Advancements in Cancer Diagnosis Using Digital Imaging System: A Review

Abstract

Imaging and microscopic concepts are the most preferable mode of a cancer diagnosis. It is a wide area of research. Digital Pathology is one of the prompt technique which can provide an accurate and optimized solution. Different imaging techniques like magnetic resonance tomography, mammography, sonography, ultrasound, and biopsy histopathologic images are playing a very important role in computer-aided cancer diagnosis. Biomedical imaging follows the cancer clinical protocols and provides authentic information about various biomarkers and its features, morphology, structure, functions, and metabolism. Further advancements in computation and storage technology lead to enhance imaging speed to meet physiological requirements in real-time. Basic phases of computer-aided diagnosis are pre-processing of sample data, segmentation, feature extraction, classification, and report generation. Many performance evaluation parameters help to rank the models. Different machine learning algorithms like k-means clustering, support vector machine, KNN (K Nearest Neighbors) and various Neural Network models are helping for prediction, as a result, good accuracy with high precision can be achieved. Early diagnosis of cancer via screening based on imaging has one of the most important roles to reduce the mortality of cancer patients. CAD will always be cost-effective and assist pathologists and oncologists to take the correct decision after that follow the appropriate treatment.

Keywords:Computer-Aided Diagnosis, Biopsy histopathologic images, Machine learning