Abstract

 Background & Aims: Cancer is a serious problem for human being. A necessary prerequisite for any therapeutic modality is early diagnosis. Fractal dimension measurement of nuclear boundary of cancer cells can be used as a feature extraction in the field of automated cancer diagnosis.

 Materials & Methods: 41 positive urine cytology and 33 negative ones were selected randomly. Digitalized images prepared using objective 100X and a digital video head preprocessed subsequently and the fractal dimension of nuclear boundaries was measured by a software.

 Results:�Findings revealed a statistically significant difference between fractal dimensions of nuclear boundaries of cancerous versus non-cancerous in urinary smears. Selecting a proper cut off point (1.732 ± 0.006 in this study) can discriminate malignant and non-malignant epithelial cells in urinary smears.

 Conclusion: Based on diagnostic accuracy measures (sensitivity and specificity), probability of disease measures (predictive value of a positive and negative test results), and likelihood ratio of positive and negative tests, it seems fractal dimension of nuclear cell boundaries in urinary smears can be used as a feature extraction in the field of automated cancer diagnosis.

 Keywords:�Automated cancer diagnosis, Screening machine design, Fractal dimension

  Address : Pathology Department, Imam Khomeini Hospital, Urmia, Iran Tel : (+98441) 3469931

  E-mail: ilkhanib@gmail.com

  Source: UMJ 2009: 20(2): 156 ISSN: 1027-3727