Polycystic Ovary Syndrome (PCOS) is a widespread endocrine disorder affecting women worldwide, necessitating accurate and efficient diagnostic tools. Traditional deep learning approaches have shown impressive results in medical image classification but often face challenges in generalization and computational efficiency. In this study, we implement and compared quantum-enhanced machine learning approaches for PCOS image classification, including Hybrid Quantum-Classical Convolutional Neural Network (HQCNN), Hybrid Quantum-Classical Vision Transformer (HQViT), Variational Quantum Autoencoder (VQAE), Quantum Spiking Neural Networks (QSNN), and Quantum Graph Neural Networks (QGNN). We evaluate these models based on accuracy, precision, recall, F1 score, Area Under the Receiver Operating Characteristic Curve (AUC-ROC), and training time. Our results demonstrate that QSNN achieves superior classification performance, attaining 99.48% accuracy. However, QGNN exhibits limitations in generalization. These findings emphasize the potential of quantum machine learning for medical diagnostics while identifying areas for further optimization. Future work will focus on enhancing quantum feature extraction and exploring real quantum hardware implementation.