A hierarchical stochastic modelling approach for reconstructing lung tumour geometry from 2D CT images

Afshar, Parnian; Ahmadi, Abbas; Mohebi, Azadeh; Zarandi, M. H. Fazel

doi:10.1080/0952813x.2018.1509894

Cited by 7 publications

(13 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Employing the interpolation method with MC can further reduce the error so that interpolation decreases the error by about 0.5, 0.8 relative to the fairing for two metrics. However, the recent work of Afshar et al [ 2 ] with 5.23, 1.29 at HD, ED, is superior to other standard methods, including MC and MC + fairing. This algorithm's most significant differences are with MC, at about 3 and 2 improvements at HD, ED.…”

Section: Empirical Evaluationsmentioning

confidence: 97%

“…Based on the literature, most studies have been conducted on the issue of 3D brain tumor reconstruction. In 2018, P. Afshar et al [ 2 ] have developed a method using a sequence of 2D images for 3D reconstruction of tumor geometry. This method involves two steps: tumor segmentation from CT images and 3D shape reconstruction.…”

Section: Related Workmentioning

confidence: 99%

“…For 3D reconstruction, first, a new approach is proposed to interpolate some intermediate slices between the main slices, then the famous marching cubes algorithm is used to reconstruct the tumor geometry. Finally, the surface is smoothed using an explicit fairing algorithm [ 2 ].…”

Section: Related Workmentioning

confidence: 99%

“…We consider four approaches: marching cubes (MC) [ 47 ], marching cubes + fairing [ 15 ], Interpolation + marching cubes [ 6 ] and Afshar et al [ 2 ] to compare the proposed 3D reconstruction model. In addition to HD, we use Euclidean Distance (ED) as an evaluation metric, in which HD calculates the shape accuracy, while ED serves as pixel-wise distance.…”

Section: Empirical Evaluationsmentioning

confidence: 99%

“…Although these examinations have promising results, most of them are not valuable for 3D lung tumor reconstruction because they either require relatively fine detail in all of 3D dimensions or do not consider the randomization nature of tumor shape. To the best of our knowledge, a recent study [ 2 ] has been presented for 3D lung tumor reconstruction. This traditionally based work uses mathematical operations for solving the problem.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A GAN-based method for 3D lung tumor reconstruction boosted by a knowledge transfer approach

Rezaei

Ahmadi

2023

Multimed Tools Appl

Self Cite

View full text Add to dashboard Cite

Three-dimensional (3D) image reconstruction of tumors has been one of the most effective techniques for accurately visualizing tumor structures and treatment with high resolution, which requires a set of two-dimensional medical images such as CT slices. In this paper we propose a novel method based on generative adversarial networks (GANs) for 3D lung tumor reconstruction by CT images. The proposed method consists of three stages: lung segmentation, tumor segmentation and 3D lung tumor reconstruction. Lung and tumor segmentation are performed using snake optimization and Gustafson-Kessel (GK) clustering. In the 3D reconstruction part first, features are extracted using the pre-trained VGG model from the tumors that detected in 2D CT slices. Then, a sequence of extracted features is fed into an LSTM to output compressed features. Finally, the compressed feature is used as input for GAN, where the generator is responsible for high-level reconstructing the 3D image of the lung tumor. The main novelty of this paper is the use of GAN to reconstruct a 3D lung tumor model for the first time, to the best of our knowledge. Also, we used knowledge transfer to extract features from 2D images to speed up the training process. The results obtained from the proposed model on the LUNA dataset showed better results than state of the art. According to HD and ED metrics, the proposed method has the lowest values of 3.02 and 1.06, respectively, as compared to those of other methods. The experimental results show that the proposed method performs better than previous similar methods and it is useful to help practitioners in the treatment process.

show abstract

Section: Empirical Evaluationsmentioning

confidence: 97%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Empirical Evaluationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A GAN-based method for 3D lung tumor reconstruction boosted by a knowledge transfer approach

Rezaei

Ahmadi

2023

Multimed Tools Appl

Self Cite

View full text Add to dashboard Cite

show abstract

GAN‐LSTM‐3D: An efficient method for lung tumour 3D reconstruction enhanced by attention‐based LSTM

Hong,

Modirrousta,

Hossein Nasirpour

et al. 2023

CAAI Trans on Intel Tech

View full text Add to dashboard Cite

Three‐dimensional (3D) image reconstruction of tumours can visualise their structures with precision and high resolution. In this article, GAN‐LSTM‐3D method is proposed for 3D reconstruction of lung cancer tumours from 2D CT images. Our method consists of three phases: lung segmentation, tumour segmentation, and tumour 3D reconstruction. Lung segmentation is done using snake optimisation followed by tumour segmentation using Gustafson‐Kessel (GK) clustering method. The outputs of GK (2D lung cancer images) are fed to a pre‐trained Visual Geometry Group (VGG) for feature extraction. The VGG outputs are used as input for an attention‐based LSTM which performs feature unpacking. The output of LSTM units is given to generator network of a Generative Adversarial Networks (GAN) model to carry out 3D reconstruction of (normal/cancerous) images with high quality. During training, the discriminator network of the GAN is used to judge the generator outputs. The authors to the best of their knowledge were the first to use GAN for 3D reconstruction of lung cancer tumours which is the primary contribution of this article. Moreover, existing studies are mostly focused on brain tumours and are not suitable for lung tumour reconstruction. Focusing on lung tumours is the second contribution of this article. Evaluation on LUNA data collection against existing methods like MC, MC + fairing etc. reveals the superiority of our method in terms of Hamming and Euclidean distance metrics. Additionally, the computational complexity of the proposed method is lower compared to evaluated methods.

show abstract

Persistent mucus plugs in proximal airways are consequential for airflow limitation in asthma

Huang,

Elicker,

Henry

et al. 2023

JCI Insight

View full text Add to dashboard Cite

show abstract

A hierarchical stochastic modelling approach for reconstructing lung tumour geometry from 2D CT images

Cited by 7 publications

References 13 publications

A GAN-based method for 3D lung tumor reconstruction boosted by a knowledge transfer approach

A GAN-based method for 3D lung tumor reconstruction boosted by a knowledge transfer approach

GAN‐LSTM‐3D: An efficient method for lung tumour 3D reconstruction enhanced by attention‐based LSTM

Persistent mucus plugs in proximal airways are consequential for airflow limitation in asthma

Contact Info

Product

Resources

About