images from pre-treatment to post-treatment [ 44 ]. However, few of them focused on CT-based radiomics analysis, although it has been demonstrated that multiple radiomics analysis based on CT images can facilitate the prediction of lymph node metastasis [ 37 , 49 , 50 ], distant metastasis [ 51 ], therapy response [ 52 , 53 ] and prognostic outcomes [ 28 ]. Two previous studies have performed CT-based radiomics analysis for pCR prediction but came out with controversial results [ 27 , 54 ].…”

Deep Learning for Differentiating Benign From Malignant Parotid Lesions on MR Images

“…In a recent report, Zhao et al. ( 25 ) developed a deep-learning autoLNDS (lymph node detection and segmentation) model based on mp-MRI. The model can detect and segment LNs (lymph nodes) quickly, yield good clinical efficiency and reduce the difference among physicians with different levels of experience.…”

Deep-learning-assisted detection and segmentation of rib fractures from CT scans: Development and validation of FracNet

“…Recent advances in artificial intelligence and computer vision lead to a rapid development of deep learning technology [1] in medical image analysis and digital medicine [2] , [3] , [4] , [5] , [6] , [7] . With end-to-end learning of deep representation, deep supervised learning, as a unified methodology, achieved remarkable success in numerous 2D and 3D medical image tasks, e.g., classification [8] , detection [9] , segmentation [10] . With the rise of deep learning, infrastructures, algorithms and data (with annotations) are known to be the keys to its success.…”