Abstract
Lung cancer continues to rank as the leading cause of cancer deaths worldwide. One of the most promising techniques for early detection of cancerous cells relies on sputum cell analysis. This was the motivation behind the design and the development of a new computer aided diagnosis (CAD) system for early detection of lung cancer based on the analysis of sputum color images. The proposed CAD system encompasses four main processing steps. First is the preprocessing step which utilizes a Bayesian classification method using histogram analysis. Then, in the second step, mean shift segmentation is applied to segment the nuclei from the cytoplasm. The third step is the feature analysis. In this step, geometric and chromatic features are extracted from the nucleus region. These features are used in the diagnostic process of the sputum images. Finally, the diagnosis is completed using an artificial neural network and support vector machine (SVM) for classifying the cells into benign or malignant. The performance of the system was analyzed based on different criteria such as sensitivity, specificity and accuracy. The evaluation was carried out using Receiver Operating Characteristic (ROC) curve. The experimental results demonstrate the efficiency of the SVM classifier over other classifiers, with 97% sensitivity and accuracy as well as a significant reduction in the number of false positive and false negative rates.
Highlights
Lung cancer ranks as one of the most common causes of death amongst all diseases
In the sputum cell segmentation, we have analyzed the results of the mean shift in gray level feature space and compared them to the results obtained from the Hopfield Neural Network (HNN) proposed in [11]
It obtains a high number of true positives (TP) and true negatives (TN), and reduced number of false positives (FP) and false negatives (FN) which leads to successful classification where all the performance criteria measurements are increased and the classification error is highly reduced
Summary
Lung cancer ranks as one of the most common causes of death amongst all diseases. While there have been a lot of approaches to minimize the fatalities caused by this disease, early detection is considered the best step towards effective treatment. The proposed CAD system was tested on 100 sputum color images for early lung cancer detection, the experimental results were substantially improved, with high values of sensitivity, specificity and accuracy, in addition to an accurate detection of the cancerous cells when compared with the pathologist’s diagnosis results. The novelty of this work is defined as follows: a state-of-the-art complete CAD system is implemented based on the sputum color image analysis, and optimal deployment and combination of existing image processing and analysis techniques for building the computer aided diagnosis (CAD) system is used. The contributions can be summarized as follows: (1) Detection of sputum cell using a Bayesian classification framework; (2) Best color space after analysis of the images with histogram analysis; (3) Mean shift technique for the sputum cell segmentation; (4) Feature extraction, where a set of features are extracted from the nucleus region to be used in the diagnosis process.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
