An online segmentation tool for cervicographic image analysis

Cervicography is an important visual screening method for cervical cancer prevention. Automatic segmentation of clinically significant regions in acquired images may provide valuable assistance toward research in cervical cancer detection. This paper presents a Web-accessible cervicographic image segmentation system that incorporates several novel segmentation algorithms developed for particular tissue types and landmarks. The system combines the advantages of two commonly used programming languages, Matlab and Java. It relieves the research groups in academic institutes from the heavy burden of re-developing the sophisticated segmentation algorithms originally implemented in Matlab, while allowing medical experts to evaluate the segmentation algorithms using, perhaps, their own image data acquired at remote locations. It offers attractive properties of flexibility, extensibility and Web-accessibility in a prototype image processing application. The system is integrated with other applications that have been developed for uterine cervix image analysis at the U.S. National Library of Medicine. The architecture and concept of this system are generalizable and can be applied to different medical image processing tasks.

[1]  Hayit Greenspan,et al.  Automated and Interactive Lesion Detection and Segmentation in Uterine Cervix Images , 2010, IEEE Transactions on Medical Imaging.

[2]  Siddharth Samsi,et al.  A Java-based Web Interface to Matlab , 2003 .

[3]  Hayit Greenspan,et al.  Automatic detection of specular reflections in uterine cervix images , 2006, SPIE Medical Imaging.

[4]  Shelly Lotenberg,et al.  Automatic Detection of Anatomical Landmarks in Uterine Cervix Images , 2009, IEEE Transactions on Medical Imaging.

[5]  S. Gordon,et al.  Content-based indexing and retrieval of uterine cervix images , 2004, 2004 23rd IEEE Convention of Electrical and Electronics Engineers in Israel.

[6]  Jose Jeronimo,et al.  Tissue classification using cluster features for lesion detection in digital cervigrams , 2008, SPIE Medical Imaging.

[7]  Sunanda Mitra,et al.  Ultra high resolution image coding and ROI viewing using line-based backward coding of wavelet trees (L-BCWT) , 2009, 2009 Picture Coding Symposium.

[8]  Mark Schiffman,et al.  ASCUS-LSIL Triage Study , 2000, Acta Cytologica.

[9]  Hannu Koivisto,et al.  Global Condition Monitoring System - Implementing MATLAB-Based Analysis Services , 2004, ICINCO.

[10]  Jose Jeronimo,et al.  A tool for collection of region based data from uterine cervix images for correlation of visual and clinical variables related to cervical neoplasia , 2004, Proceedings. 17th IEEE Symposium on Computer-Based Medical Systems.

[11]  Jose Jeronimo,et al.  Comparative performance analysis of cervix ROI extraction and specular reflection removal algorithms for uterine cervix image analysis , 2007, SPIE Medical Imaging.

[12]  A. Štafl,et al.  Cervicography: a new method for cervical cancer detection. , 1981, American journal of obstetrics and gynecology.

[13]  Shelly Lotenberg,et al.  Shape Priors for Segmentation of the Cervix Region Within Uterine Cervix Images , 2008, SPIE Medical Imaging.

[14]  Daniel P. Lopresti,et al.  Balancing the Role of Priors in Multi-Observer Segmentation Evaluation , 2009, J. Signal Process. Syst..

[15]  Allan Hildesheim,et al.  Description of a seven-year prospective study of human papillomavirus infection and cervical neoplasia among 10000 women in Guanacaste, Costa Rica,. , 2004, Revista panamericana de salud publica = Pan American journal of public health.