AI-Driven Sub-Type Classification in Ovarian Cancer
Chinnarammiahgari Sahithi, Sapna Kushwah, Avadh Kishor, Pramod Kumar Singh · 2024
Ovarian cancer is a significant public health concern, ranking as the eighth most frequent cancer in women worldwide. However, it is more prevalent in some regions, such as India, where it is the fourth most common cancer in women. It poses a significant risk due to its high fatality rates, which are related to late-stage diagnosis. Early identification of ovarian cancer is challenging, owing to its diverse nature, which includes numerous subtypes with overlapping characteristics. Furthermore, traditional histological diagnosis, which is based on tissue samples, is time-consuming and subject to interobserver variability among pathologists. Given these intricacies, we have proposed a novel approach that can classify the sub-type classification of endometrial ovarian cancer using pre-trained models like VGG19 and EfficientNetV2. The experimental results discern the potential of models to overcome these limitations and achieve a more accurate and efficient sub-type classification of endometrial and ovarian cancer and could significantly improve early diagnosis and pave the way for personalized treatment strategies.