A novel CAD system to automatically detect cancerous lung nodules using wavelet transform and SVM

Ayman A Abu Baker, Yazeed Ghadi

Abstract


This paper presents an ongoing effort to detect cancerous nodules in Computed Tomography Images (CT-Images). Due to the images size and resolution, CT-images are considered to be diagnosed by radiologist. Therefore, this will increase the radiologist fatigue and may miss some of the cancerous lung nodule lesions. In this paper a new novel CAD system that will used to detect cancerous nodule is proposed. The proposed algorithm is divided to four stages. In the first stage, an enhancement algorithm is implement to highlight the suspicious regions. Then in the second stage, the region on interest will be detected. The adaptive SVM and Wavelet transform techniques are used to reduce the detected false positive regions. The proposed algorithm is tested and evaluated on 60 normal and cancerous lung nodule CT-Images. As a result, this new approach can efficiently detect the cancerous lung nodules with TP ration 94.5% but with slightly high number of detected FP regions which is 7 cluster/image.

Keywords


Cancer detection; Computed tomography; DICOM; Wavelet features; Wavelet transform



DOI: http://doi.org/10.11591/ijece.v10i5.pp%25p
Total views : 9 times

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.