Enhancement Filter for Computer-Aided Detection of Pulmonary Nodules on Thoracic CT images

Yu, Yang; Zhao, Hong

doi:10.1109/isda.2006.253783

Cited by 8 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The suspected pulmonary nodule region exhibits the form of a circular or oval object whereas vascular tissue structures presents a line‐like elongated structure. Therefore, this property can be used to distinguish different shape structures present in lung lobes …”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Fully automatic detection of lung nodules in CT images using a hybrid feature set

et al. 2017

View full text Add to dashboard Cite

Purpose The aim of this study was to develop a novel technique for lung nodule detection using an optimized feature set. This feature set has been achieved after rigorous experimentation, which has helped in reducing the false positives significantly. Method The proposed method starts with preprocessing, removing any present noise from input images, followed by lung segmentation using optimal thresholding. Then the image is enhanced using multiscale dot enhancement filtering prior to nodule detection and feature extraction. Finally, classification of lung nodules is achieved using Support Vector Machine (SVM) classifier. The feature set consists of intensity, shape (2D and 3D) and texture features, which have been selected to optimize the sensitivity and reduce false positives. In addition to SVM, some other supervised classifiers like K‐Nearest‐Neighbor (KNN), Decision Tree and Linear Discriminant Analysis (LDA) have also been used for performance comparison. The extracted features have also been compared class‐wise to determine the most relevant features for lung nodule detection. The proposed system has been evaluated using 850 scans from Lung Image Database Consortium (LIDC) dataset and k‐fold cross‐validation scheme. Results The overall sensitivity has been improved compared to the previous methods and false positives per scan have been reduced significantly. The achieved sensitivities at detection and classification stages are 94.20% and 98.15%, respectively, with only 2.19 false positives per scan. Conclusions It is very difficult to achieve high performance metrics using only a single feature class therefore hybrid approach in feature selection remains a better choice. Choosing right set of features can improve the overall accuracy of the system by improving the sensitivity and reducing false positives.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Therefore, this property can be used to distinguish different shape structures present in lung lobes. 45 For circular structures we have:…”

Section: B Image Enhancement and Nodule Detectionmentioning

confidence: 99%

Fully automatic detection of lung nodules in CT images using a hybrid feature set

et al. 2017

View full text Add to dashboard Cite

show abstract

“…This paper focuses on identifying MAs and HMAs based on the analysis of eigenvalues of the image Hessian matrix. The advantage of this approach is that basing on the eigenvalues not only detects vessel-like, but also sheet-like or blob-like structures [14,15]. This paper is organized as follows: Section 2 presents the overview of both automated and semi automated approach towards DR detection.…”

Section: Introductionmentioning

confidence: 99%

Diabetic Retinopathy Detection Based on Eigenvalues of the Hessian Matrix

Rubini¹,

Kunthavai²

2015

Procedia Computer Science

View full text Add to dashboard Cite

Diabetic Retinopathy (DR) is a medical condition caused by fluctuating insulin level in the blood which causes vision loss in case of severity. Timely treatment of such risks requires identification of the first clinical symptoms like microaneurysms (MAs) and hemorrhages (HMAs). The presence of those symptoms are visible in the digital color photographs of the retina and appear as round dark red spots in the image. In this paper, two approaches in the detection of MAs and HMAs are proposed. First, the semi automated approach applies semi automated hessian-based candidate selection algorithm (SHCS) followed by thresholding to detect true MAs and HMAs. The automated approach applies automated hessian-based candidate selection algorithm (AHCS) followed by feature extraction and SVM classification that uses twenty images for training manually annotated by medical domain experts. Implementations of both the approaches have been tested on real world images from retinal scan. From the results, the detection rate of automated algorithm when compared with that of the semi automated algorithm has been found to be significantly lesser with a probability p<0.005.

show abstract

“…In general, the method based on the Hessian Eigen values analysis is capable of detecting not only tubular structures, but also blob-like and sheet-like structures within the image [3]. This only requires finding proper formulas for "blobness" and "sheetness" as functions of λ i .…”

Section: Introductionmentioning

confidence: 99%