Patient-Wise Versus Nodule-Wise Classification of Annotated Pulmonary Nodules using Pathologically Confirmed Cases

Abstract

This paper presents a novel framework for combining well known shape, texture, size and resolution informatics descriptor of solitary pulmonary nodules (SPNs) detected using CT scan. The proposed methodology evaluates the performance of classifier in differentiating benign, malignant as well as metastasis SPNs with 246 chests CT scan of patients. Both patient-wise as well as nodule-wise available diagnostic report of 80 patients was used in differentiating the SPNs and the results were compared. For patient-wise data, generated a model with efficiency of 62.55% with labeled nodules and using semisupervised approach, labels of rest of the unknown nodules were predicted and finally classification accuracy of 82.32% is achieved with all labeled nodules. For nodule-wise data, ground truth database of labeled nodules is expanded from a very small ground truth using content based image retrieval (CBIR) method and achieved a precision of 98%. Proposed methodology not only avoids unnecessary biopsies but also efficiently label unknown nodules using prediagnosed cases which can certainly help the physicians in diagnosis.

An Appraisal of Lung Nodules Automatic Classification Algorithms for CT Images

“…Aggarwal and Vig [39] combined shape, texture, size, resolution informatics, statistical and intensity features descriptor, and principal component analysis (PCA) was performed for feature selection. They achieved accuracy of 82.32% for patient-wise data.…”

“…Some selected papers are based on user-defined features, including shape [29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60], texture [29,33,34,35,36,37,38,39,41,44,45,46,47,48,50,54,59,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,…”

“…Some selected papers are based on generic features, including scale-invariant feature transform (SIFT) [17,70,71,72,96,107], local binary pattern (LBP) [41,70,73,74,76,107], histogram of oriented gradient (HOG) [43,50,70,94,107], Gabor [39,41,72,73,74,107], and speeded-up robust features (SURF) [76,99].…”

“…The main methods for classifier construction are machine learning-based, which uses the training algorithm and labeled data. Many kinds of classifiers are used in selected papers, such as those based on support vector machines (SVM) [29,34,35,36,37,39,41,46,47,49,62,63,64,65,70,71,75,80,81,92,94,96,99,100,101,102,107,110,115,130], bayesian classifier [35,40,46,62,98,117], rule based [106], k-means [57,58,86], linear discriminant analysis (LDA) [33,46,54,57,58,68,72,76,82,88], artificial neural networks (ANNs) [44,45,50,55,62,66,69,77,78,79,104,112,121,122], convolutional Neural Networks (CNNs) […”

An Improved Approach to Term Weighting in Hierarchical Web Page Classification

“…Web pages are essential hypertext. Besides text and multimedia components, Web pages also include context features such as hyperlinks, HTML tags and metadata [21]. Various context features have been used in Web classification.…”