Orchestral fully convolutional networks for small lesion segmentation in brain MRI

Xu, Botian; Chai, Yaqiong; Galarza, Cristina; Vu, Chau; Tamrazi, Benita; Gaonkar, Bilwaj; Macyszyn, Luke; Coates, Thomas D.; Leporé, Natasha; Wood, John C.

doi:10.1109/isbi.2018.8363714

Cited by 15 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…imaging background and one medical graduate student). The number and location of each WMH were confirmed by an experienced board certificated neuroradiologist, as described in detail in a previous study (Xu et al, 2018), with accumulated WMHs of 132 distributed mostly in frontal and parietal lobes (Table 1). While the lesions were typically quite small and were not expected to affect the tractography, we performed quality control on the tracts generated by the TSA to ensure their integrity, particularly around the lesions.…”

Section: Hemoglobin Bmentioning

confidence: 98%

Tract‐specific analysis and neurocognitive functioning in sickle cell patients without history of overt stroke

Chai

Coloigner

et al. 2021

Brain and Behavior

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Hemoglobin Bmentioning

confidence: 98%

Tract‐specific analysis and neurocognitive functioning in sickle cell patients without history of overt stroke

Chai

Coloigner

et al. 2021

Brain and Behavior

Self Cite

View full text Add to dashboard Cite

show abstract

“…Secondly, inspired by the success of multi-task learning [41,42], Deep Contour-aware Networks (DCAN) [43] and Boundaryaware FCN [44] employ contour-aware loss functions for better discrimination between boundaries and the background. In addition, Multilevel Contextual 3D CNNs [45], DeepMedic [46], Orchestral Fully Convolutional Networks (OFCNs) [47] and Hybrid Loss guided Fully Convolutional Networks (HL-FCNs) [48] adopt model ensemble for better robustness.…”

Section: Introductionmentioning

confidence: 99%

3-D RoI-Aware U-Net for Accurate and Efficient Colorectal Tumor Segmentation

Huang

Dou

Wang

et al. 2021

IEEE Trans. Cybern.

View full text Add to dashboard Cite

Segmentation of colorectal cancerous regions from 3DMagnetic Resonance (MR) images is a crucial procedure for radiotherapy which conventionally requires accurate delineation of tumour boundaries at an expense of labor, time and reproducibility. While deep learning based methods serve good baselines in 3D image segmentation tasks, small applicable patch size limits effective receptive field and degrades segmentation performance. In addition, Regions of interest (RoIs) localization from large whole volume 3D images serves as a preceding operation that brings about multiple benefits in terms of speed, target completeness, reduction of false positives. Distinct from sliding window or non-joint localizationsegmentation based models, we propose a novel multitask framework referred to as 3D RoI-aware U-Net (3D RU-Net), for RoI localization and in-region segmentation where the two tasks share one backbone encoder network. With the region proposals from the encoder, we crop multi-level RoI in-region features from the encoder to form a GPU memory-efficient decoder for detailpreserving segmentation and therefore enlarged applicable volume size and effective receptive field. To effectively train the model, we designed a Dice formulated loss function for the global-to-local multi-task learning procedure. Based on the efficiency gains demonstrated by the proposed method, we went on to ensemble models with different receptive fields to achieve even higher performance costing minor extra computational expensiveness. Extensive experiments were subsequently conducted on 64 cancerous cases with a four-fold cross-validation, and the results showed significant superiority in terms of accuracy and efficiency over conventional state-of-the art frameworks. In conclusion, the proposed method has a huge potential for extension to other 3D object segmentation tasks from medical images due to its inherent generalizability. The code for the proposed method is publicly available.3D CNN, region of interest, multi-task learning, tumor segmentation, colorectal cancer.

show abstract

“…Stroke has non-reversible effects on the human body so there is a need for early diagnosis and treatment [26]. Also, white matter lesions are common in neural disorders such as multiple sclerosis, small vessel disease and anaemia [124]. Identification and measurement of regions of interest is an important step in the identification of neurological disorders [124,26].…”

Section: Brain Mri Segmentationmentioning

confidence: 99%

A Patch-Based Convolutional Neural Network for Localized MRI Brain Segmentation

Vambe

Viriri

Gwetu

2022

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

The authors' contributions in the paper above are as follows:-1st Author : Concept formulation, literature review, design and implementation of algorithms and preparing and editing the write-up.-2nd and 3rd Authors : Concept improvement, providing advice, highlighting important considerations with respect to methods and techniques, and proof-reading manuscripts.I would like to express profound gratitude to my supervisors, that is, Prof. Serestina Viriri and Dr Mandlenkosi Gwetu, for their invaluable counsel, guidance and feedback during the work related to this dissertation. I would also like to extend heartfelt gratitude to several academics for their motivation and advice during my research. Last but not least I would like to thank my family for their continued support every step of the way. v 2.5D Two and a Half Dimensional. 5

show abstract

Orchestral fully convolutional networks for small lesion segmentation in brain MRI

Cited by 15 publications

References 17 publications

Tract‐specific analysis and neurocognitive functioning in sickle cell patients without history of overt stroke

Tract‐specific analysis and neurocognitive functioning in sickle cell patients without history of overt stroke

3-D RoI-Aware U-Net for Accurate and Efficient Colorectal Tumor Segmentation

A Patch-Based Convolutional Neural Network for Localized MRI Brain Segmentation

Contact Info

Product

Resources

About