Vertebrae Localization, Segmentation and Identification Using a Graph Optimization and an Anatomic Consistency Cycle

Meng, Di; Mohammed, Eslam; Boyer, Edmond; Pujades, Sergi

doi:10.1007/978-3-031-21014-3_32

Cited by 10 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the selected cases, we automatically segmented the thoracolumbar spine vertebrae using Meng et al 2023 method [22] and the skin with a binary segmentation approach on each slice of the volumetric image. The results were then turned into meshes using the marching-cubes algorithm [2].…”

Section: Nmdid Datasetmentioning

confidence: 99%

See 1 more Smart Citation

3D Inference of the Scoliotic Spine from Depth Maps of the Back

Comte,

Pujades,

Courvoisier

et al. 2024

Lecture Notes in Computational Vision and Biomechanics

View full text Add to dashboard Cite

Recent advances combining outer images and deep-learning algorithms (DLA) show promising results in the detection and the characterization of the Adolescent Idiopathic Scoliosis (AIS). However, these methods are providing a limited 2D characterization while scoliosis is defined in 3D.In this study we propose an inference method that takes as input a depth map of the back of a person and outputs the 3D shape estimation of the thoracolumbar spine. Our DLA method predicts 3D vertebrae positions with an average 3D error of 7.1mm (std: 4.7mm). From the predicted 3D positions, scoliosis can be located and estimated with a mean absolute error (MAE) of 5.5 • (std: 6.2 • ) in the frontal plane. Moreover, sagittal alignments can be estimated with a MAE of 6.4 • (std: 5.5 • ) in kyphosis and 8.3 • (std: 6.8 • ) in lordosis. In addition, our non-ionizing approach can detect scoliosis with an accuracy of 89%.

show abstract

Section: Nmdid Datasetmentioning

confidence: 99%

“…From the 3D vertebra model positions obtained from the EOS Imaging system [7] and the automatic CT segmentation [22], we computed the 3D spine characteristics.…”

Section: Spine 3d Characteristicsmentioning

confidence: 99%

3D Inference of the Scoliotic Spine from Depth Maps of the Back

Comte,

Pujades,

Courvoisier

et al. 2024

Lecture Notes in Computational Vision and Biomechanics

View full text Add to dashboard Cite

show abstract

“…We used an anatomically informed framework trained on the VerSE datasets 46 to segment the different vertebrae on CT imaging 47 . The authors of the framework showed a dice score of 91.04 on a large test set.…”

Section: Epidemological Analysis Spinal Cord Injury Communitymentioning

confidence: 99%

The Implantable System That Restores Hemodynamic Stability After Spinal Cord Injury

Phillips,

Gandhi,

Hankov

et al. 2024

Preprint

View full text Add to dashboard Cite

A spinal cord injury (SCI) causes immediate and sustained hemodynamic instability that threatens neurological recovery and impacts quality of life. Here, we establish the clinical burden of chronic hypotensive complications due to SCI and expose the ineffective treatment of these complications with conservative measures. To address this clinical burden, we developed a purpose-built implantable system based on biomimetic epidural electrical stimulation (EES) of the spinal cord that immediately triggered robust pressor responses. The system durably reduced the severity of hypotensive complications in people with SCI, removed the necessity for conservative treatments, improved quality of life, and enabled superior engagement in activities of daily living. Central to the development of this therapy was the head-to-head demonstration in the same participants that EES must target the last three thoracic segments, and not the lumbosacral segments, to achieve the safe and effective regulation of blood pressure in people with SCI. These findings establish a path forward to treat the underappreciated, treatment-resistant hypotensive complications due to SCI.

show abstract

“…3). Previous works 31,32 also showed the limitation of the single methods and required pre-processing of the medical image, modification of framework, or incorporation of prior knowledge. We noticed that Mask R-CNN may have missed the prediction for those vertebral bodies, which have the incomplete boundary since the first step of object detection fails to identify the bounding box of the incomplete vertebral body, while U-Net is more reliable to find vertebral bodies as it performs semantic segmentation, that is binary classification at pixel level (Fig.…”

Section: Ensemble Rulesmentioning

confidence: 99%

VertXNet: an ensemble method for vertebral body segmentation and identification from cervical and lumbar spinal X-rays

Chen,

Mo,

Readie

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Accurate annotation of vertebral bodies is crucial for automating the analysis of spinal X-ray images. However, manual annotation of these structures is a laborious and costly process due to their complex nature, including small sizes and varying shapes. To address this challenge and expedite the annotation process, we propose an ensemble pipeline called VertXNet. This pipeline currently combines two segmentation mechanisms, semantic segmentation using U-Net, and instance segmentation using Mask R-CNN, to automatically segment and label vertebral bodies in lateral cervical and lumbar spinal X-ray images. VertXNet enhances its effectiveness by adopting a rule-based strategy (termed the ensemble rule) for effectively combining segmentation outcomes from U-Net and Mask R-CNN. It determines vertebral body labels by recognizing specific reference vertebral instances, such as cervical vertebra 2 (‘C2’) in cervical spine X-rays and sacral vertebra 1 (‘S1’) in lumbar spine X-rays. Those references are commonly relatively easy to identify at the edge of the spine. To assess the performance of our proposed pipeline, we conducted evaluations on three spinal X-ray datasets, including two in-house datasets and one publicly available dataset. The ground truth annotations were provided by radiologists for comparison. Our experimental results have shown that the proposed pipeline outperformed two state-of-the-art (SOTA) segmentation models on our test dataset with a mean Dice of 0.90, vs. a mean Dice of 0.73 for Mask R-CNN and 0.72 for U-Net. We also demonstrated that VertXNet is a modular pipeline that enables using other SOTA model, like nnU-Net to further improve its performance. Furthermore, to evaluate the generalization ability of VertXNet on spinal X-rays, we directly tested the pre-trained pipeline on two additional datasets. A consistently strong performance was observed, with mean Dice coefficients of 0.89 and 0.88, respectively. In summary, VertXNet demonstrated significantly improved performance in vertebral body segmentation and labeling for spinal X-ray imaging. Its robustness and generalization were presented through the evaluation of both in-house clinical trial data and publicly available datasets.

show abstract

Vertebrae Localization, Segmentation and Identification Using a Graph Optimization and an Anatomic Consistency Cycle

Cited by 10 publications

References 30 publications

3D Inference of the Scoliotic Spine from Depth Maps of the Back

3D Inference of the Scoliotic Spine from Depth Maps of the Back

The Implantable System That Restores Hemodynamic Stability After Spinal Cord Injury

VertXNet: an ensemble method for vertebral body segmentation and identification from cervical and lumbar spinal X-rays

Contact Info

Product

Resources

About