3D active surfaces for liver segmentation in multisequence MRI images

Bereciartua, Arantza; Picón, Artzai; Galdrán, Adrián; Iriondo, Pedro

doi:10.1016/j.cmpb.2016.04.028

Cited by 11 publications

(10 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For CT, state-of-the-art algorithms typically have shown good performance, with mean Dice scores of 0.93-0.95 (13)(14)(15). For MRI, a wide performance range has been reported, with mean Dice scores ranging from 0.85 to 0.95 depending on the specific MR technique and contrast material used (9,(16)(17)(18)(19)34,35). In this work, we explored the feasibility of a single multimodal CNN to perform this task with comparable accuracy across different imaging modalities and techniques.…”

Section: Discussionmentioning

confidence: 99%

“…While efforts have been made to develop segmentation algorithms for a single modality or for a particular phase of intravenous contrast material enhancement (9,(13)(14)(15)(16)(17)(18)(19), a generalized algorithm that is robust across multiple imaging modalities, techniques, sequences, signal weightings, and phases of contrast enhancement would be beneficial for many clinical applications. For example, a patient might undergo different types of CT and MRI examinations during routine clinical care.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Automated CT and MRI Liver Segmentation and Biometry Using a Generalized Convolutional Neural Network

Wang¹,

Mamidipalli²,

Retson³

et al. 2019

Radiology: Artificial Intelligence

101

View full text Add to dashboard Cite

To assess feasibility of training a convolutional neural network (CNN) to automate liver segmentation across different imaging modalities and techniques used in clinical practice and to apply this technique to enable automation of liver biometry. Materials and Methods: A two-dimensional U-Net CNN was trained for liver segmentation in two stages by using 330 abdominal MRI and CT examinations. First, the neural network was trained with unenhanced multiecho spoiled gradient-echo images from 300 MRI examinations to yield multiple signal weightings. Then, transfer learning was used to generalize the CNN with additional images from 30 contrast material-enhanced MRI and CT examinations. Performance of the CNN was assessed by using a distinct multiinstitutional dataset curated from multiple sources (498 subjects). Segmentation accuracy was evaluated by computing Dice scores. These segmentations were used to compute liver volume from CT and T1-weighted MRI examinations and to estimate hepatic proton density fat fraction (PDFF) from multiecho T2*-weighted MRI examinations. Quantitative volumetry and PDFF estimates were compared between automated and manual segmentation by using Pearson correlation and Bland-Altman statistics. Results: Dice scores were 0.94 ± 0.06 for CT (n = 230), 0.95 ± 0.03 (n = 100) for T1-weighted MRI, and 0.92 ± 0.05 for T2*weighted MRI (n = 168). Liver volume measured with manual and automated segmentation agreed closely for CT (95% limits of agreement: −298 mL, 180 mL) and T1-weighted MRI (95% limits of agreement: −358 mL, 180 mL). Hepatic PDFF measured by the two segmentations also agreed closely (95% limits of agreement: −0.62%, 0.80%). Conclusion: By using a transfer-learning strategy, this study has demonstrated the feasibility of a CNN to be generalized to perform liver segmentation across different imaging techniques and modalities. With further refinement and validation, CNNs may have broad applicability for multimodal liver volumetry and hepatic tissue characterization.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

Automated CT and MRI Liver Segmentation and Biometry Using a Generalized Convolutional Neural Network

Wang¹,

Mamidipalli²,

Retson³

et al. 2019

Radiology: Artificial Intelligence

101

View full text Add to dashboard Cite

show abstract

“…Manual and automated image segmentation from multisequence MR data is a challenging problem because of the presence of different contrasts requiring significant amounts of training data and/or using complementary imaging information within multiple sequences . Sophisticated solutions have been proposed, but each require segmenting individual channels, which considerably increases the computational complexity.…”

Section: Discussionmentioning

confidence: 99%

“…The multisequence nature of MR imaging presents an ideal framework for developing specialized methods within the emerging area of multi‐image automatic segmentation strategies, in which using multiple images in segmentation models can substantially improve the accuracy and robustness of algorithms . Increasingly, automated MR segmentation algorithms are becoming more accurate and faster, producing localization and shape of the targeted object(s) in several minutes, which required experts/operators substantially longer to complete .…”

Section: Introductionmentioning

confidence: 99%

Local contrast‐enhanced MR images via high dynamic range processing

Chandra

Engstrom

Fripp

et al. 2018

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…However, 2D approaches lack information in the third axis. This information when dealing with 3D objects, as it is the case with human tissues and organs, may provide details that will lead to a more precise solution [ 44 ].…”

Section: Methodsmentioning

confidence: 99%

MRI Deep Learning-Based Solution for Alzheimer’s Disease Prediction

Saratxaga

Moya

Picón

et al. 2021

JPM

View full text Add to dashboard Cite

Background: Alzheimer’s is a degenerative dementing disorder that starts with a mild memory impairment and progresses to a total loss of mental and physical faculties. The sooner the diagnosis is made, the better for the patient, as preventive actions and treatment can be started. Although tests such as the Mini-Mental State Tests Examination are usually used for early identification, diagnosis relies on magnetic resonance imaging (MRI) brain analysis. Methods: Public initiatives such as the OASIS (Open Access Series of Imaging Studies) collection provide neuroimaging datasets openly available for research purposes. In this work, a new method based on deep learning and image processing techniques for MRI-based Alzheimer’s diagnosis is proposed and compared with previous literature works. Results: Our method achieves a balance accuracy (BAC) up to 0.93 for image-based automated diagnosis of the disease, and a BAC of 0.88 for the establishment of the disease stage (healthy tissue, very mild and severe stage). Conclusions: Results obtained surpassed the state-of-the-art proposals using the OASIS collection. This demonstrates that deep learning-based strategies are an effective tool for building a robust solution for Alzheimer’s-assisted diagnosis based on MRI data.

show abstract

3D active surfaces for liver segmentation in multisequence MRI images

Cited by 11 publications

References 33 publications

Automated CT and MRI Liver Segmentation and Biometry Using a Generalized Convolutional Neural Network

Automated CT and MRI Liver Segmentation and Biometry Using a Generalized Convolutional Neural Network

Local contrast‐enhanced MR images via high dynamic range processing

MRI Deep Learning-Based Solution for Alzheimer’s Disease Prediction

Contact Info

Product

Resources

About