This dissertation addresses information security application advancement through designing and analyzing quantum circuit approaches. It focuses on both non-parametric circuit and parameterized quantum circuits (PQCs) models, addressing information security challenges across three core areas. First, it introduces a novel depth-efficient quantum multiplier based on the Toom-Cook 20.5-way method utilizing high- and half-degree multiplication techniques to build multiplication with time and space efficiency for cryptographic multiplication computations. Second, the research presents a depth-optimized elliptic curve point multiplication (ECPM) circuit designed for Shor-based quantum cryptanalysis circuits in binary field elliptic curves. By minimizing quantum cost and optimizing ECPM circuit depth, the proposed approach enhances computational efficiency, providing a robust foundation for evaluating the security of cryptographic algorithms, in particular, elaborating on the quantum attacks aspect in information security. Thirdly, we investigate hybrid classical-quantum approaches for steganalysis by integrating PQCs with classical neural networks in machine learning models. This approach leverages the strengths of quantum and classical integration to improve the detection and classification of concealed information within digital images, addressing steganalysis accuracy improvement challenges in information security.