Minimum redundancy maximum relevance (mRMR) based feature selection from endoscopic images for automatic gastrointestina
- PDF / 866,413 Bytes
- 11 Pages / 439.642 x 666.49 pts Page_size
- 106 Downloads / 217 Views
Minimum redundancy maximum relevance (mRMR) based feature selection from endoscopic images for automatic gastrointestinal polyp detection Mustain Billah1
· Sajjad Waheed1
Received: 18 May 2019 / Revised: 21 April 2020 / Accepted: 28 May 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In this paper, a computer based system has been proposed as a support to gastrointestinal polyp detection. It can detect and classify gastrointestinal polyps from endoscopic video. Color wavelet (CW) features and convolutional neural network (CNN) features of endoscopic video frames are extracted. Mutual information based feature selection technique-Minimum redundancy maximum relevance (mRMR) is used to scale down feature vector. Instead of using a single classifier, Bootstrap Aggregrating (Bagging)- an ensemble classifier is used. Proposed system has been assessed against different public databases and our own datasets. Evaluation shows that, the system outperforms the existing methods. Keywords Minimum redundancy maximum relevance (mRMR) · Video endoscopy · Ensemble classifier · Feature selection · Convolutional Neural Network (CNN) · Color Wavelet (CW) · Feature extraction
1 Introduction GASTROINTESTINAL cancer is one of the most dominating causes of death in the whole world. It actually originates from gastrointestinal polyps. Irregular growth of tissue on gastric mucosa turns into gastrointestinal polyps. Gradually this polyps increases in size but in most of the cases they do not produce symptoms. Therefore, prevention of cancer is supposed to the early detection of polyps. Video endoscopy is the most popular diagnostic modality for early detection of gastrointestinal polyps. A tiny camera is inserted directly to examine the innermost region of gastrointestinal tract to detect and remove polyps. Several Mustain Billah
[email protected] Sajjad Waheed [email protected] 1
Department of Information and Communication Technology (ICT), Mawlana Bhashani Science and Technology University (MBSTU), Tangail, Bangladesh
Multimedia Tools and Applications
human factors including doctor’s experience, lack of concentration, video length influences accuracy of this diagnosis procedure. Such misdetection of polyps can lead to malignant cancerous tumors in the future. Here comes the usability and advantage of computer aided polyp detection system. Such an scheme can cut down polyp misdetection rate. It can detect and classify polyps. It also supports doctors in acquisition of crucial regions. It can generate detailed report about any particular region to asses the necessity of re-examining the region with more attention. Additionally, duration of this bitter proceudure for the subject can be minimized and the cost of process can also be scaled down. Numerous methods have been practiced for computer aided polyp detection system in literature. These systems differ from the context of diverse feature extraction techniques, heterogeneous datasets, different color models and distinct classifiers
Data Loading...
