Advancing Ovarian Cancer Detection Through Explainable AI and Multimodal Machine Learning Integration

Ramisetti Umamaheswari1 , T.Lahari2 , Shaik Arshiya Zehra3 , Nakka Hemasree4 , Gunthakanti Harish Babu5
1Assistant Professor Dept of Information Technology, SV College of Engineering, Tirupati, India.
2B.Tech, Dept of Information Technology, SV College of Engineering, Tirupati, India.
3B.Tech, Dept of Information Technology, SV College of Engineering, Tirupati, India.
4B.Tech, Dept of Information Technology, SV College of Engineering, Tirupati, India.
5B.Tech, Dept of Information Technology, SV College of Engineering, Tirupati, India.

Abstract-Ovarian cancer remains a formidable challenge in gynecological oncology due to its asymptomatic progression and the limitations of
traditional screening methods. Existing systems leverage comprehensive preprocessing and various machine learning and deep learning algorithms to predict ovarian cancer using clinical and biomarker data. Tehniques such as feature selection and dimensionality reduction enhance model robustness, while classifiers and ensemble methods yieldimproved, albeit variable, accuracies. Nonetheless, thecurrent approaches encounter limitations including sensitivity to data transformations, reduced efficacy of RNNs for tabular medical data, and the complexity of optimizing ensemble and autoencoder-based models. The envisioned future system proposes the integration of Explainable AI and multimodal data fusion, encompassing clinical, imaging, and genetic insights, to advance ovarian cancer prediction. This approach notonly aims to elevate diagnostic accuracy but also enhances result interpretability and trust. Real-world deployment in clinical decision support systems will facilitate early and informed intervention, ultimately advancing patient outcomes. By overcoming existing limitationsand utilizing transparent, robust frameworks, the proposed system is poised to transform early ovarian cancer detection and support healthcare professionals in critical decision-making.Keywords: Ovarian cancer, Deep Learning, Ensemble and Autoencoder, Explainabletransformations, Clinical Imaging.