Related Experiment Videos
A novel method for breast cancer prognosis using wavelet packet based neural network.
Sepehr H Jamarani1, Gholamali Rezai-Rad, Hamid Behnam
1Department of Biomedical Eng., Science and Research Branch, Islamic Azad University, Tehran, IRAN. jamarani@shemroon.com.
Summary
This study introduces a novel method for early breast cancer detection using artificial neural networks (ANN) and wavelet-based image analysis to identify microcalcifications in mammograms. The approach enhances diagnostic accuracy for early-stage breast cancer.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Signal Processing
Background:
- Early detection of breast cancer is crucial for effective treatment and improved patient outcomes.
- Microcalcifications in mammograms are key indicators of early-stage breast cancer.
- Traditional methods for microcalcification detection can be challenging due to image noise and subtle features.
Purpose of the Study:
- To develop and evaluate an automated approach for detecting microcalcifications in digital mammograms.
- To combine wavelet-based subband decomposition with artificial neural networks (ANN) for enhanced diagnostic accuracy.
- To assess the performance of the proposed method using established mammographic databases and clinical data.
Main Methods:
- Digital mammograms were decomposed into different frequency subbands using wavelet packet transforms.
- Low-frequency subbands were suppressed, and high-frequency subbands containing microcalcification information were reconstructed.
- The reconstructed images were used as input for an artificial neural network (ANN) for classification.
- The methodology was validated using the Nijmegen and MIAS mammographic databases and local hospital images.
Main Results:
- The proposed approach successfully detected microcalcifications by isolating high-frequency image components.
- Performance was evaluated using Receiver Operating Characteristic (ROC) curves.
- Quantitative results, including the area under the ROC curve, demonstrated the effectiveness of the method.
Conclusions:
- The combination of wavelet-based subband decomposition and artificial neural networks offers a promising approach for early breast cancer diagnosis.
- This automated method can aid radiologists in identifying microcalcifications, potentially leading to earlier and more accurate diagnoses.
- Further validation on larger datasets could solidify its clinical utility in breast cancer screening programs.