Computer-aided Diagnosis of Soft Tissue Tumors on High-resolution Ultrasonography with Geometrical and Morphological Features

Chen, Chih-Yen; Chiou, Hong-Jen; Chou, Yi-Hong; Chiou, See-Ying; Wang, Hsin-Kai; Chou, Szu-Yuan; Chiang, Huihua Kenny

doi:10.1016/j.acra.2008.12.016

« Previous Next »Academic Radiology
Volume 16, Issue 5 , Pages 618-626, May 2009

Computer-aided Diagnosis of Soft Tissue Tumors on High-resolution Ultrasonography with Geometrical and Morphological Features

Institute of Biomedical Engineering, National Yang Ming University, No 155, Sec 2, Linong Street, Beitou District, Taipei City 112, Taiwan (C.-Y.C., H.K.C.); the Department of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan (H.-J.C., Y.-H.C, S.-Y.C., H.-K.W); General Hospital and National Yang Ming University, School of Medicine, Taipei, Taiwan (H.-J.C.); the Department of Obstetrics and Gynecology, Wan-Fang Hospital and Shuang-Ho Hospital, Taipei Medical University, Taipei, Taiwan (S.-Y.C.); and the Department of Education and Research, Taipei City Hospital, Taipei, Taiwan (H.K.C.)

Received 23 September 2008; accepted 30 December 2008.

Rationale and Objectives

The aim of this study was to develop a semi-automated computer-aided diagnosis (CAD) system based on high-resolution ultrasonography for classifying benign and malignant soft tissue tumors (STTs).

Materials and Methods

One hundred seven patients with STTs (70 benign and 37 malignant) were enrolled, and regions of interest were manually delineated for analysis. Sixteen tumor shape features, including five geometric features and 11 morphologic features (six old and five new normalized radial length [NRL] features) were individually evaluated using Student's t test and the area under the receiver-operating characteristic curve (A_z). Then linear discriminant analysis with stepwise feature selection was used to construct a semi-automated CAD system with old NRL features, new NRL features, and all features combined. Additionally, two experienced radiologists participated in malignancy grading of tumors. To investigate the associations among CAD results, pathologic results, and radiologists' rankings, Spearman's rank correlation coefficient was used in the statistical analysis.

Results

The results showed that 11 features had P values < .05, and five of the proposed features were significant. The optimal CAD system achieved accuracy of 87.9%, sensitivity of 89.2%, specificity of 87.1%, and an A_z value of 0.93. Correlation between pathologic results and radiologists' rankings was obtained (radiologist A: r=0.62, P < .01; radiologist B: r=0.61, P < .01). In addition, a higher correlation between pathologic results and CAD results (r=0.73, P < .01) was demonstrated.

Conclusion

This semi-automated CAD method based on tumor shape features can successfully distinguish between benign and malignant STTs. It can also provide a second opinion to ultrasound for the diagnosis of STTs.

Key Words: Soft tissue tumor, high-resolution ultrasonography, HRUS, computer-aided diagnosis, CAD, normalized radial length, NRL