Limits...
Computer-aided lung nodule recognition by SVM classifier based on combination of random undersampling and SMOTE.

Sui Y, Wei Y, Zhao D - Comput Math Methods Med (2015)

Bottom Line: However, problems of unbalanced datasets often have detrimental effects on the performance of classification.Eight features including 2D and 3D features are extracted for training and classification.Experimental results show that for different sizes of training datasets our RU-SMOTE-SVM classifier gets the highest classification accuracy among the four kinds of classifiers, and the average classification accuracy is more than 92.94%.

View Article: PubMed Central - PubMed

Affiliation: Software College, Northeastern University, Shenyang 110004, China.

ABSTRACT
In lung cancer computer-aided detection/diagnosis (CAD) systems, classification of regions of interest (ROI) is often used to detect/diagnose lung nodule accurately. However, problems of unbalanced datasets often have detrimental effects on the performance of classification. In this paper, both minority and majority classes are resampled to increase the generalization ability. We propose a novel SVM classifier combined with random undersampling (RU) and SMOTE for lung nodule recognition. The combinations of the two resampling methods not only achieve a balanced training samples but also remove noise and duplicate information in the training sample and retain useful information to improve the effective data utilization, hence improving performance of SVM algorithm for pulmonary nodules classification under the unbalanced data. Eight features including 2D and 3D features are extracted for training and classification. Experimental results show that for different sizes of training datasets our RU-SMOTE-SVM classifier gets the highest classification accuracy among the four kinds of classifiers, and the average classification accuracy is more than 92.94%.

Show MeSH

Related in: MedlinePlus

Nodule and nonnodule sequent images.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4419492&req=5

fig5: Nodule and nonnodule sequent images.

Mentions: We choose 120 thoracic CT scans for the experiments. To set the dataset, we extracted nodule and nonnodule regions from the lung images, and they are all examined by expert radiologists. We created the nodule and nonnodule regions in forms of volume data, that is, a pixels ×  b pixels ×  c layers; a, b, and c stand for size of the nodule or nonnodule in x, y and, z direction, respectively, the range of a and b is 10~50 pixels, and the range of c is 5~13. We create 150 nodules and 908 nonnodules for the dataset. Figure 5 shows 6 nodule and 6 nonnodule sequent images of the dataset, and groups (a) and (b) show nodule and nonnodule images, respectively.


Computer-aided lung nodule recognition by SVM classifier based on combination of random undersampling and SMOTE.

Sui Y, Wei Y, Zhao D - Comput Math Methods Med (2015)

Nodule and nonnodule sequent images.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4419492&req=5

fig5: Nodule and nonnodule sequent images.
Mentions: We choose 120 thoracic CT scans for the experiments. To set the dataset, we extracted nodule and nonnodule regions from the lung images, and they are all examined by expert radiologists. We created the nodule and nonnodule regions in forms of volume data, that is, a pixels ×  b pixels ×  c layers; a, b, and c stand for size of the nodule or nonnodule in x, y and, z direction, respectively, the range of a and b is 10~50 pixels, and the range of c is 5~13. We create 150 nodules and 908 nonnodules for the dataset. Figure 5 shows 6 nodule and 6 nonnodule sequent images of the dataset, and groups (a) and (b) show nodule and nonnodule images, respectively.

Bottom Line: However, problems of unbalanced datasets often have detrimental effects on the performance of classification.Eight features including 2D and 3D features are extracted for training and classification.Experimental results show that for different sizes of training datasets our RU-SMOTE-SVM classifier gets the highest classification accuracy among the four kinds of classifiers, and the average classification accuracy is more than 92.94%.

View Article: PubMed Central - PubMed

Affiliation: Software College, Northeastern University, Shenyang 110004, China.

ABSTRACT
In lung cancer computer-aided detection/diagnosis (CAD) systems, classification of regions of interest (ROI) is often used to detect/diagnose lung nodule accurately. However, problems of unbalanced datasets often have detrimental effects on the performance of classification. In this paper, both minority and majority classes are resampled to increase the generalization ability. We propose a novel SVM classifier combined with random undersampling (RU) and SMOTE for lung nodule recognition. The combinations of the two resampling methods not only achieve a balanced training samples but also remove noise and duplicate information in the training sample and retain useful information to improve the effective data utilization, hence improving performance of SVM algorithm for pulmonary nodules classification under the unbalanced data. Eight features including 2D and 3D features are extracted for training and classification. Experimental results show that for different sizes of training datasets our RU-SMOTE-SVM classifier gets the highest classification accuracy among the four kinds of classifiers, and the average classification accuracy is more than 92.94%.

Show MeSH
Related in: MedlinePlus