Every year, millions of women throughout the world are affected by cancer of the cervix. Earlier diagnosis and treatment by optimal medical guidance are essential to mitigate the impacts of the condition similar to other issues. Smear images are among the most effective methods for detecting cervical malignancy. This paper presents a Deep Hyper Tuned Convolutional Neural Network (Deep-HTCNN) utilizing a chaotic theory-based modified grey wolf optimization (CGWO) technique which is suitable for accurate categorization of cervical cancer with smear images. The Grey Wolf Optimizer (GWO) is a nature-inspired meta-heuristic method driven by the social hunting habits of grey wolves. This work incorporates the chaos hypothesis with the GWO algorithm to enhance its global convergence rate. The findings indicated that utilizing a suitable chaotic map enables CGWO to significantly surpass traditional GWO, demonstrating superior performance relative to other frameworks.