A 3D reconstruction and visualization framework from a sequence of 2D cross-sectional images has aroused extensive concern particularly in medical science and engineering. In this research, we design parallel reconstruction and visualization frameworks for medical data reconstruction and visualization. Reconstruction methods for medical data normally require high-performance computing. In this paper we make a web-based 3D reconstruction service and present a parallel computing framework for real-time processing. The visualization framework design visualizes a generated 3D model. Moreover, we implement a web-based 3D medical data reconstruction and visualization application that integrates both proposed frameworks and design our framework integrated with X3D technology and a modular approach. The web-based X3D framework integrated with a modular approach can extend cross-platform and inter-application data transfer ability and support run-time architecture to represent and communicate 3D scenes and objects.