Publications

Search or filter publications

Filter by type:

Filter by publication type

Filter by year:

to

Results

  • Showing results for:
  • Reset all filters

Search results

  • Journal article
    Zhuang X, Li L, Payer C, Stern D, Urschler M, Heinrich M, Oster J, Wang C, Smedby O, Bian C, Yang X, Heng P-A, Mortazi A, Bagci U, Yang G, Sun C, Galisot G, Ramel J-Y, Brouard T, Tong Q, Si W, Liao X, Zeng G, Shi Z, Zheng G, Wang C, MacGillivray T, Newby D, Rhode K, Ourselin S, Mohiaddin R, Keegan J, Firmin D, Yang Get al., 2019,

    Evaluation of algorithms for multi-modality whole heart segmentation: An open-access grand challenge

    , Medical Image Analysis, Vol: 58, ISSN: 1361-8415

    Knowledge of whole heart anatomy is a prerequisite for many clinical applications. Whole heart segmentation (WHS),which delineates substructures of the heart, can be very valuable for modeling and analysis of the anatomy and functionsof the heart. However, automating this segmentation can be challenging due to the large variation of the heart shape,and different image qualities of the clinical data. To achieve this goal, an initial set of training data is generally neededfor constructing priors or for training. Furthermore, it is difficult to perform comparisons between different methods,largely due to differences in the datasets and evaluation metrics used. This manuscript presents the methodologiesand evaluation results for the WHS algorithms selected from the submissions to the Multi-Modality Whole Heart Segmentation (MM-WHS) challenge, in conjunction with MICCAI 2017. The challenge provided 120 three-dimensionalcardiac images covering the whole heart, including 60 CT and 60 MRI volumes, all acquired in clinical environmentswith manual delineation. Ten algorithms for CT data and eleven algorithms for MRI data, submitted from twelvegroups, have been evaluated. The results showed that the performance of CT WHS was generally better than thatof MRI WHS. The segmentation of the substructures for different categories of patients could present different levelsof challenge due to the difference in imaging and variations of heart shapes. The deep learning (DL)-based methodsdemonstrated great potential, though several of them reported poor results in the blinded evaluation. Their performance could vary greatly across different network structures and training strategies. The conventional algorithms,mainly based on multi-atlas segmentation, demonstrated good performance, though the accuracy and computationalefficiency could be limited. The challenge, including provision of the annotated training data and the blinded evaluation for submitted algorithms on the test data, conti
  • Journal article
    Li L, Wu F, Yang G, Xu L, Wong T, Mohiaddin R, Firmin D, Keegan J, Zhuang Xet al.,

    Atrial scar quantification via multi-scale CNN in the graph-cuts framework

    , Medical Image Analysis, ISSN: 1361-8415

    Late gadolinium enhancement magnetic resonance imaging (LGE MRI) appears to be a promising alternative for scarassessment in patients with atrial fibrillation (AF). Automating the quantification and analysis of atrial scars can bechallenging due to the low image quality. In this work, we propose a fully automated method based on the graph-cutsframework, where the potentials of the graph are learned on a surface mesh of the left atrium (LA) using a multi-scaleconvolutional neural network (MS-CNN). For validation, we have included fifty-eight images with manual delineations.MS-CNN, which can efficiently incorporate both the local and global texture information of the images, has been shownto evidently improve the segmentation accuracy of the proposed graph-cuts based method. The segmentation could befurther improved when the contribution between the t-link and n-link weights of the graph is balanced. The proposedmethod achieves a mean accuracy of 0.856 ± 0.033 and mean Dice score of 0.702 ± 0.071 for LA scar quantification.Compared to the conventional methods, which are based on the manual delineation of LA for initialization, our methodis fully automatic and has demonstrated significantly better Dice score and accuracy (p < 0.01). The method is promisingand can be potentially useful in diagnosis and prognosis of AF.
  • Journal article
    Mills M, Pelting V, Harris LM, Smith J, Aiton N, Rabe H, Fernandez-Alvarez JRet al., 2019,

    Comparison of MRI and neurosonogram 1-and 2-dimensional morphological measurements of the newborn corpus callosum

    , PEDIATRIC RESEARCH, Vol: 86, Pages: 355-359, ISSN: 0031-3998
  • Journal article
    Inglese M, Katherine L O, Lesley H, Tara D B, Eric O A, Adam D W, Matthew G-Set al.,

    Reliability of dynamic contrast enhanced magnetic resonance imaging data in primary brain tumours: a comparison of Tofts and Shutter Speed Models

    , Neuroradiology, ISSN: 0028-3940

    PurposeTo investigate the robustness of pharmacokinetic modelling of DCE-MRI brain tumourdata and to ascertain reliable perfusion parameters through a model selection processand a stability test.MethodsDCE-MRI data of 14 patients with primary brain tumours were analysed using the Toftsmodel (TM), the extended Tofts model (ETM), the shutter speed model (SSM) and theextended shutter speed model (ESSM). A no-effect model (NEM) was implemented toassess overfitting of data by the other models.For each lesion, the Akaike Information Criteria (AIC) was used to build a 3D modelselection map. The variability of each pharmacokinetic parameter extracted from thismap was assessed with a noise propagation procedure, resulting in voxel-wisedistributions of the coefficient of variation (CV).ResultsThe model selection map over all patients showed NEM had the best fit in 35.5% ofvoxels, followed by ETM (32%), TM (28.2%), SSM (4.3%) and ESSM (<0.1%). Inanalysing the reliability of Ktrans, when considering regions with a CV<20%, ≈25% ofvoxels were found to be stable across all patients. The remaining 75% of voxels wereconsidered unreliable.ConclusionsThe majority of studies quantifying DCE-MRI data in brain tumours only consider asingle model and whole-tumour statistics for the output parameters. Appropriate modelselection, considering tissue biology and its effects on blood brain barrier permeabilityand exchange conditions, together with an analysis on the reliability and stability of thecalculated parameters, is critical in processing robust brain tumour DCE-MRI data.
  • Conference paper
    Ali AR, Li J, O'Shea SJ, Yang G, Trappenberg T, Ye Xet al., 2019,

    A Deep Learning Based Approach to Skin Lesion Border Extraction with a Novel Edge Detector in Dermoscopy Images

    © 2019 IEEE. Lesion border detection is considered a crucial step in diagnosing skin cancer. However, performing such a task automatically is challenging due to the low contrast between the surrounding skin and lesion, ambiguous lesion borders, and the presence of artifacts such as hair. In this paper we propose a two-stage approach for skin lesion border detection: (i) segmenting the skin lesion dermoscopy image using U-Net, and (ii) extracting the edges from the segmented image using a novel approach we call FuzzEdge. The proposed approach is compared with another published skin lesion border detection approach, and the results show that our approach performs better in detecting the main borders of the lesion and is more robust to artifacts that might be present in the image. The approach is also compared with the manual border drawings of a dermatologist, resulting in an average Dice similarity of 87.7%.
  • Journal article
    Chappell K, Brujic D, Van Der Straeten C, Meeson R, Gedroyc W, McRobbie D, Ristic Met al., 2019,

    Detection of maturity and ligament injury using magic angle directional imaging

    , Magnetic Resonance in Medicine, ISSN: 0740-3194

    PurposeTo investigate whether magnetic field–related anisotropies of collagen may be correlated with postmortem findings in animal models.MethodsOptimized scan planning and new MRI data‐processing methods were proposed and analyzed using Monte Carlo simulations. Six caprine and 10 canine knees were scanned at various orientations to the main magnetic field. Image intensities in segmented voxels were used to compute the orientation vectors of the collagen fibers. Vector field and tractography plots were computed. The Alignment Index was defined as a measure of orientation distribution. The knees were subsequently assessed by a specialist orthopedic veterinarian, who gave a pathological diagnosis after having dissected and photographed the joints.ResultsUsing 50% less scans than reported previously can lead to robust calculation of fiber orientations in the presence of noise, with much higher accuracy. The 6 caprine knees were found to range from very immature (< 3 months) to very mature (> 3 years). Mature specimens exhibited significantly more aligned collagen fibers in their patella tendons compared with the immature ones. In 2 of the 10 canine knees scanned, partial cranial caudal ligament tears were identified from MRI and subsequently confirmed with encouragingly high consistency of tractography, Alignment Index, and dissection results.ConclusionThis method can be used to detect injury such as partial ligament tears, and to visualize maturity‐related changes in the collagen structure of tendons. It can provide the basis for new, noninvasive diagnostic tools in combination with new scanner configurations that allow less‐restricted field orientations.
  • Journal article
    Zhang N, Yang G, Gao Z, Xu C, Zhang Y, Shi R, Keegan J, Xu L, Zhang H, Fan Z, Firmin Det al., 2019,

    Deep learning for diagnosis of chronic myocardial infarction on nonenhanced cardiac cine MRI

    , Radiology, ISSN: 0033-8419

    BackgroundRenal impairment is common in patients with coronary artery disease and, if severe, late gadolinium enhancement (LGE) imaging for myocardial infarction (MI) evaluation cannot be performed.PurposeTo develop a fully automatic framework for chronic MI delineation via deep learning on non–contrast material–enhanced cardiac cine MRI.Materials and MethodsIn this retrospective single-center study, a deep learning model was developed to extract motion features from the left ventricle and delineate MI regions on nonenhanced cardiac cine MRI collected between October 2015 and March 2017. Patients with chronic MI, as well as healthy control patients, had both nonenhanced cardiac cine (25 phases per cardiac cycle) and LGE MRI examinations. Eighty percent of MRI examinations were used for the training data set and 20% for the independent testing data set. Chronic MI regions on LGE MRI were defined as ground truth. Diagnostic performance was assessed by analysis of the area under the receiver operating characteristic curve (AUC). MI area and MI area percentage from nonenhanced cardiac cine and LGE MRI were compared by using the Pearson correlation, paired t test, and Bland-Altman analysis.ResultsStudy participants included 212 patients with chronic MI (men, 171; age, 57.2 years ± 12.5) and 87 healthy control patients (men, 42; age, 43.3 years ± 15.5). Using the full cardiac cine MRI, the per-segment sensitivity and specificity for detecting chronic MI in the independent test set was 89.8% and 99.1%, respectively, with an AUC of 0.94. There were no differences between nonenhanced cardiac cine and LGE MRI analyses in number of MI segments (114 vs 127, respectively; P = .38), per-patient MI area (6.2 cm2 ± 2.8 vs 5.5 cm2 ± 2.3, respectively; P = .27; correlation coefficient, r = 0.88), and MI area percentage (21.5% ± 17.3 vs 18.5% ± 15.4; P = .17; correlation coefficient, r = 0.89).ConclusionThe proposed deep learning f
  • Journal article
    Sun Y, Reynolds HM, Wraith D, Williams S, Finnegan ME, Mitchell C, Murphy D, Haworth Aet al., 2019,

    Automatic stratification of prostate tumour aggressiveness using multiparametric MRI: a horizontal comparison of texture features

    , ACTA ONCOLOGICA, Vol: 58, Pages: 1118-1126, ISSN: 0284-186X
  • Journal article
    Klemt C, Nolte D, Ding Z, Rane L, Quest RA, Finnegan ME, Walker M, Reilly P, Bull Aet al., 2019,

    Anthropometric scaling of anatomical datasets for subject-specific musculoskeletal modelling of the shoulder

    , Annals of Biomedical Engineering, Vol: 47, Pages: 924-936, ISSN: 0090-6964

    Linear scaling of generic shoulder models leads to substantial errors in model predictions. Customisation of shoulder modelling through magnetic resonance imaging (MRI) improves modelling outcomes, but model development is time and technology intensive. This study aims to validate 10 MRI-based shoulder models, identify the best combinations of anthropometric parameters for model scaling, and quantify the improvement in model predictions of glenohumeral loading through anthropometric scaling from this anatomical atlas. The shoulder anatomy was modelled using a validated musculoskeletal model (UKNSM). Ten subject-specific models were developed through manual digitisation of model parameters from high-resolution MRI. Kinematic data of 16 functional daily activities were collected using a 10-camera optical motion capture system. Subject-specific model predictions were validated with measured muscle activations. The MRI-based shoulder models show good agreement with measured muscle activations. A tenfold cross-validation using the validated personalised shoulder models demonstrates that linear scaling of anthropometric datasets with the most similar ratio of body height to shoulder width and from the same gender (p < 0.04) yields best modelling outcomes in glenohumeral loading. The improvement in model reliability is significant (p < 0.02) when compared to the linearly scaled-generic UKNSM. This study may facilitate the clinical application of musculoskeletal shoulder modelling to aid surgical decision-making.
  • Journal article
    Finnegan ME, Visanji NP, Romero-Canelon I, House E, Rajan S, Mosselmans JFW, Hazrati LN, Dobson J, Collingwood JFet al., 2019,

    Synchrotron XRF imaging of Alzheimer's disease basal ganglia reveals linear dependence of high-field magnetic resonance microscopy on tissue iron concentration

    , Journal of Neuroscience Methods, ISSN: 0165-0270

    Background: Chemical imaging of the human brain has great potential for diagnostic and monitoring purposes. The heterogeneity of human brain iron distribution, and alterations to this distribution in Alzheimer's disease, indicate iron as a potential endogenous marker. The influence of iron on certain magnetic resonance imaging (MRI) parameters increases with magnetic field, but is under-explored in human brain tissues above 7 T. New Method: Magnetic resonance microscopy at 9.4 T is used to calculate parametric images of chemically-unfixed post-mortem tissue from Alzheimer's cases (n = 3) and healthy controls (n = 2). Iron-rich regions including caudate nucleus, putamen, globus pallidus and substantia nigra are analysed prior to imaging of total iron distribution with synchrotron X-ray fluorescence mapping. Iron fluorescence calibration is achieved with adjacent tissue blocks, analysed by inductively coupled plasma mass spectrometry or graphite furnace atomic absorption spectroscopy. Results: Correlated MR images and fluorescence maps indicate linear dependence of R 2 , R 2 * and R 2 ’ on iron at 9.4 T, for both disease and control, as follows: [R 2 (s −1 ) = 0.072[Fe] + 20]; [R 2 *(s −1 ) = 0.34[Fe] + 37]; [R 2 ’(s −1 ) = 0.26[Fe] + 16] for Fe in μg/g tissue (wet weight). Comparison with Existing Methods: This method permits simultaneous non-destructive imaging of most bioavailable elements. Iron is the focus of the present study as it offers strong scope for clinical evaluation; the approach may be used more widely to evaluate the impact of chemical elements on clinical imaging parameters. Conclusion: The results at 9.4 T are in excellent quantitative agreement with predictions from experiments performed at lower magnetic fields.

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: http://wlsprd.imperial.ac.uk:80/respub/WEB-INF/jsp/search-t4-html.jsp Request URI: /respub/WEB-INF/jsp/search-t4-html.jsp Query String: id=1107&limit=10&respub-action=search.html Current Millis: 1573951404130 Current Time: Sun Nov 17 00:43:24 GMT 2019