Development of Ground Truth Data for Automatic Lumbar Spine MRI Image Segmentation

Natalia, Friska; Meidia, Hira; Afriliana, Nunik; Kafri, Ala S. Al; Sudirman, Sud; Simpson, Andrew; Sophian, Ali; Al-Jumaily, Mohammed; Al-Rashdan, Wasfi; Bashtawi, Mohammad

doi:10.1109/hpcc/smartcity/dss.2018.00239

Cited by 16 publications

(10 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results are consistent with the previous experiment we reported in [21]. The IVD has the highest confidence and consistency values compared to the other three.…”

Section: Ground Truth Data Quality Metricssupporting

confidence: 93%

“…On average, five to ten minutes are spent to label each of the 1,545 slices. We employ the strategy we have designed previously in [21] to develop our ground truth dataset using five participant/labelers. We analyze the quality of the developed ground-truth dataset using the confidence and consistency metrics that we presented in that paper.…”

Section: B Image Labels and Ground-truth Datamentioning

confidence: 99%

See 1 more Smart Citation

Boundary Delineation of MRI Images for Lumbar Spinal Stenosis Detection Through Semantic Segmentation Using Deep Neural Networks

et al. 2019

View full text Add to dashboard Cite

We propose a methodology to aid clinicians in performing lumbar spinal stenosis detection through semantic segmentation and delineation of magnetic resonance imaging (MRI) scans of the lumbar spine using deep learning. Our dataset contains MRI studies of 515 patients with symptomatic back pains. Each study is annotated by expert radiologists with notes regarding the observed characteristics and condition of the lumbar spine. We have developed a ground truth dataset, containing image labels of four important regions in the lumbar spine, to be used as the training and test images to develop classification models for segmentation. We developed two novel metrics, namely confidence, and consistency, to assess the quality of the ground truth dataset through a derivation of the Jaccard Index. We experimented with semantic segmentation of our dataset using SegNet. Our evaluation of the segmentation and the delineation results show that our proposed methodology produces a very good performance as measured by several contourbased and region-based metrics. In addition, using the Cohen's kappa and frequency-weighted confidence metrics, we can show that 1) the model's performance is within the range of the worst and the best manual labeling results and 2) the ground-truth dataset has an excellent inter-rater agreement score. We also presented two representative delineation results of the worst and best segmentation based on their BF-score to show visually how accurate and suitable the results are for computer-aided-diagnosis purposes.

show abstract

“…The results are consistent with the previous experiment we reported in [21]. The IVD has the highest confidence and consistency values compared to the other three.…”

Section: Ground Truth Data Quality Metricssupporting

confidence: 93%

Section: B Image Labels and Ground-truth Datamentioning

confidence: 99%

Boundary Delineation of MRI Images for Lumbar Spinal Stenosis Detection Through Semantic Segmentation Using Deep Neural Networks

et al. 2019

View full text Add to dashboard Cite

show abstract

“…In our work using the dataset [3], the root canal space is called AAP (called the canal/space), which represents the area between the anterior and posterior. The width of the AAP depends on where in the spine the measurement is taken.…”

Section: Figure 1 Spinal Stenosismentioning

confidence: 99%

Lumbar Spinal Stenosis Analysis with Deep Learning Based Decision Support Systems

Altun

Alkan

2023

Gazi University Journal of Science

View full text Add to dashboard Cite

Lumbar spinal stenosis is a disease with negative consequences and usually occurs in 3 vertebrae, disc and canal located in the lower back. In this region, the nerves in the canal can be exposed to pressure for various reasons, and diseases occur. Surgical operation is required to treat canal narrowing, and the exact location and size of the spinal stenosis is vital importance for the operation. The UNet model, which is an example of this network, can be further deeper using different deep learning networks. In this study, it is aimed to be the basis for the creation of a system that helps in the diagnosis of canal stenosis by using a deeper network. The ResUNET model, in which ResNet is used as the backbone, achieved an average IoU of 0.987. This result reveals that MR images can be used in segmentation for the diagnosis of Lumbar spinal stenosis.

show abstract

“…This is also reflected in the popularity of image analysis methods in the literature that use sagittal MRI images to detect LSS [6,[13][14][15][16]. However, a more accurate assessment of the actual location and the extent of the LSS can only be obtained through inspection of the suspected IVD in traverse view (as illustrated in Fig 3) [17,18]. Traverse images taken from planes that cut closest to the half-height of the L3/L4, L4/L5, and L5/S1 IVDs are generally considered as the best images to use when the neuroradiologist inspects the disc because The task of selecting these traverse images falls into the category of image classification and can be solved using machine learning (ML).…”

Section: Introductionmentioning

confidence: 99%

Automated selection of mid-height intervertebral disc slice in traverse lumbar spine MRI using a combination of deep learning feature and machine learning classifier

et al. 2022

Self Cite

View full text Add to dashboard Cite

Abnormalities and defects that can cause lumbar spinal stenosis often occur in the Intervertebral Disc (IVD) of the patient’s lumbar spine. Their automatic detection and classification require an application of an image analysis algorithm on suitable images, such as mid-sagittal images or traverse mid-height intervertebral disc slices, as inputs. Hence the process of selecting and separating these images from other medical images in the patient’s set of scans is necessary. However, the technological progress in making this process automated is still lagging behind other areas in medical image classification research. In this paper, we report the result of our investigation on the suitability and performance of different approaches of machine learning to automatically select the best traverse plane that cuts closest to the half-height of an IVD from a database of lumbar spine MRI images. This study considers images features extracted using eleven different pre-trained Deep Convolution Neural Network (DCNN) models. We investigate the effectiveness of three dimensionality-reduction techniques and three feature-selection techniques on the classification performance. We also investigate the performance of five different Machine Learning (ML) algorithms and three Fully Connected (FC) neural network learning optimizers which are used to train an image classifier with hyperparameter optimization using a wide range of hyperparameter options and values. The different combinations of methods are tested on a publicly available lumbar spine MRI dataset consisting of MRI studies of 515 patients with symptomatic back pain. Our experiment shows that applying the Support Vector Machine algorithm with a short Gaussian kernel on full-length image features extracted using a pre-trained DenseNet201 model is the best approach to use. This approach gives the minimum per-class classification performance of around 0.88 when measured using the precision and recall metrics. The median performance measured using the precision metric ranges from 0.95 to 0.99 whereas that using the recall metric ranges from 0.93 to 1.0. When only considering the L3/L4, L4/L5, and L5/S1 classes, the minimum F1-Scores range between 0.93 to 0.95, whereas the median F1-Scores range between 0.97 to 0.99.

show abstract

Development of Ground Truth Data for Automatic Lumbar Spine MRI Image Segmentation

Cited by 16 publications

References 10 publications

Boundary Delineation of MRI Images for Lumbar Spinal Stenosis Detection Through Semantic Segmentation Using Deep Neural Networks

Boundary Delineation of MRI Images for Lumbar Spinal Stenosis Detection Through Semantic Segmentation Using Deep Neural Networks

Lumbar Spinal Stenosis Analysis with Deep Learning Based Decision Support Systems

Automated selection of mid-height intervertebral disc slice in traverse lumbar spine MRI using a combination of deep learning feature and machine learning classifier

Contact Info

Product

Resources

About