A CI feature-based pulmonary nodule segmentation using three-domain mean shift clustering

Nie, Shengdong; Chen, Zhaoxue; Li, Lihong

doi:10.1109/icwapr.2007.4420699

Cited by 5 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The loss compression methods of current medical image data contain adaptive predictive coding scheme, discrete cosine transform (DCT), sub-band coding and vector quantization (VQ). It has two classes, that is transform and non transformation, for example JPEG, MPEG and fractal compression [3].…”

Section: Diversified Methods Of Image Condensationmentioning

confidence: 99%

Medical Image Condensation Based on Wavelet Transform

Zhang

2015

AMM

View full text Add to dashboard Cite

This paper based on the peculiarity of wavelet transform that its transform only in vacuum region and frequency region, decompose image use the theory of wavelet, obtain a series sub-image of different resolution ratio. The value of higeresolution ratio sub-image is all verge on 0, the phenomenon is more obviously in high frequency, so that, the mainly proportion is low frequency to a image. Use wavelet decomposition get rid of the high frequency, only reservation low frequency, to realize the aim of condensation image. Through the simulation of contradistinctive image of cerebra framework remotion between three-dimensional ultrasonic imaging in course of OPS and MIR preceding OPS validated the feasibility by Matlab.

show abstract

Section: Diversified Methods Of Image Condensationmentioning

confidence: 99%

Medical Image Condensation Based on Wavelet Transform

Zhang

2015

AMM

View full text Add to dashboard Cite

show abstract

“…The classification results of three input dimensions were fused. The fusion formula is as function (2).…”

Section: Multilevel Contextual Network and Model Fusionmentioning

confidence: 99%

“…The early lung cancer lesions are characterized by pulmonary nodules, 2 which are mostly shown as approximately circular and high-density lung shadows on CT scan images. 3 Therefore, the detection of lung nodules is the first step in lung cancer screening.…”

Section: Introductionmentioning

confidence: 99%

“…The nodules can be roughly divided into three types 6 : solid nodules, ground glass nodules, hollow nodules, as shown in Figure 1. (2) Because the internal structure of the lung parenchyma is complex, the position changes of the lung nodules also bring great difficulties to our lung nodule detection work. Figure 2 shows several typical positions, Figure 2A,B shows the adhesion of the lung nodules to the pleural wall, and Figure 2(C) shows the connection between the lung nodules and blood vessels.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A improved detection method for lung nodule based on multi‐scale 3Dconvolutional neural network

Tan

Zhu

et al. 2022

Concurrency and Computation

View full text Add to dashboard Cite

Low detection sensitivity and high false positives are two main challenges from traditional lung nodule detection model. In order to improve reliability of detection model, we proposed a multi‐scale three‐dimensional convolutional neural network lung nodule detection method which can simultaneously detect lung nodules and reduce false positives. First, a 3D UNet++‐like architecture with an encoding‐decoding structure is regarded as a feature extraction network and combines it with the region proposal network. The multi‐feature fusion method can fully learn features at different levels through the cross‐layer connection between the encoder and the decoder. Second, the layers of the 3D UNet++‐like architecture are connected by residual blocks, which enhance feature reuse and speeds up the convergence speed of the network. Finally, the three input sizes are input into the 3D neural network, and their classification results are merged, thus to find the final nodule determination result. Based on the LUNA16 dataset, the experiment result shows that average sensitivity is 87.3% and be increased 7.8% than the UNet++ network in our model. And if the number of candidate nodules is 48, the sensitivity is as high as 96.2%, it increases 8.1% than the VGG16 network. Obviously, that the improved model can significantly improve detection sensitivity and reduce false positives, which can provide a valuable theoretical reference for clinical medicine.

show abstract

“…The chart below show, the outline of image is mainly in low frequency part, detail is in high frequency part through two-dimensional wavelet decomposition. Data processing is discrete in numerical application, so the change of size j 2 is limited, narrow shampling rate make j 2 cannot arbitrarily small, it cannot arbitrarily big to be limited by computing resource, generally the finest scale normalized to 1 and the coarsest scale is supposed N 2 , N is analysis of the series [7].…”

Section: The Matlab Achieve Of Picture Enhancementmentioning

confidence: 99%

The MATLAB Realization of Toning up Digital Image Based on Wavelet Transform

Zhang

Xie

2013

AMM

View full text Add to dashboard Cite

This paper mainly description image enhance based on wavelet transform. The author gave the theory of wavelet analysis, the application of wavelet transform in image enhancement. The impression drawing is given by means of MATLAB simulation. Through tone up to disperse wavelet transform, rebuild coefficient, the siginal of image is more boost up in low contrast section than high contrast section, the margin is distinct.

show abstract

A CI feature-based pulmonary nodule segmentation using three-domain mean shift clustering

Cited by 5 publications

References 7 publications

Medical Image Condensation Based on Wavelet Transform

Medical Image Condensation Based on Wavelet Transform

A improved detection method for lung nodule based on multi‐scale 3Dconvolutional neural network

The MATLAB Realization of Toning up Digital Image Based on Wavelet Transform

Contact Info

Product

Resources

About