| Autor: |
Kumar, Anil, De Mukhopadhyay, Keya, Verma, Ashok Kumar, Mamoria, Pushpa, Tariq, Wali, Dawn, V. J. |
| Předmět: |
|
| Zdroj: |
African Journal of Biomedical Research; Sep2024, Vol. 27 Issue 3, p693-699, 7p |
| Abstrakt: |
Breast cancer remains one of the leading causes of cancer-related deaths among women globally. Early detection through advanced imaging techniques is crucial for improving survival rates. This study evaluates the integration of an artificial intelligence (AI) system into 3D mammography (digital breast tomosynthesis, DBT) for breast cancer diagnosis. A total of 5,000 women who underwent DBT screening between January 2021 and June 2023 were included. The AI system, based on a deep learning convolutional neural network, was trained on a dataset of annotated DBT images and compared against the interpretations of two experienced radiologists. Key performance metrics such as sensitivity, specificity, accuracy, and area under the curve (AUC) were analyzed. Results showed that the AI system achieved higher sensitivity (94.2%) and specificity (92.5%) than the radiologists, with an AUC of 0.968, indicating superior diagnostic performance. Additionally, AI-assisted readings significantly reduced radiologist interpretation time by 44%, suggesting workflow efficiency improvements. While the AI system showed promising results in improving detection accuracy and efficiency, further studies in diverse populations are needed to validate its clinical application. This research highlights the potential of AI as a valuable tool in breast cancer diagnosis, aiding radiologists in enhancing diagnostic accuracy and reducing time to diagnosis. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
