<p>Ovarian cysts are basically fluid-filled sacs which develop inside the ovaries being the part of the reproductive system of females. The ovaries are responsible for producing eggs and hormones like estrogen and progesterone and ultrasound images are required for accurate detection. Several recent studies have focused on applying deep learning methods to improve diagnostic accuracy and efficiency along with efforts to develop non-invasive imaging approaches for the early detection of ovarian cysts. To generate more accurate results a fuzzy inference-based MobileNetV2 model was implemented which helped in detecting if the cyst had less chance, medium chance or more chance using the ultrasound images. The dataset consisted of 265 ultrasound images representing 8 classes of ovarian cysts. The training phase involved data augmentation and model training using the Adam optimizer over 24 epochs achieving a training accuracy of 93.16%. A synthetic image generator was implemented to refine a clear view of different synthetic images from the original images. ImageDataGenerator and normalization using Keras library in Python enhances data augmentation leading to more accurate results. Overall the model obtained an accuracy of 97% on the data set which was tested. The model achieved a precision of 97.5%, recall of 96.4% and F1-score of 96.5%. The use of synthetic images contributed to a 30% improvement in overall accuracy. Key challenges included managing class imbalance within the dataset and ensuring sufficient variation in the synthetic images generated for effective training.</p>