| Autor: |
Wang, Jinhua, Wang, Liang, Yang, Zhongxian, Tan, Wanchang, Luo, Min, Liu, Yubao |
| Předmět: |
|
| Zdroj: |
Multimedia Tools & Applications; May2024, Vol. 83 Issue 18, p55075-55090, 16p |
| Abstrakt: |
The present study aims to evaluate the effectiveness of the multifractal analysis in classifying benign and malignant tumor cells on MRI (Magnetic Resonance Imaging) images from patients diagnosed with breast cancer. In this study, the three-dimensional multifractal analysis method was adapted to comprehensively analyze MRI breast images and evaluate chemotherapy in terms of effectiveness. Twenty-four Chinese patients (Guangdong, Shenzhen, People's Republic of China (PRC)) between the ages of 18 and 60 were included in the study. All patients were divided in two groups. Group 1 included 10 women whose MRI breast images were used to estimate the multifractal scaling exponents, with lacunarity taken as a multifractal measure. Group 2 included 14 patients whose MRI breast images underwent multifractal analysis to evaluate chemotherapy in terms of effectiveness. The MRI images were taken before and after chemotherapy. The combined multifractal analysis is effective for classifying malignant and benign neoplasms. Based on the results of the study, the multifractal spectrum width has proven an informative parameter in quantitative analysis of cancer treatment effectiveness. It correlates with the rate of blood supply to malignant tumors. The baseline Hurst exponent was below 0.4, indicating a low long-term correlation in the original data. However, after chemotherapy treatment, the exponent significantly increased to a range of 0.7–0.9, suggesting enhanced long-term correlation and a potential association with the effectiveness of chemotherapy in the context of breast cancer. Receiver operating characteristic (ROC) curve analysis for fractal dimension and lacunarity showed 70.8% sensitivity and 65.9% specificity and 72.4% sensitivity and 89.3% specificity, respectively. Therefore, lacunarity is more effective as a parameter in neoplastic cell identification. The present findings provide prediction capability for breast cancer at various stages. The multifractal method exhibited high efficiency of tumor classification in MRI. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
Full text from SpringerLink