Multiple ROI selection based focal liver lesion classification in ultrasound images

Jeon, Jae Hyun; Choi, Jae Young; Lee, Sihyoung; Ro, Yong Man

doi:10.1016/j.eswa.2012.07.053

Cited by 44 publications

(30 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A few tumorous regions of interest (ROI(s)) are marked and segregated by radiologists through a GUI developed by the authors. Many researchers have used different sizes of ROI(s) which has a considerable effect on the analysis [11,12]. However, the optimal ROI size varies according to the feature extraction methodology and its final application.…”

Section: Background Theorymentioning

confidence: 99%

Computer Aided Diagnosis System-A Decision Support System for Clinical Diagnosis of Brain Tumours

Tiwari¹,

Sachdeva²,

Ahuja³

et al. 2017

IJCIS

View full text Add to dashboard Cite

The iso, hypo or hyper intensity, similarity of shape, size and location complicates the identification of brain tumors. Therefore, an adequate Computer Aided Diagnosis (CAD) system is designed for classification of brain tumor for assisting inexperience radiologists in diagnosis process. A multifarious database of real post contrast T1-weighted MR images from 10 patients has been taken. This database consists of primary brain tumors namely Meningioma (MENI-class 1), Astrocytoma (AST-class 2), and Normal brain regions (NORM-class 3). The region of interest(s) (ROIs) of size 20 x 20 is extracted by the radiologists from each image in the database. A total of 371 texture and intensity features are extracted from these ROI(s). An Artificial Neural Network (ANN) is used to classify these three classes as it shows better classification results on multivariate non-linear, complicated, rule based domains, and decision making domains. It is being observed that ANN provides much accurate results in terms of individual classification accuracy and overall classification accuracy. The four discrete experiments have been performed. Initially, the experiment was performed by extracting 263 features and an overall classification accuracy 78.10% is achieved, however, it was noticed that MENI (class-1) was highly misclassified with AST (class-2). Further, to improve the overall classification accuracy and individual classification accuracy specifically for MENI (class-1), LAWs textural energy measures (LTEM) are added in the feature bank (263+108=371). An individual class accuracy of 91.40% is obtained for MENI (class-1), 91.43% for AST (class-2), 94.29% for NORM (class-3) and an overall classification accuracy of 92.43% is achieved. The results are calculated with and without addition of LTEM feature with Principle component analysis (PCA)-ANN. LTEM-PCA-ANN approach improved results with an overall accuracy of 93.34%. The texture patterns obtained were clear enough to differentiate between MENI (class-1) and AST (class-2) despite of necrotic and cystic component and location and size of tumor. LTEM detected fundamental texture properties such as level, edge, spot, wave and ripple in both horizontal and vertical directions which boosted the texture energy.

show abstract

Section: Background Theorymentioning

confidence: 99%

Computer Aided Diagnosis System-A Decision Support System for Clinical Diagnosis of Brain Tumours

Tiwari¹,

Sachdeva²,

Ahuja³

et al. 2017

IJCIS

View full text Add to dashboard Cite

show abstract

“…A crucial step for obtaining good generalization performance is correct choice of the regularization parameter C and kernel parameter γ. The optimal values for C and γ are obtained by extensive search, carried out in the parameter space for the values of C є {2 -4 , 2 -3 … 2 15 }, γ є {2 -12 , 2 -11 … 2 4 } using 10 fold cross validation on training data.…”

Section: Classification Module 2431 Svm Classifiermentioning

confidence: 99%

“…The researchers [13][14][15] experimented classification by considering malignant lesions as single class; however, the classification of malignant lesions as HCC or MET lesions is clinically significant for effective treatment of liver malignancies 4 . As per the best of the author's information, only two studies 5,6 are reported for five-class classification of FLLs using B-Mode US images.…”

Section: Brief Details Of Studies For Classification Of Flls Using B-mentioning

confidence: 99%

See 1 more Smart Citation

PCA-SVM based CAD System for Focal Liver Lesions using B-Mode Ultrasound Images

Virmani¹,

Kumar²,

Kalra³

et al. 2013

DSJ

View full text Add to dashboard Cite

“…Experimental results show that the proposed method with the dense FV has achieved an area under the receiver of characteristics (AUC) of 96.77%, sensitivity and specificity of 98.04% and 93.75% for the placental maturity staging, respectively. Our experimental results also demonstrate that the dense feature outperforms the traditional sparse feature for placental maturity staging.Over the past decade, ultrasound (US) imaging has been extensively applied in prenatal diagnosis and prognosis since it is radiation-free, direct-use, and low-cost [1][2][3][4][5][6][7] …”

mentioning

confidence: 99%

Discriminative learning for automatic staging of placental maturity via multi-layer fisher vector

Chen

et al. 2015

2015 10th International Conference on Information, Communications and Signal Processing (ICICS)

View full text Add to dashboard Cite

Currently, placental maturity is performed using subjective evaluation, which can be unreliable as it is highly dependent on the observations and experiences of clinicians. To address this problem, this paper proposes a method to automatically stage placenta maturity from B-mode ultrasound (US) images based on dense sampling and novel feature descriptors. Specifically, our proposed method first densely extracts features with a regular grid based on dense sampling instead of a few unreliable interest points. Followed by, these features are clustered using generative Gaussian mixture model (GMM) to obtain high order statistics of the features. The clustering representatives (i.e., cluster means) are encoded by Fisher vector (FV) for staging accuracy enhancement. Differing from the previous studies, a multi-layer FV is investigated to exploit the spatial information rather than the single layer FV. Experimental results show that the proposed method with the dense FV has achieved an area under the receiver of characteristics (AUC) of 96.77%, sensitivity and specificity of 98.04% and 93.75% for the placental maturity staging, respectively. Our experimental results also demonstrate that the dense feature outperforms the traditional sparse feature for placental maturity staging.Over the past decade, ultrasound (US) imaging has been extensively applied in prenatal diagnosis and prognosis since it is radiation-free, direct-use, and low-cost [1][2][3][4][5][6][7]

show abstract

Multiple ROI selection based focal liver lesion classification in ultrasound images

Cited by 44 publications

References 19 publications

Computer Aided Diagnosis System-A Decision Support System for Clinical Diagnosis of Brain Tumours

Computer Aided Diagnosis System-A Decision Support System for Clinical Diagnosis of Brain Tumours

PCA-SVM based CAD System for Focal Liver Lesions using B-Mode Ultrasound Images

Discriminative learning for automatic staging of placental maturity via multi-layer fisher vector

Contact Info

Product

Resources

About