A Distributed Cloud-Centric Framework for Scalable CT/MRI Image Processing with Automated Segmentation

Department of MCA Dayananda Sagar College of Engineering Bangalore, India
Department of MCA Dayananda Sagar College of Engineering Bangalore, India
Department of MCA Dayananda Sagar College of Engineering Bangalore, India
Assistant Professor, Department of MCA Dayananda Sagar College of Engineering Bangalore, India

This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Modern computed tomography (CT) and magnetic resonance imaging (MRI) systems generate large volumes of high-resolution medical images, placing significant computational demands on conventional processing infrastructures. As imaging studies increase in size and frequency, operations such as slice normalization, artifact removal, and tumor segmentation become major performance bottlenecks. Traditional single-node systems often lack the scalability and computational resources required to process such data efficiently.

To address these challenges, this paper proposes a distributed cloud-based framework for scalable CT and MRI image analysis. The framework leverages Apache Spark for parallel preprocessing and GPU-accelerated inference for efficient tumor segmentation. Distributed execution enables the system to handle large imaging workloads while reducing preprocessing latency compared to sequential approaches.

The proposed framework is validated using a publicly available brain MRI dataset. Experimental results demonstrate that the system efficiently processes large numbers of image slices and generates reliable outputs, including preprocessed images, segmentation masks, and performance metrics. Parallelization significantly improves preprocessing speed while maintaining consistent segmentation performance.

Furthermore, the framework supports privacy-aware deployment through optional secure computation, making scalable and confidential medical image analysis more practical in clinical environments.