<![CDATA[This study aims to develop an automatic classification model for Polycystic Ovary Syndrome (PCOS) using deep learning with transfer learning based on the MobileNetV2 architecture. The dataset consists of 1,987 ovarian ultrasound images that underwent preprocessing and augmentation. The model was initialized with pre-trained ImageNet weights, trained using binary cross-entropy loss and the Adam optimizer, and evaluated using accuracy, precision, recall, and F1-score metrics. The training and testing were conducted on a cloud computing platform with resource-efficient settings. The results demonstrate that the model can classify normal and PCOS ovarian images with 99% accuracy, 0.99 precision, 0.99 recall, and 0.99 F1-score. The confusion matrix indicates very few misclassifications, with four normal images incorrectly predicted as PCOS and seven PCOS images misclassified as normal. These findings confirm that MobileNetV2 is effective, efficient, and stable for classifying low-resolution medical images. The proposed model has the potential to serve as a practical automatic diagnostic tool based on ultrasound imaging, which can be implemented on resource-constrained devices as well as cloud platforms to support medical decision-making.]]>
Copyrights © 2025
