A deep learning-shape driven level set synergism for pulmonary nodule segmentation

Roy, Rukhmini; Chakraborti, Tapabrata; Chowdhury, Ananda S.

doi:10.1016/j.patrec.2019.03.004

Cited by 52 publications

(21 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The meaning of the numerical sequence in the table has been explained in the experimental data description section. (17)…”

Section: Availability Of Data and Materialsmentioning

confidence: 99%

“…Ye et al [16] proposed a deep learning computer artificial intelligence system for early identification of GGO nodules. Roy et al [17] proposed a collaborative combination of deep learning and shape-driven level set for automatic and accurate segmentation of pulmonary nodules. Wang et al [18] proposed a central focus convolution neural network to segment pulmonary nodules.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Segmentation of small ground glass opacity pulmonary nodules based on Markov random field energy and Bayesian probability difference

et al. 2020

View full text Add to dashboard Cite

Background: Image segmentation is an important part of computer-aided diagnosis (CAD), the segmentation of small ground glass opacity (GGO) pulmonary nodules is beneficial for the early detection of lung cancer. For the segmentation of small GGO pulmonary nodules, an integrated active contour model based on Markov random field energy and Bayesian probability difference (IACM_MRFEBPD) is proposed in this paper. Methods: First, the Markov random field (MRF) is constructed on the computed tomography (CT) images, then the MRF energy is calculated. The MRF energy is used to construct the region term. It can not only enhance the contrast between pulmonary nodule and the background region, but also solve the problem of intensity inhomogeneity using local spatial correlation information between neighboring pixels in the image. Second, the Gaussian mixture model is used to establish the probability model of the image, and the model parameters are estimated by the expectation maximization (EM) algorithm. So the Bayesian posterior probability difference of each pixel can be calculated. The probability difference is used to construct the boundary detection term, which is 0 at the boundary. Therefore, the blurred boundary problem can be solved. Finally, under the framework of the level set, the integrated active contour model is constructed. Results: To verify the effectiveness of the proposed method, the public data of the lung image database consortium and image database resource initiative (LIDC-IDRI) and the clinical data of the Affiliated Jiangmen Hospital of Sun Yat-sen University are used to perform experiments, and the intersection over union (IOU) score is used to evaluate the segmentation methods. Compared with other methods, the proposed method achieves the best results with the highest average IOU of 0.7444, 0.7503, and 0.7450 for LIDC-IDRI test set, clinical test set, and all test sets, respectively. Conclusions: The experiment results show that the proposed method can segment various small GGO pulmonary nodules more accurately and robustly, which is helpful for the accurate evaluation of medical imaging.

show abstract

“…The meaning of the numerical sequence in the table has been explained in the experimental data description section. (17)…”

Section: Availability Of Data and Materialsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Segmentation of small ground glass opacity pulmonary nodules based on Markov random field energy and Bayesian probability difference

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Deep learning methods have recently been highlighted. Benefiting from their powerful identification and classification capabilities, deep learning methods have been introduced into medical image segmentation given a large volume of training data or labeling samples [27]. U-Net is a typical deep learning method that adopts a U-type structure and skip connection in the architecture [28], [29].…”

Section: Introductionmentioning

confidence: 99%

A Novel Framework for Improving Pulse-Coupled Neural Networks With Fuzzy Connectedness for Medical Image Segmentation

et al. 2020

View full text Add to dashboard Cite

A pulse-coupled neural network (PCNN) is a promising image segmentation approach that requires no training. However, it is challenging to successfully apply a PCNN to medical image segmentation due to common but difficult scenarios such as irregular object shapes, blurred boundaries, and intensity inhomogeneity. To improve this situation, a novel framework incorporating fuzzy connectedness (FC) is proposed. First, a comparative study of the traditional PCNN models is carried out to analyze the framework and firing mechanism. Then, the characteristic matrix of fuzzy connectedness (CMFC) is presented for the first time. The CMFC can provide more intensity information and spatial relationships at the pixel level, which is helpful for producing a more reasonable firing mechanism in the PCNN models. Third, by integrating the CMFC into the PCNN framework models, a construction scheme of FC-PCNN models is designed. To illustrate this concept, a general solution that can be applied to different PCNN models is developed. Next, the segmentation performances of the proposed FC-PCNN models are evaluated by comparison with the traditional PCNN models, the traditional segmentation methods, and deep learning methods. The test images include synthetic and real medical images from the Internet and three public medical image datasets. The quantitative and visual comparative analysis demonstrates that the proposed FC-PCNN models outperform the traditional PCNN models and the traditional segmentation methods and achieve competitive performance to the deep learning methods. In addition, the proposed FC-PCNN models have favorable capability to eliminate inference from surrounding artifacts.

show abstract

“…Therefore, they were frequently used in the works which focus on lesion detection and annotation in medical images. For example, SegNet was employed in [20] and [21] for the annotation of pulmonary nodule lesions in CT images and polyp lesions in colonoscopic images, respectively. U-Net was utilized in [9] to annotate breast tumors in MRI images.…”

Section: Introductionmentioning

confidence: 99%

Depth Information-Based Automatic Annotation of Early Esophageal Cancers in Gastroscopic Images Using Deep Learning Techniques

et al. 2020

View full text Add to dashboard Cite

The early diagnoses of esophageal cancer are of great significance in the clinic because they are critical for reducing mortality. At present, the diagnoses are mainly performed by artificial detection and annotations based on gastroscopic images. However, these procedures are very challenging to clinicians due to the large variability in the appearance of early cancer lesions. To reduce the subjectivity and fatigue in manual annotations and to improve the efficiency of diagnoses, computer-aided annotation methods are highly required. In this work, we proposed a novel method that utilized deep learning (DL) techniques to realize the automatic annotation of early esophageal cancer (EEC) lesions in gastroscopic images. The depth map of gastroscopic images was initially extracted by a DL network. Then, this additional depth information was fused with the original RGB gastroscopic images, which were then sent to another DL network to obtain precise annotations of EEC regions. In total, 4231 gastroscopic images of 732 patients were used to build and validate the proposed method. A total of 3190 of those images were EEC images, and the remaining 1041 were non-EEC images. The experimental results show that the combination of depth information and RGB information improved the annotation performance. The final EEC detection rate and mean Dice Similarity Coefficient (DSC) of our method were 97.54% and 74.43%, respectively. Compared with other state-of-the-art DL-based methods, the proposed method showed better annotation performances and fewer false positive outputs. Therefore, our method offers a good prospect in aiding the clinical diagnoses of EEC.INDEX TERMS Gastroscopic image, early esophageal cancer, lesion annotation, deep learning, depth map.

show abstract

A deep learning-shape driven level set synergism for pulmonary nodule segmentation

Cited by 52 publications

References 16 publications

Segmentation of small ground glass opacity pulmonary nodules based on Markov random field energy and Bayesian probability difference

Segmentation of small ground glass opacity pulmonary nodules based on Markov random field energy and Bayesian probability difference

A Novel Framework for Improving Pulse-Coupled Neural Networks With Fuzzy Connectedness for Medical Image Segmentation

Depth Information-Based Automatic Annotation of Early Esophageal Cancers in Gastroscopic Images Using Deep Learning Techniques

Contact Info

Product

Resources

About