Polycystic ovary syndrome (PCOS) is a hormonal disorder that affects overall health and daily life. It increases the risk of serious conditions like heart disease, endometrial cancer, and ovarian cancer. Despite advancements in PCOS diagnosis using artificial intelligence (AI), limited research has established a direct association between PCOS parameters and gynaecological cancers such as ovarian and endometrial cancers. This work proposes a comprehensive framework to address this gap to explore these associations and effectively predict gynaecological tumour risks. Then this study presents the development and evaluation of deep learning models for ovarian and endometrium cancer detection using benchmark datasets. For ovarian cancer detection, the MMOTU dataset comprising 1639 ultrasound images was utilized. Models including ResNet50, VGG16, DenseNet121, EfficientNet, and the proposed EENet were trained and tested on the ovarian dataset, with EENet achieving the highest accuracy of 98% significantly outperforming state-of-the-art models such as EfficientNetV2-M by 6.96%.