Automatic identification of ring enhancing lesion pattern in cases of brain infection and metastasis brain tumor based on invariant moment features classification

Riries Rulaningtyas, Lailatul Muqmiroh, Leonard Ferdian, Abidah Alfi, Isma Iba

Research output: Contribution to journalArticlepeer-review

Abstract

Brain infection and metastasis brain tumor in CT Scan examination have similar ring-enhancing lesion patterns. This research aims to develop a program that aids the radiologists to identify these brain disorders. The radiologists often have difficulties and mostly subjective when distinguishing the ring-enhancing lesion whether brain infection or metastatic brain tumor, especially in patients with no previous history of the disease. With these limitations, this research produces a Computer Aided Diagnose (CAD) system in order to assist the radiologists in brain disorders observing from the images of CT scan brain scanning. The CAD system is supported by features extraction method which is generated using the combinations of Hu's invariant moment features. The decision maker uses the backpropagation neural network method to classify the brain disorders based on their invariant moment features that could help the radiologists when identifying the brain abnormalities. The results of brain abnormalities identification including normal, brain infection, and metastasis brain tumor yielded performance with 88.9% accuracy, 86% sensitivity, and 100% specificity.

Original languageEnglish
Pages (from-to)42-52
Number of pages11
JournalMalaysian Journal of Science
Volume38
DOIs
Publication statusPublished - 29 Nov 2019

Keywords

  • Backpropagation
  • CT scan
  • Identification
  • Invariant moment features
  • Neural network

Fingerprint

Dive into the research topics of 'Automatic identification of ring enhancing lesion pattern in cases of brain infection and metastasis brain tumor based on invariant moment features classification'. Together they form a unique fingerprint.

Cite this