A survey on machine learning based brain retrieval algorithms in medical image analysis
- PDF / 1,109,188 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 92 Downloads / 195 Views
REVIEW PAPER
A survey on machine learning based brain retrieval algorithms in medical image analysis Arpit Kumar Sharma 1 & Amita Nandal 1 & Arvind Dhaka 1 & Rahul Dixit 2 Received: 23 May 2020 / Accepted: 30 July 2020 # IUPESM and Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract In recent times, researchers showed huge interest in machine learning approaches that attempts to develop the information representations via computational modules. Past decade gained momentum by deep learning approaches and their potential of enhancing the performance for numerous automation operations with superior future research applications. The novelties in medical image processing initialized the unique perspective to diagnose the human body with superior resolution and enhanced accuracy. This paper offers a comprehensive work on existing methodologies that attain optimum results in their respective domains. There exist various Magnetic Resonance Imaging (MRI) brain scan classifiers to obtain efficient features extraction images. The fundamental step in these methods includes several actions to be performed by using different approaches in order to characterize the anomalous developments in MRI scans of brain. Mostly, current techniques are utilizing deep learning feature extraction algorithm from MRI brain scans to obtain their relevant features. Currently, deep learning algorithms associated with medical imaging results in achieving remarkable performance enhancement in diagnosis as well as characterization of complex pathologies in case of brain tumors. This paper provides existing research gaps in identification, segmentation and feature extraction among current approaches. This paper also suggests the future directions to increase the efficiency of current models. Keywords Brain tumor . Convolutional neural networks . Deep learning . Feature extraction . Machine learning . MRI . And segmentation
1 Introduction Brain tumor is an uncontrolled and unnatural development of cells which affects the human functions and spread into other body organs. Human skull is a volume restricted and rigid body fragment therefore; any unexpected growth influences * Arvind Dhaka [email protected] Arpit Kumar Sharma [email protected] Amita Nandal [email protected] Rahul Dixit [email protected] 1
Department of Computer and Communication Engineering, Manipal University Jaipur, Jaipur, India
2
Department of Computer Science Engineering, IIIT Pune, Maharashtra, India
the functions of organs or parts that are involved with the brain [1]. World Health Organization (WHO) published the world cancer report, which states that 2% of brain tumor is responsible for human cancer; but it produces extreme complications as well as morbidity [2]. The revised WHO characterization on brain tumour have shown the widespread use as future perspective in the research works on epidemiological as well as clinical procedures. It also can deliver the new insights into etiological features due to the patterns discrimination of obtai
Data Loading...
