Automatic bone segmentation in whole-body CT images

Klein, André; Warszawski, Jan; Hillengaß, Jens; Maier‐Hein, Klaus H.

doi:10.1007/s11548-018-1883-7

Cited by 73 publications

(60 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Even this simplification, however, still results in increased planning time, which may make the process unappealing in a busy clinic or for a multicenter clinical trial. Auto-contouring software is developing at a rapid pace, and whole-body bone segmentation has already been implemented [ 30 ]. Popular methods include atlas-based methods [ 31 ] and convolutional neural networks [ 32 ], and both may also aid in the simultaneous contouring of other OARs including the kidneys, lungs, bowels, and others.…”

Section: Discussionmentioning

confidence: 99%

Dosimetric Evaluation Between the Conventional Volumetrically Modulated Arc Therapy (VMAT) Total Body Irradiation (TBI) and the Novel Simultaneous Integrated Total Marrow Approach (SIMBa) VMAT TBI

Stanley¹,

McConnell²,

Iqbal³

et al. 2021

Cureus

View full text Add to dashboard Cite

The purpose of this study was to assess the treatment planning feasibility of volumetrically modulated arc therapy total body irradiation (VMAT TBI) using a simultaneous integrated marrow and body approach (SIMBa). We also aimed to compare SIMBa TBI with the more conventional VMAT TBI approach using the entire body as the target. The goal of using an integrated approach like SIMBa is to balance the known clinical benefit of TBI with the toxicity decrease of Total Marrow Irradiation (TMI) using two prescription volumes. In anticipation of a clinical trial to investigate a novel conditioning regimen that uses SIMBa, our institution retrospectively analyzed the dosimetric differences between 20 clinical VMAT TBI which were replanned using SIMBa. MethodsTwenty patients who previously received conventional VMAT TBI at our institution with a dose of 12 Gy in six fractions were re-planned using SIMBa with a planning aim of delivering a uniform dose of 12 Gy to at least 90% of the PTV_BodyEval. The planning aims of SIMBa were to deliver a uniform dose of 12 Gy to at least 90% of the PTV_Marrow and 8 Gy to at least 90% of the PTV_TotalBody while limiting the mean lung dose to less than 8 Gy. The plans were normalized so that 100% of the PTV_Marrow received at least 90% of the dose with the PTV_TotalBody optimized to stay as close to 100% at 90% as possible. ResultsAll 20 patient plans achieved 12 Gy/8 Gy to at least 90% of the PTV_Marrow and PTV_TotalBody, respectively, with max doses of <16 Gy (130%). As compared with the delivered TBI, the following reductions in mean dose were notable: small bowel 21.3±4.2%, lung 16.3±7.9%, heart 25.3±8.6%, and kidney 16.4±6.2%. Coverage of the sanctuary sites was maintained despite a significant reduction to sensitive organs at risk (OARs). ConclusionThis study supports that VMAT TBI treatment planning with SIMBa is feasible. In this sample, SIMBa provided dosimetrically similar doses to marrow and sanctuary site doses as TBI while achieving lower doses to OARs. A clinical trial is needed to investigate the clinical implications of VMAT TBI with SIMBa.

show abstract

Section: Discussionmentioning

confidence: 99%

Dosimetric Evaluation Between the Conventional Volumetrically Modulated Arc Therapy (VMAT) Total Body Irradiation (TBI) and the Novel Simultaneous Integrated Total Marrow Approach (SIMBa) VMAT TBI

Stanley¹,

McConnell²,

Iqbal³

et al. 2021

Cureus

View full text Add to dashboard Cite

show abstract

“…In the present work, we focus on foot anatomies that have been generated by 2D image segmentation of CT scans followed by 3D reconstruction of each component (bone). We consider the steps of 3D reconstruction and segmentation pre-processing that are out of the scope of this work (see for example [20][21][22][23][24] for techniques to perform such steps), hence we consider the 28 bones of the foot-and-ankle as the input set I = {P 1 , ..., P 28 } with P i = {x j ∈ R 3 , j = 1, ..., n i }. n i represents the number of 3D points in the cloud associated with the bone i.…”

Section: Definitions and Notationmentioning

confidence: 99%

Computational and Image Processing Methods for Analysis and Automation of Anatomical Alignment and Joint Spacing in Reconstructive Surgery

Chaudhary

et al. 2021

Preprint

View full text Add to dashboard Cite

Purpose: Reconstructive surgeries to treat a number of musculoskeletal conditions, from arthritis to severe trauma, involve implant placement and reconstructive planning components. Anatomically matched 3D printed implants are becoming increasingly patient-specific; however the preoperative planning and design process requires several hours of manual effort from highly trained engineers and clinicians. Our work mitigates this problem by proposing algorithms for the automatic re-alignment of unhealthy anatomies, leading to more efficient, affordable, and scalable treatment solutions. Methods: Our solution combines global alignment techniques such as iterative closest points (ICP) with novel joint space refinement algorithms. The latter is achieved by a low dimensional characterization of the joint space, computed from the distribution of the distance between adjacent points in a joint. Results: Experimental validation is presented on real clinical data from human subjects. Compared with ground truth healthy anatomies, our algorithms can reduce misalignment errors by 22% in translation and 19% in rotation for the full foot-and-ankle and 37% in translation and 39% in rotation for the hind-foot only, achieving a performance comparable to expert technicians. Conclusion: Our methods and histogram-based metric allow for automatic and unsupervised alignment of anatomies, a major step toward a fully automated and data driven re-positioning, designing, and diagnosing tool.

show abstract

“…In the case of deep learning algorithms, promising results were achieved via convolutional neural networks using two-dimensional kernels in a slice-by-slice segmentation ( 9 ) or via a pseudo-3D segmentation ( 10 ) . The latter approach used the successful 2D U-Net developed by Ronneberger et al .…”

Section: Introductionmentioning

confidence: 99%

Semi-Automated 3d Segmentation of Pelvic Region Bones in Ct Volumes for the Annotation of Machine Learning Datasets

Jeuthe

Sánchez

Magnusson

et al. 2021

Radiation Protection Dosimetry

View full text Add to dashboard Cite

Automatic segmentation of bones in computed tomography (CT) images is used for instance in beam hardening correction algorithms where it improves the accuracy of resulting CT numbers. Of special interest are pelvic bones, which—because of their strong attenuation—affect the accuracy of brachytherapy in this region. This work evaluated the performance of the JJ2016 algorithm with the performance of MK2014v2 and JS2018 algorithms; all these algorithms were developed by authors. Visual comparison, and, in the latter case, also Dice similarity coefficients derived from the ground truth were used. It was found that the 3D-based JJ2016 performed better than the 2D-based MK2014v2, mainly because of the more accurate hole filling that benefitted from information in adjacent slices. The neural network-based JS2018 outperformed both traditional algorithms. It was, however, limited to the resolution of 1283 owing to the limited amount of memory in the graphical processing unit (GPU).

show abstract

Automatic bone segmentation in whole-body CT images

Cited by 73 publications

References 15 publications

Dosimetric Evaluation Between the Conventional Volumetrically Modulated Arc Therapy (VMAT) Total Body Irradiation (TBI) and the Novel Simultaneous Integrated Total Marrow Approach (SIMBa) VMAT TBI

Dosimetric Evaluation Between the Conventional Volumetrically Modulated Arc Therapy (VMAT) Total Body Irradiation (TBI) and the Novel Simultaneous Integrated Total Marrow Approach (SIMBa) VMAT TBI

Computational and Image Processing Methods for Analysis and Automation of Anatomical Alignment and Joint Spacing in Reconstructive Surgery

Semi-Automated 3d Segmentation of Pelvic Region Bones in Ct Volumes for the Annotation of Machine Learning Datasets

Contact Info

Product

Resources

About