Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Sinergi
Gampala, Veerraju
Unknown Affiliation
Civil Engineering, Building, Construction & Architecture Control & Systems Engineering Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Enhanced Classification of Multi Abnormal Brain Tumors Detection Using Customized Inception V3 Arumalla, Nagaraju; Gampala, Veerraju
SINERGI Vol 30, No 1 (2026)
Publisher : Universitas Mercu Buana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22441/sinergi.2026.1.020

Abstract

A brain tumor (BT) is considered to be one of the most fatal diseases in the world, which also demands a very precise and early detection to be successfully addressed. The irregularities in the brain can be detected with the help of a magnetic resonance image, or MRI. Menigoma, glioma, pituitary tumours, and no-tumor are four categories of BT to be classified in this work according to an enhanced transfer learning (TL) approach, generated by the pretrained Inception V3 model. The preprocessing pipeline is new and includes data augmentation to reduce overfitting, a bilateral filter to remove noise, background cropping, and image scaling. The proposed method achieves training accuracy of 94.9% and validation accuracy of 93.8%. With a change in the hyperparameter (k-value), the validation and training accuracies improve to 95.3% and 96.8%, respectively. Furthermore, the model has a high level of generalization, where sensitivity is 92.8 percent, and specificity is 93.5 percent. The combination of transfer learning with the high-level enhancement and strengthening of pictures is novel. Nevertheless, among the factors that can affect generalizability, the variety and size of datasets are important. This model should be confirmed through further research using larger, more diverse datasets and explored in the context of clinical interpretability.
Co-Authors Arumalla, Nagaraju