Publications by authors named "Lan-Hsin Cheng"
Biomed Eng Online
May 2023
Article Synopsis
- An AI system based on deep learning was developed to help differentiate between malignant and benign calcifications in spot magnification mammograms, aiming to lower the number of unnecessary biopsies.
- The study analyzed a total of 2,508 mammogram images, with the AI's performance evaluated using receiver operating characteristic (ROC) analysis, achieving a strong area under the curve of 0.888.
- The results indicated that the AI could potentially eliminate 80.8% of benign biopsies, while maintaining high sensitivity (88.4%) and specificity (80.8%), showcasing its accuracy in classifying suspicious calcifications.
