Autodelineation of cervical cancers using multiparametric magnetic resonance imaging and machine learning

Torheim, Turid; Malinen, Eirik; Hole, Knut Håkon; Lund, Kjersti V.; Indahl, Ulf Geir; Lyng, Heidi; Kvaal, Knut; Futsæther, Cecilia Marie

doi:10.1080/0284186x.2017.1285499

Cited by 27 publications

(18 citation statements)

References 24 publications

(52 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 83–85 Use of multiple atlases further improves accuracy. 86 Cervix target segmentation on MRI using machine learning results in mean sensitivity and specificity of 85–93% 87 and is faster than atlas based strategies. 88 Accuracy of autosegmentation is not perfect and visual verification is still required.…”

Section: Technical Challenges In the Realization Of Real-time Mrigartmentioning

confidence: 99%

Realizing the potential of magnetic resonance image guided radiotherapy in gynaecological and rectal cancer

White

Scurr

Wetscherek

et al. 2019

BJR

View full text Add to dashboard Cite

CT-based radiotherapy workflow is limited by poor soft tissue definition in the pelvis and reliance on rigid registration methods. Current image-guided radiotherapy and adaptive radiotherapy models therefore have limited ability to improve clinical outcomes. The advent of MRI-guided radiotherapy solutions provides the opportunity to overcome these limitations with the potential to deliver online real-time MRI-based plan adaptation on a daily basis, a true “plan of the day.” This review describes the application of MRI guided radiotherapy in two pelvic tumour sites likely to benefit from this approach.

show abstract

Section: Technical Challenges In the Realization Of Real-time Mrigartmentioning

confidence: 99%

Realizing the potential of magnetic resonance image guided radiotherapy in gynaecological and rectal cancer

White

Scurr

Wetscherek

et al. 2019

BJR

View full text Add to dashboard Cite

show abstract

“…To the best of our knowledge, a few fully automatic segmentation methods have been applied to diagnose cervical cancer. In this work [3], a Fisher's linear discriminant analysis approach was used for cervical cancer segmentation. Related works on pelvic organ segmentation leveraged machine learning techniques for automated tissue labeling using Computed Tomography (CT) and MRI data.…”

Section: Introductionmentioning

confidence: 99%

“…The approach achieved better results compared to each individual classifier model. However, these methods [3][4][5][6] are somewhat limited as they require hand-crafted features and may not be robust against varying image appearances. This motivates us to use Deep Learning (DL) method based on Convolutional Neural Networks (CNNs), where features are extracted and learned automatically for robust segmentation.…”

Section: Introductionmentioning

confidence: 99%

“…Ascending and descending feedbacks between each parent node and its children nodes are dynamically communicated and gradually refined to boost the overall performance of the tree. To address the limitations of aforementioned methods [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17] while leveraging their appealing aspects, we unprecedentedly propose a Cluster-based Dynamic Multi-scale Forest (C K+1 DMF) for automatic cervical cancer segmentation. The proposed C K+1 DMF comprises ( + 1) dynamic multiscale trees (DMTs): a principal tree trained using all data samples, and subordinate trees, each trained with a more homogeneous cluster of data samples.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Multi-scale CNN Forest Learning for Automatic Cervical Cancer Segmentation

Bnouni

Rekik

Rhim

et al. 2018

Machine Learning in Medical Imaging

View full text Add to dashboard Cite

Deep-learning based labeling methods have gained unprecedented popularity in different computer vision and medical image segmentation tasks. However, to the best of our knowledge, these have not been used for cervical tumor segmentation. More importantly, while the majority of innovative deeplearning works using convolutional neural networks (CNNs) focus on developing more sophisticated and robust architectures (e.g., ResNet, U-Net, GANs), there is very limited work on how to aggregate different CNN architectures to improve their relational learning at multiple levels of CNN-to-CNN interactions. To address this gap, we introduce a Dynamic Multi-Scale CNN Forest (C K+1 DMF), which aims to address three major issues in medical image labeling and ensemble CNN learning: (1) heterogeneous distribution of MRI training patches, (2) a bidirectional flow of information between two consecutive CNNs as opposed to cascading CNNs-where information passes in a directional way from current to the next CNN in the cascade, and (3) multiscale anatomical variability across patients. To solve the first issue, we group training samples into clusters, then design a forest with (+ 1) trees: a principal tree of CNNs trained using all data samples and subordinate trees, each trained using a cluster of samples. As for the second and third issues, we design each dynamic multiscale tree (DMT) in the forest such that each node in the tree nests a CNN architecture. Two successive CNN nodes in the tree pass bidirectional contextual maps to progressively improve the learning of their relational non-linear mapping. Besides, as we traverse a path from the root node to a leaf node in the tree, the architecture of each CNN node becomes shallower to take in smaller training patches. Our C K+1 DMF significantly (p<0.05) outperformed several conventional and ensemble CNN architectures, including conventional CNN (improvement by 10.3%) and CNN-based DMT (improvement by 5%).

show abstract

“…This substantiates the need for more robust and objective methodologies and/or guidelines for delineation. In this issue, Torheim et al explored a computer-assisted method to extract cervical tumor volume from multi-parametric MRI and machine learning [26]. T1-and T2-weighted alongside dynamic contrast enhanced (DCE) MRI scans were used in discriminant analysis to identify tumor from normal tissue.…”

mentioning

confidence: 99%

Bridging imaging and therapy: the role of medical physics in development of precision cancer care

et al. 2017

Self Cite

View full text Add to dashboard Cite

Autodelineation of cervical cancers using multiparametric magnetic resonance imaging and machine learning

Cited by 27 publications

References 24 publications

Realizing the potential of magnetic resonance image guided radiotherapy in gynaecological and rectal cancer

Realizing the potential of magnetic resonance image guided radiotherapy in gynaecological and rectal cancer

Dynamic Multi-scale CNN Forest Learning for Automatic Cervical Cancer Segmentation

Bridging imaging and therapy: the role of medical physics in development of precision cancer care

Contact Info

Product

Resources

About