Display omitted
•Multi-center, multi-vendor, multi-protocol prostate MRI dataset was made available for evaluation of segmentation algorithms.•Evaluated 11 substantially different segmentation ...algorithms with respect to algorithm performance on multi-center data.•Algorithms were evaluated relative to human observers.•Challenge results show that segmentation of prostate MRI images is not a solved issue.
Prostate MRI image segmentation has been an area of intense research due to the increased use of MRI as a modality for the clinical workup of prostate cancer. Segmentation is useful for various tasks, e.g. to accurately localize prostate boundaries for radiotherapy or to initialize multi-modal registration algorithms. In the past, it has been difficult for research groups to evaluate prostate segmentation algorithms on multi-center, multi-vendor and multi-protocol data. Especially because we are dealing with MR images, image appearance, resolution and the presence of artifacts are affected by differences in scanners and/or protocols, which in turn can have a large influence on algorithm accuracy. The Prostate MR Image Segmentation (PROMISE12) challenge was setup to allow a fair and meaningful comparison of segmentation methods on the basis of performance and robustness. In this work we will discuss the initial results of the online PROMISE12 challenge, and the results obtained in the live challenge workshop hosted by the MICCAI2012 conference. In the challenge, 100 prostate MR cases from 4 different centers were included, with differences in scanner manufacturer, field strength and protocol. A total of 11 teams from academic research groups and industry participated. Algorithms showed a wide variety in methods and implementation, including active appearance models, atlas registration and level sets. Evaluation was performed using boundary and volume based metrics which were combined into a single score relating the metrics to human expert performance. The winners of the challenge where the algorithms by teams Imorphics and ScrAutoProstate, with scores of 85.72 and 84.29 overall. Both algorithms where significantly better than all other algorithms in the challenge (p<0.05) and had an efficient implementation with a run time of 8min and 3s per case respectively. Overall, active appearance model based approaches seemed to outperform other approaches like multi-atlas registration, both on accuracy and computation time. Although average algorithm performance was good to excellent and the Imorphics algorithm outperformed the second observer on average, we showed that algorithm combination might lead to further improvement, indicating that optimal performance for prostate segmentation is not yet obtained. All results are available online at http://promise12.grand-challenge.org/.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK
The objective of this study was to evaluate early changes in magnetic resonance imaging (MRI) and clinical disease activity measures as predictors of later structural progression in early rheumatoid ...arthritis (RA).
This was a post hoc analysis of data pooled across treatments from a three-arm (tofacitinib monotherapy, tofacitinib with methotrexate MTX, or MTX monotherapy) trial of MTX-naïve patients with early, active RA. Synovitis, osteitis and erosions were assessed with the Outcome Measures in Rheumatology (OMERACT) RA MRI scoring system (RAMRIS) and RAMRIQ (automated quantitative RA MRI assessment system; automated RAMRIS) at months 0, 1, 3, 6 and 12. Radiographs were assessed at months 0, 6 and 12, and clinical endpoints were assessed at all timepoints. Univariate and multivariate analyses explored the predictive value of early changes in RAMRIS/RAMRIQ parameters and disease activity measures, with respect to subsequent radiographic progression.
Data from 109 patients with a mean RA duration of 0.7 years were included. In univariate analyses, changes in RAMRIS erosions at months 1 and 3 significantly predicted radiographic progression at month 12 (both p < 0.01); changes in RAMRIQ synovitis and osteitis at months 1 and 3 were significant predictors of RAMRIS erosions and radiographic progression at month 12 (all p < 0.01). In subsequent multivariate analyses, RAMRIS erosion change at month 1 (p < 0.05) and RAMRIQ osteitis changes at months 1 and 3 (both p < 0.01) were significant independent predictors of radiographic progression at month 12. Univariate analyses demonstrated that changes in Clinical Disease Activity Index (CDAI) and Disease Activity Score in 28 joints, erythrocyte sedimentation rate (DAS28-4ESR) at months 1 and 3 were not predictive of month 12 radiographic progression.
MRI changes seen as early as 1 month after RA treatment initiation have the potential to better predict long-term radiographic progression than changes in disease activity measures.
ClinicalTrials.gov, NCT01164579 .
MIV-711 is a novel selective cathepsin K inhibitor with beneficial effects on bone and cartilage in preclinical osteoarthritis models.
To evaluate the efficacy, safety, and tolerability of MIV-711 in ...participants with symptomatic, radiographic knee osteoarthritis.
26-week randomized, double-blind, placebo-controlled phase 2a study with a 26-week open-label safety extension substudy. (EudraCT: 2015-003230-26 and 2016-001096-73).
Six European sites.
244 participants with primary knee osteoarthritis, Kellgren-Lawrence grade 2 or 3, and pain score of 4 to 10 on a numerical rating scale (NRS).
MIV-711, 100 (n = 82) or 200 (n = 81) mg daily, or matched placebo (n = 77). Participants (46 who initially received 200 mg/d and 4 who received placebo) received 200 mg of MIV-711 daily during the extension substudy.
The primary outcome was change in NRS pain score. The key secondary outcome was change in bone area on magnetic resonance imaging (MRI). Other secondary end points included cartilage thickness on quantitative MRI and type I and II collagen C-telopeptide biomarkers. Outcomes were assessed over 26 weeks.
Changes in NRS pain scores with MIV-711 were not statistically significant (placebo, -1.4; MIV-711, 100 mg/d, -1.7; MIV-711, 200 mg/d, -1.5). MIV-711 significantly reduced medial femoral bone area progression (P = 0.002 for 100 mg/d and 0.004 for 200 mg/d) and medial femoral cartilage thinning (P = 0.023 for 100 mg/d and 0.125 for 200 mg/d) versus placebo and substantially reduced bone and cartilage biomarker levels. Nine serious adverse events occurred in 6 participants (1 in the placebo group, 3 in the 100 mg group, and 2 in the 200 mg group); none were considered to be treatment-related.
The trial was relatively short.
MIV-711 was not more effective than placebo for pain, but it significantly reduced bone and cartilage progression with a reassuring safety profile. This treatment may merit further evaluation as a disease-modifying osteoarthritis drug.
Medivir.
Accurate automated segmentation of cartilage should provide rapid reliable outcomes for both epidemiological studies and clinical trials. We aimed to assess the precision and responsiveness of ...cartilage thickness measured with careful manual segmentation or a novel automated technique.
Agreement of automated segmentation was assessed against 2 manual segmentation datasets: 379 magnetic resonance images manually segmented in-house (training set), and 582 from the Osteoarthritis Initiative with data available at 0, 1, and 2 years (biomarkers set). Agreement of mean thickness was assessed using Bland-Altman plots, and change with pairwise Student t test in the central medial femur (cMF) and tibia regions (cMT). Repeatability was assessed on a set of 19 knees imaged twice on the same day. Responsiveness was assessed using standardized response means (SRM).
Agreement of manual versus automated methods was excellent with no meaningful systematic bias (training set: cMF bias 0.1 mm, 95% CI ± 0.35; biomarkers set: bias 0.1 mm ± 0.4). The smallest detectable difference for cMF was 0.13 mm (coefficient of variation 3.1%), and for cMT 0.16 mm(2.65%). Reported change using manual segmentations in the cMF region at 1 year was -0.031 mm (95% CI -0.022, -0.039), p < 10
, SRM -0.31 (-0.23, -0.38); and at 2 years was -0.071 (-0.058, -0.085), p < 10
, SRM -0.43 (-0.36, -0.49). Reported change using automated segmentations in the cMF at 1 year was -0.059 (-0.047, -0.071), p < 10
, SRM -0.41 (-0.34, -0.48); and at 2 years was -0.14 (-0.123, -0.157, p < 10
, SRM -0.67 (-0.6, -0.72).
A novel cartilage segmentation method provides highly accurate and repeatable measures with cartilage thickness measurements comparable to those of careful manual segmentation, but with improved responsiveness.
Purpose
Automated delineation of structures and organs is a key step in medical imaging. However, due to the large number and diversity of structures and the large variety of segmentation algorithms, ...a consensus is lacking as to which automated segmentation method works best for certain applications. Segmentation challenges are a good approach for unbiased evaluation and comparison of segmentation algorithms.
Methods
In this work, we describe and present the results of the Head and Neck Auto‐Segmentation Challenge 2015, a satellite event at the Medical Image Computing and Computer Assisted Interventions (MICCAI) 2015 conference. Six teams participated in a challenge to segment nine structures in the head and neck region of CT images: brainstem, mandible, chiasm, bilateral optic nerves, bilateral parotid glands, and bilateral submandibular glands.
Results
This paper presents the quantitative results of this challenge using multiple established error metrics and a well‐defined ranking system. The strengths and weaknesses of the different auto‐segmentation approaches are analyzed and discussed.
Conclusions
The Head and Neck Auto‐Segmentation Challenge 2015 was a good opportunity to assess the current state‐of‐the‐art in segmentation of organs at risk for radiotherapy treatment. Participating teams had the possibility to compare their approaches to other methods under unbiased and standardized circumstances. The results demonstrate a clear tendency toward more general purpose and fewer structure‐specific segmentation algorithms.
Full text
Available for:
BFBNIB, FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
Objectives
To determine the repeatability and response to therapy of dynamic contrast-enhanced (DCE) MRI biomarkers of synovitis in the hand and wrist of rheumatoid arthritis (RA) patients, and in ...particular the performance of the transfer constant
K
trans
, in a multicentre trial setting.
Methods
DCE-MRI and RA MRI scoring (RAMRIS) were performed with meticulous standardisation at baseline and 6 and 24 weeks in a substudy of fostamatinib monotherapy in reducing synovitis compared with placebo or adalimumab. Analysis employed statistical shape modelling to avoid biased regions-of-interest, kinetic modelling and heuristic analyses. Repeatability was also evaluated.
Results
At early study termination, DCE-MRI data had been acquired from 58 patients in 19 imaging centres.
K
trans
intra-subject coefficient of variation (N = 14) was 30%.
K
trans
change demonstrated inferiority of fostamatinib (N = 11) relative to adalimumab (N = 10) after 6 weeks (treatment ratio = 1.92, p = 0.003), and failed to distinguish fostamatinib from placebo (N = 10, p = 0.79). RAMRIS showed superiority of fostamatinib relative to placebo at 6 weeks (p = 0.023), and did not distinguish fostamatinib from adalimumab at either 6 (p = 0.175) or 24 (p = 0.230) weeks.
Conclusion
This demonstrated repeatability of
K
trans
and its ability to distinguish treatment groups show that DCE-MRI biomarkers are suitable for use in multicentre RA trials.
Key Points
• DCE-MRI biomarkers are feasible in large multicentre studies of joint inflammation.
• DCE-MRI K
trans
showed fostamatinib inferior to adalimumab after 6 weeks.
• K
trans
repeatability coefficient of variation was 30
%
multicentre.
Structured Case Base Knowledge using Unsupervised Learning Chourib, Ikram; Guillard, Gwenael; Farah, Imed Riadh ...
2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP),
2022-May-24
Conference Proceeding
In medicine, medical knowledge representation and reasoning processes are the main issues of Medical Decision Support Systems, MDSS. This paper proposes an approach allowing to design a new ...representation of medical knowledge model using the Case-Based Reasoning approach and the Unsupervised Learning (K-Means clustering) method. The proposed approach is evaluated using the Stroke Knowledge Base. Obtained results indicate that applying the clustering approach on the case base and, then, using the K-Nearest Neighbor for each cluster is more efficient than the only use of the K-Nearest Neighbor without applying the clustering on the case base. The provided representation and the achieved results constitute an appropriate approach in order to reach several purposes of MDSS in terms of clustering, similarity estimation and quality assessment.
This article attempts to provide an overview of the basic concepts, structure and terms related to case-based reasoning, because we aim to work on an evolution of the CBR process which allows us to ...integrate the knowledge base, ensure the consistency of information, integrate the modes of reasoning: similarity, classification and hybrid. In addition, this article discusses the concept and issues related to similarity measure, which is very important in medical system and using it allows to reduce the risks of health error for patients. However, it seems very complex because the measure of similarity must take into account the diversity of the types of information (qualitative, textual, quantitative, etc.) for each patient. And at the end, we represent an evaluation of similarity measure which was applied in (GS.52).
The close inter-relationship between shape and functionality of osteo-articular complexes is fundamental for understanding, modeling, and imaging. It is thus desirable to put forward theoretical and ...methodological principles that take benefit of this relationship. In this paper, we propose a new representation of spherical motions that is applied to ball-and-socket joint such as hip and shoulder. We have been developing Spherical Medial Representation (SM-Rep) based on skeletonisation over the sphere. While M-Rep provides a paradigm for 3D surface and image analysis, the resulting surface remains in a 3D vector space. Since the geometry of the unit sphere is quite different, especially geodesies, SM-Rep supplies consistent elements to overcome this issue. Example of application of this framework to hip is given.