Review Paper on VLSI Implementation of Tumor Detection using Machine Learning Technique

Abstract

The rapid growth of medical imaging technologies has led to an increasing demand for fast, accurate, and energy-efficient tumor detection systems. Machine Learning (ML) and Deep Learning (DL) models have shown exceptional performance in classifying and detecting tumors in modalities such as MRI, CT, and PET. However, software-based implementations often require high computational resources, making them unsuitable for real-time and portable healthcare devices. To address this challenge, researchers are moving toward Very-Large-Scale Integration (VLSI) implementations of ML-based tumor detection algorithms. This review paper presents a comprehensive analysis of recent advancements in VLSI architectures designed for tumor detection using ML techniques. It highlights various hardware-optimized models such as Support Vector Machines, Convolutional Neural Networks, Gradient Boosting Machines, Decision Trees, and hybrid ML–VLSI systems. Key design parameters including power consumption, hardware complexity, latency, memory usage, and scalability are systematically reviewed. Furthermore, the paper discusses architectural optimization strategies such as pipelining, parallelism, quantization, approximate computing, and multiplier-less design approaches to achieve low-power operation. Finally, open research challenges and future directions are identified, emphasizing the need for high-accuracy, low-power ML hardware capable of supporting next-generation point-of-care medical diagnostic systems.