Segmentation of Femoral Cartilage from Knee Ultrasound Images Using Mask R-CNN

Kompella, Gayatri; Antico, Maria; Sasazawa, Fumio; Jeevakala, S.; Ram, Keerthi; Fontanarosa, Davide; Pandey, Ajay K.; Sivaprakasam, Mohanasankar

doi:10.1109/embc.2019.8857645

Cited by 19 publications

(22 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent osteoarthritis research studies on fully automatic methods are mostly focused on MR images as they have excellent soft tissue contrast and distinct resolution on a knee joint. MRI is also a noninvasive technique that does not require ionizing radiation [ 8 , 10 , 35 – 37 ]. Although ultrasound imaging is a noninvasive, portable option that does not require ionizing radiation, its application is limited, especially on the segmentation tasks, due to the low contrast ratio and presence of speckle noise [ 38 – 40 ].…”

Section: Imaging-based Deep Learningmentioning

confidence: 99%

“…Kompella et al [ 37 ] adopted the state-of-the-art Mask R-CNN (regional convolutional neural network) for automated femoral cartilage (FC) segmentation from ultrasound 2D image scans. The ResNet-50 with the feature pyramid network was chosen as the backbone of the architecture with region proposal network to extract the region of interest (ROI).…”

Section: Application Of 2d Deep Learning In Knee Osteoarthritis Assessmentmentioning

confidence: 99%

See 1 more Smart Citation

Emergence of Deep Learning in Knee Osteoarthritis Diagnosis

Yeoh

Lai

Goh

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Osteoarthritis (OA), especially knee OA, is the most common form of arthritis, causing significant disability in patients worldwide. Manual diagnosis, segmentation, and annotations of knee joints remain as the popular method to diagnose OA in clinical practices, although they are tedious and greatly subject to user variation. Therefore, to overcome the limitations of the commonly used method as above, numerous deep learning approaches, especially the convolutional neural network (CNN), have been developed to improve the clinical workflow efficiency. Medical imaging processes, especially those that produce 3-dimensional (3D) images such as MRI, possess ability to reveal hidden structures in a volumetric view. Acknowledging that changes in a knee joint is a 3D complexity, 3D CNN has been employed to analyse the joint problem for a more accurate diagnosis in the recent years. In this review, we provide a broad overview on the current 2D and 3D CNN approaches in the OA research field. We reviewed 74 studies related to classification and segmentation of knee osteoarthritis from the Web of Science database and discussed the various state-of-the-art deep learning approaches proposed. We highlighted the potential and possibility of 3D CNN in the knee osteoarthritis field. We concluded by discussing the possible challenges faced as well as the potential advancements in adopting 3D CNNs in this field.

show abstract

Section: Imaging-based Deep Learningmentioning

confidence: 99%

Section: Application Of 2d Deep Learning In Knee Osteoarthritis Assessmentmentioning

confidence: 99%

Emergence of Deep Learning in Knee Osteoarthritis Diagnosis

Yeoh

Lai

Goh

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…For the aforementioned reasons, deep learning (DL) algorithms have been recently gaining relevance in aiding in US image analysis Liu, Wang, Yang, Lei, Liu, Li, Ni and Wang (2019); Stoel (2020); van Sloun, Cohen and Eldar (2019) since they are able to directly extract complicated information from input images. Convolutional neural networks (CNNs) have already shown promising results for cartilage segmentation Kompella, Antico, Sasazawa, Jeevakala, Ram, Fontanarosa, Pandey and Sivaprakasam (2019); Antico, Sasazawa, Takeda, Jaiprakash, Wille, Pandey, Crawford, Carneiro and Fontanarosa (2020); Dunnhofer, Antico, Sasazawa, Takeda, Camps, Martinel, Micheloni, Carneiro and Fontanarosa (2020). This suggests that DL could potentially help tackling the challenges of automatic cartilage-thickness estimation.…”

Section: Introductionmentioning

confidence: 99%

A deep-learning framework for metacarpal-head cartilage-thickness estimation in ultrasound rheumatological images

Fiorentino

Cipolletta

Filippucci

et al. 2022

Computers in Biology and Medicine

View full text Add to dashboard Cite

“…This strategy showed good effectiveness and accuracy in tumor identification. Kompella et al achieved better results using a pretrained network with both COCO and ImageNet dataset when segmenting the cartilage and even better accuracy after image preprocessing, 21 indicating the importance of image preprocessing. Cai et al successfully used Mask R-CNN to detect pulmonary nodules with high accuracy, which indicates the potential of 2-stage object detection on medical image classification and detection.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Internal Target Volume (ITV) Prediction Using Cone-Beam CT Images in Lung Stereotactic Body Radiotherapy

Zhang

et al. 2022

Technol Cancer Res Treat

View full text Add to dashboard Cite

Purpose:This study aims to develop a deep learning (DL)-based (Mask R-CNN) method to predict the internal target volume (ITV) in cone beam computed tomography (CBCT) images for lung stereotactic body radiotherapy (SBRT) patients and to evaluate the prediction accuracy of the model using 4DCT as ground truth. Methods and Materials: This study enrolled 78 phantom cases and 156 patient cases who received SBRT treatment. We used a novel DL model (Mask R-CNN) to identify and delineate lung tumor ITV in CBCT images. The results of the DL-based method were compared quantitatively with the ground truth (4DCT) using 4 metrics, including Dice Similarity Coefficient (DSC), Relative Volume Index (RVI), 3D Motion Range (R3D), and Hausdorff Surface Distance (HD). Paired t-tests were used to determine the differences between the DL-based method and manual contouring. Results: The DSC value for 4DCTMIP versus CBCT is 0.86 ± 0.16 and for 4DCTAVG versus CBCT is 0.83 ± 0.18, indicating a high similarity of tumor delineation in CBCT and 4DCT. The mean Accuracy Precision (mAP), R3D, RVI, and HD values for phantom evaluation are 0.94 ± 0.04, 1.37 ± 0.36, 0.79 ± 0.02, and 6.79 ± 0.68, respectively. For patient evaluation, the mAP, R3D, RVI, and HD achieved averaged values of 0.74 ± 0.23, 2.39 ± 1.59, 1.27 ± 0.47, and 17.00 ± 19.89, respectively. These results showed a good correlation between predicted ITV and manually contoured ITV. The phantom p-value for RVI, R3D, and HD are 0.75, 0.08, 0.86, and patient p-value are 0.53, 0.07, 0.28, respectively. These p-values for phantom and patient showed no significant difference between the predicted ITV and physician's manual contouring. Conclusion:The current improved method (Mask R-CNN) yielded a good similarity between predicted ITV in CBCT and the manual contouring in 4DCT, thus indicating great potential for using CBCT for patient ITV contouring.

show abstract

Segmentation of Femoral Cartilage from Knee Ultrasound Images Using Mask R-CNN

Cited by 19 publications

References 11 publications

Emergence of Deep Learning in Knee Osteoarthritis Diagnosis

Emergence of Deep Learning in Knee Osteoarthritis Diagnosis

A deep-learning framework for metacarpal-head cartilage-thickness estimation in ultrasound rheumatological images

Deep Learning-Based Internal Target Volume (ITV) Prediction Using Cone-Beam CT Images in Lung Stereotactic Body Radiotherapy

Contact Info

Product

Resources

About