Open AccessSurvey on Recent Works in Computed Tomography based Computer ‑ Aided Diagnosis of Liver using Deep Learning Techniques
Author(s) -
E. Emerson Nithiyaraj,
S. Arivazhagan
Publication year - 2020
Publication title -
international journal of innovative science and research technology
Language(s) - English
Resource type - Journals
ISSN - 2456-2165
DOI - 10.38124/ijisrt20jul058
Subject(s) - deep learning , artificial intelligence , computer science , cad , modality (human–computer interaction) , segmentation , transfer of learning , medical imaging , computer aided diagnosis , feature (linguistics) , domain (mathematical analysis) , machine learning , pattern recognition (psychology) , engineering , engineering drawing , mathematical analysis , linguistics , philosophy , mathematics
Computed tomography (CT) scanning is a non-invasive diagnostic imaging technique that provides more detailed information about the liver than standard X-rays. Unlike ultrasound (US) examination, the quality of the CT image is not highly operator dependent. Plenty of works has been done using computer-aided diagnosis (CAD) for liver using conventional machine learning algorithms with better results. Recent advances especially in deep learning technology, can detect, classify, segment patterns in medical images where the advancements in deep learning has been shifted to medical domain also. One of the core abilities of deep learning is that they could learn feature representations automatically from data instead of feeding hand crafted features based on application. In this review, the basics of deep learning is introduced and their success in liver segmentation and lesion detection, classification using CT imaging modality is reviewed and their different network architectures is also discussed. Transfer learning is an interesting approach in deep learning which is also discussed. So, deep learning and CAD system has made a huge impact and has produced enhanced performance in healthcare industry.