1
|
Kuiper RJA, Colaris JW, Stockmans F, van Es EM, Viergever MA, Seevinck PR, Weinans H, Sakkers RJB. Impact of bone and cartilage segmentation from CT and MRI on both bone forearm osteotomy planning. Int J Comput Assist Radiol Surg 2023; 18:2307-2318. [PMID: 37219804 PMCID: PMC10632286 DOI: 10.1007/s11548-023-02929-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/03/2022] [Accepted: 04/19/2023] [Indexed: 05/24/2023]
Abstract
INTRODUCTION The use of MRI scans for pre-operative surgical planning of forearm osteotomies provides additional information of joint cartilage and soft tissue structures and reduces radiation exposure in comparison with the use of CT scans. In this study, we investigated whether using 3D information obtained from MRI with and without cartilage information leads to a different outcome of pre-operative planning. METHODS Bilateral CT and MRI scans of the forearms of 10 adolescent and young adult patients with a unilateral bone deformation were acquired in a prospective study. The bones were segmented from CT and MRI, and cartilage only from MRI. The deformed bones were virtually reconstructed, by registering the joint ends to the healthy contralateral side. An optimal osteotomy plane was determined that minimized the distance between the resulting fragments. This process was performed in threefold: using the CT and MRI bone segmentations, and the MRI cartilage segmentations. RESULTS Comparison of bone segmentation from MRI and CT scan resulted in a 0.95 ± 0.02 Dice Similarity Coefficient and 0.42 ± 0.07 mm Mean Absolute Surface Distance. All realignment parameters showed excellent reliability across the different segmentations. However, the mean differences in translational realignment between CT and MRI bone segmentations (4.5 ± 2.1 mm) and between MRI bone and MRI bone and cartilage segmentations (2.8 ± 2.1 mm) were shown to be clinically and statistically significant. A significant positive correlation was found between the translational realignment and the relative amount of cartilage. CONCLUSION This study indicates that although bone realignment remained largely similar when using MRI with and without cartilage information compared to using CT, the small differences in segmentation could induce statistically and clinically significant differences in the osteotomy planning. We also showed that endochondral cartilage might be a non-negligible factor when planning osteotomies for young patients.
Collapse
Affiliation(s)
- Ruurd J A Kuiper
- Department of Orthopaedics, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands.
- Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands.
| | - Joost W Colaris
- Department of Orthopaedics and Sports Medicine, Erasmus Medical Centre, Rotterdam, The Netherlands
| | - Filip Stockmans
- Muscles & Movement, Department of Development and Regeneration, KU Leuven Campus Kulak, Kortrijk, Belgium
| | - Eline M van Es
- Department of Orthopaedics and Sports Medicine, Erasmus Medical Centre, Rotterdam, The Netherlands
| | - Max A Viergever
- Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands
| | - Peter R Seevinck
- Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands
| | - Harrie Weinans
- Department of Orthopaedics, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands
| | - Ralph J B Sakkers
- Department of Orthopaedics, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands
| |
Collapse
|
2
|
Vladimirov N, Brui E, Levchuk A, Al-Haidri W, Fokin V, Efimtcev A, Bendahan D. CNN-based fully automatic wrist cartilage volume quantification in MR images: A comparative analysis between different CNN architectures. Magn Reson Med 2023; 90:737-751. [PMID: 37094028 DOI: 10.1002/mrm.29671] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/24/2022] [Revised: 03/17/2023] [Accepted: 03/26/2023] [Indexed: 04/26/2023]
Abstract
PURPOSE Automatic measurement of wrist cartilage volume in MR images. METHODS We assessed the performance of four manually optimized variants of the U-Net architecture, nnU-Net and Mask R-CNN frameworks for the segmentation of wrist cartilage. The results were compared to those from a patch-based convolutional neural network (CNN) we previously designed. The segmentation quality was assessed on the basis of a comparative analysis with manual segmentation. The best networks were compared using a cross-validation approach on a dataset of 33 3D VIBE images of mostly healthy volunteers. Influence of some image parameters on the segmentation reproducibility was assessed. RESULTS The U-Net-based networks outperformed the patch-based CNN in terms of segmentation homogeneity and quality, while Mask R-CNN did not show an acceptable performance. The median 3D DSC value computed with the U-Net_AL (0.817) was significantly larger than DSC values computed with the other networks. In addition, the U-Net_AL provided the lowest mean volume error (17%) and the highest Pearson correlation coefficient (0.765) with respect to the ground truth values. Of interest, the reproducibility computed using U-Net_AL was larger than the reproducibility of the manual segmentation. Moreover, the results indicate that the MRI-based wrist cartilage volume is strongly affected by the image resolution. CONCLUSIONS U-Net CNN with attention layers provided the best wrist cartilage segmentation performance. In order to be used in clinical conditions, the trained network can be fine-tuned on a dataset representing a group of specific patients. The error of cartilage volume measurement should be assessed independently using a non-MRI method.
Collapse
Affiliation(s)
- Nikita Vladimirov
- School of Physics and Engineering, ITMO University, Saint-Petersburg, Russia
| | - Ekaterina Brui
- School of Physics and Engineering, ITMO University, Saint-Petersburg, Russia
| | - Anatoliy Levchuk
- School of Physics and Engineering, ITMO University, Saint-Petersburg, Russia
- Department of Radiology, Federal Almazov North-West Medical Research Center, Saint-Petersburg, Russia
| | - Walid Al-Haidri
- School of Physics and Engineering, ITMO University, Saint-Petersburg, Russia
| | - Vladimir Fokin
- School of Physics and Engineering, ITMO University, Saint-Petersburg, Russia
- Department of Radiology, Federal Almazov North-West Medical Research Center, Saint-Petersburg, Russia
| | - Aleksandr Efimtcev
- School of Physics and Engineering, ITMO University, Saint-Petersburg, Russia
- Department of Radiology, Federal Almazov North-West Medical Research Center, Saint-Petersburg, Russia
| | - David Bendahan
- Centre de Résonance Magnétique Biologique et Médicale, Aix-Marseille Universite, CNRS, Marseille, France
| |
Collapse
|
3
|
Dalili D, Fritz J, Isaac A. 3D MRI of the Hand and Wrist: Technical Considerations and Clinical Applications. Semin Musculoskelet Radiol 2021; 25:501-513. [PMID: 34547815 DOI: 10.1055/s-0041-1731652] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/20/2022]
Abstract
In the last few years, major developments have been observed in the field of magnetic resonance imaging (MRI). Advances in both scanner hardware and software technologies have witnessed great leaps, enhancing the diagnostic quality and, therefore, the value of MRI. In musculoskeletal radiology, three-dimensional (3D) MRI has become an integral component of the diagnostic pathway at our institutions. This technique is particularly relevant in patients with hand and wrist symptoms, due to the intricate nature of the anatomical structures and the wide range of differential diagnoses for most presentations. We review the benefits of 3D MRI of the hand and wrist, commonly used pulse sequences, clinical applications, limitations, and future directions. We offer guidance for enhancing the image quality and tips for image interpretation of 3D MRI of the hand and wrist.
Collapse
Affiliation(s)
- Danoob Dalili
- Epsom and St Helier University Hospitals, London, United Kingdom
| | - Jan Fritz
- NYU Grossman School of Medicine, New York University, New York, New York
| | - Amanda Isaac
- Guy's and St. Thomas' Hospitals NHS Foundation Trust, London, United Kingdom.,School of Biomedical Engineering and Imaging Sciences, King's College London (KCL), London, United Kingdom
| |
Collapse
|
4
|
Brui E, Efimtcev AY, Fokin VA, Fernandez R, Levchuk AG, Ogier AC, Samsonov AA, Mattei JP, Melchakova IV, Bendahan D, Andreychenko A. Deep learning-based fully automatic segmentation of wrist cartilage in MR images. NMR IN BIOMEDICINE 2020; 33:e4320. [PMID: 32394453 PMCID: PMC7784718 DOI: 10.1002/nbm.4320] [Citation(s) in RCA: 13] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/25/2018] [Revised: 04/10/2020] [Accepted: 04/14/2020] [Indexed: 05/10/2023]
Abstract
The study objective was to investigate the performance of a dedicated convolutional neural network (CNN) optimized for wrist cartilage segmentation from 2D MR images. CNN utilized a planar architecture and patch-based (PB) training approach that ensured optimal performance in the presence of a limited amount of training data. The CNN was trained and validated in 20 multi-slice MRI datasets acquired with two different coils in 11 subjects (healthy volunteers and patients). The validation included a comparison with the alternative state-of-the-art CNN methods for the segmentation of joints from MR images and the ground-truth manual segmentation. When trained on the limited training data, the CNN outperformed significantly image-based and PB-U-Net networks. Our PB-CNN also demonstrated a good agreement with manual segmentation (Sørensen-Dice similarity coefficient [DSC] = 0.81) in the representative (central coronal) slices with a large amount of cartilage tissue. Reduced performance of the network for slices with a very limited amount of cartilage tissue suggests the need for fully 3D convolutional networks to provide uniform performance across the joint. The study also assessed inter- and intra-observer variability of the manual wrist cartilage segmentation (DSC = 0.78-0.88 and 0.9, respectively). The proposed deep learning-based segmentation of the wrist cartilage from MRI could facilitate research of novel imaging markers of wrist osteoarthritis to characterize its progression and response to therapy.
Collapse
Affiliation(s)
- Ekaterina Brui
- University of Information Technology Mechanics and Optics, International Research Center Nanophotonics and Metamaterials, 199034 S.-Petersburg, Russia
| | - Aleksandr Y. Efimtcev
- University of Information Technology Mechanics and Optics, International Research Center Nanophotonics and Metamaterials, 199034 S.-Petersburg, Russia
- Federal Almazov North-West Medical Research Center, 197341 S.-Petersburg, Russia
| | - Vladimir A. Fokin
- University of Information Technology Mechanics and Optics, International Research Center Nanophotonics and Metamaterials, 199034 S.-Petersburg, Russia
- Federal Almazov North-West Medical Research Center, 197341 S.-Petersburg, Russia
| | - Remi Fernandez
- APHM, Service de Radiologie, Hôpital de la Conception, Marseille, France
| | - Anatoliy G. Levchuk
- Federal Almazov North-West Medical Research Center, 197341 S.-Petersburg, Russia
| | - Augustin C. Ogier
- Aix-Marseille Universite, CNRS, Centre de Résonance Magnétique Biologique et Médicale, UMR 7339, Marseille, France
| | - Alexey A. Samsonov
- University of Wisconsin-Madison, Department of Radiology, Madison, WI 53705-2275 USA
| | - Jean. P. Mattei
- Aix-Marseille Universite, CNRS, Centre de Résonance Magnétique Biologique et Médicale, UMR 7339, Marseille, France
- Assistance Publique Hôpitaux de Marseille, Institut de l’appareil locomoteur, Service de Rhumatologie, Hôpital Sainte Marguerite, Marseille, France
| | - Irina V. Melchakova
- University of Information Technology Mechanics and Optics, International Research Center Nanophotonics and Metamaterials, 199034 S.-Petersburg, Russia
| | - David Bendahan
- Aix-Marseille Universite, CNRS, Centre de Résonance Magnétique Biologique et Médicale, UMR 7339, Marseille, France
| | - Anna Andreychenko
- University of Information Technology Mechanics and Optics, International Research Center Nanophotonics and Metamaterials, 199034 S.-Petersburg, Russia
- Research and Practical Clinical Center of Diagnostics and Telemedicine Technologies, Department of Health Care of Moscow, Moscow, Russia
| |
Collapse
|
5
|
Sivak WN, Imbriglia JE. Evaluation of Cartilage in the Wrist using Magnetic Resonance Imaging. Curr Rheumatol Rev 2019; 16:170-177. [PMID: 31804162 DOI: 10.2174/1573397115666190819153912] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/12/2019] [Revised: 06/18/2019] [Accepted: 08/20/2019] [Indexed: 11/22/2022]
Abstract
Wrist pain is a common patient complaint with a myriad of clinical conditions that can explain the underlying cause. Short of wrist arthroscopy, no technique other than formal wrist arthrotomy exists for direct examination of the hyaline cartilage coating the articular surfaces of the carpal bones. Magnetic resonance imaging (MRI) has been proven accurate in evaluating joint surfaces of large joints such as the shoulder, hip, and knee with articular cartilage surface thickness is in excess of 1 mm. However, in the carpus the thickness of the cartilage and the contours present have precluded accurate imaging. Advances in MRI technology over the last several decades are now making imaging of small joint surfaces, such as the carpus, an area worth revisiting. Herein we provide a review of these efforts with a specific focus on the evaluation of the wrist.
Collapse
Affiliation(s)
- Wesley N Sivak
- Department of Plastic Surgery, University of Pittsburgh, Pittsburgh, PA, United States
| | - Joseph E Imbriglia
- Department of Orthopaedic Surgery, University of Pittsburgh, Pittsburgh, PA, United States
| |
Collapse
|
6
|
THE ROLE OF MAGNETIC RESONANCE IMAGING IN THE DIAGNOSIS OF DEFORMING ARTHROSIS OF PROFESSIONAL ETIOLOGY IN MINERS. EUREKA: HEALTH SCIENCES 2018. [DOI: 10.21303/2504-5679.2018.00730] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
The paper analyzes the effectiveness of magnetic resonance imaging with cartilage diagram in diagnosing signs of professional deforming arthrosis of knee joints in miners working in conditions of significant physical loading.
Aim of the research – to determine of diagnostic efficiency of indicators of magnetic resonance imaging of the knee joint and cartilage diagram in miners of the main occupations suffering from deforming arthrosis.
Methods. The research is conducted in 30 miners of basic occupations: 20 mining workers of breakage face (MWBF) and 10 machinists of shearer mining machines (МSMM) have been treated in the inpatient department of occupational pathology of the Lviv Regional Clinical Hospital in 2015-2017 due to deforming arthrosis. Damages of the main anatomical elements of the knee joint with arthrosis were analyzed, visualized initially with the help of MRI, and then - cartilage diagram.
Results. According to the MRI data, in miners of the main occupations with arthrosis of the knee joint the posterior cross-shaped ligament are most commonly affected (in 75.0±9.7 % MWBF and 70.0±14.5 % МSMM), damage to the medial collateral ligament are diagnosed less frequently (in 5.0±4.9 % in the MWBF and in 10.0±9.5 % in the МSMM). On average 3.8±0.4 modified elements of the knee joint are visualized in patients, whereas 4.8±0.1 affected areas are visualized on the cartilage diagram (р<0.05). In 86.7±6.2 % patients, in the analysis of cartilage diagram, changes in all five analyzed areas are diagnosed, indicating a higher efficiency of the diagnosis of changes in the structures of the joint with DA of the professional etiology of the method of cartilage diagram compared with MRI. According to the cartilage diagram the most significant changes are noted in the hypertrophy of the femur: among all miners 62.5±0.3 ms (medial) and 62.6±0.4 ms (lateral), in the MWBF group the average time of Т2-delay is the largest in the area of the medial hypertrophy of the femur is 60.9±2.3 ms, in the МSMM group – in the area of the lateral hypertrophy of the femur: 66.7±3.3 ms, which can be linked to the peculiarities of the forced working position of miners of these professions and the kinetics of joint structures.
These results can be used to diagnose the initial lesions of joint structures with DA of professional genesis, as well as the creation of prognostic models for determining the the degree of risk of development of knee joint damage, which will allow to improve the system of personified approach to diagnostic and preventive measures in working persons in conditions of considerable physical activity and forced working position.
Collapse
|
7
|
Lee RKL, Griffith JF, Tang WK, Ng AWH, Yeung DKW. Effect of traction on wrist joint space and cartilage visibility with and without MR arthrography. Br J Radiol 2017; 90:20160932. [PMID: 28181830 DOI: 10.1259/bjr.20160932] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/02/2023] Open
Abstract
OBJECTIVE To compare the effect of traction during non-arthrographic and arthrographic MR examination of the wrist with regard to joint space width, joint fluid dispersion and cartilage surface visibility. METHODS Prospective 3-T MRI study of 100 wrists in 96 patients. The first 50 wrists underwent MR arthrography first without traction and then with traction. The following 50 wrists underwent standard MR first without traction and then with traction. On these examinations, two radiologists independently measured (i) joint space width, semi-quantitatively graded (ii) joint fluid dispersion between opposing cartilage surfaces and (iii) articular cartilage surface visibility. The three parameters were compared between the two groups. RESULTS Traction led to an increase in joint space width at nearly all joints in all patients (p < 0.05), although more so in the arthrography (∆ = 0.08-0.79 mm, all p < 0.05) than in the non-arthrography (∆ = 0.001-0.61 mm, all p < 0.05) group. Joint fluid dispersion and cartilage surface visibility improved after traction in nearly all joints (p < 0.05) in all patients and more so in the arthographic than in the non-arthrography group. CONCLUSION Traction did significantly improve cartilage surface visibility for standard MRI of the wrist although the effect was not as great as that seen with MR arthography or MR arthrography with traction. Advances in knowledge: This is the first study to show the beneficial effect of traction during standard non-arthrography MRI of the wrist and compare the effect of traction between non-arthrographic and arthrographic MRI of the wrist.
Collapse
Affiliation(s)
- Ryan K L Lee
- Department of Imaging and Interventional Radiology, Prince of Wales Hospital, The Chinese University of Hong Kong
| | - James F Griffith
- Department of Imaging and Interventional Radiology, Prince of Wales Hospital, The Chinese University of Hong Kong
| | - W K Tang
- Department of Imaging and Interventional Radiology, Prince of Wales Hospital, The Chinese University of Hong Kong
| | - Alex W H Ng
- Department of Imaging and Interventional Radiology, Prince of Wales Hospital, The Chinese University of Hong Kong
| | - David K W Yeung
- Department of Imaging and Interventional Radiology, Prince of Wales Hospital, The Chinese University of Hong Kong
| |
Collapse
|