STRAMPN: Histopathological image dataset for ovarian cancer detection incorporating AI-based methods

Ovarian cancer, characterized by uncontrolled cell growth in the ovaries, poses a significant threat to women’s reproductive health. Often referred to as the “silent killer,” it is notorious for its elusive nature, as symptoms do not manifest until the disease has advanced to critical stages. Recognizing the urgent need for early detection, this research paper aimed to enhance the identification of ovarian cancer during its initial phases. To bolster the dataset and improve the chances of accurate classification, a comprehensive approach was undertaken. Leveraging available online images, an extensive pre-processing and data augmentation methodology was employed to enrich the dataset. By expanding the dataset size and ensuring its diversity, the research sought to capture a broader range of cancerous manifestations and mitigate potential biases. Utilizing MATLAB, a suite of six state-of-the-art classifiers were employed to categorize the augmented images. To assess the efficacy of the classifiers, a holdout method was adopted for cross-validation. Remarkably, the results showcased an exceptional accuracy rate of 99%, underscoring the effectiveness of the methodology in detecting ovarian cancer at its incipient stages. The implications of this research are far-reaching, as the early identification of ovarian cancer holds immense potential for improved prognosis and treatment outcomes. By shedding light on the significance of expanding and diversifying datasets and leveraging advanced classification techniques, this study contributes to the growing body of knowledge aimed at combating ovarian cancer and underscores the importance of early intervention in reducing mortality rates associated with this insidious disease.