Explainable Brain Tumor Detection Using Deep Learning Models with Quantitative Explainability Metrics

Alvita Mary D silva; Shwetha S; Leesha H U; Monith Monnappa U M; Dharshan B R

Explainable Brain Tumor Detection Using Deep Learning Models with Quantitative Explainability Metrics

Author(s)	Alvita Mary D silva, Prof. Shwetha S, Leesha H U, Monith Monnappa U M, Dharshan B R
Country	India
Abstract	Brain tumor detection using Magnetic Resonance Imaging (MRI) plays a crucial role in early diagnosis and treatment planning. Deep learning models have demonstrated high accuracy in automating this task; however, their black-box nature limits clinical trust. This research presents a comprehensive and explainable brain tumor detection framework using four state-of-the-art deep learning architectures: Xception, ResNet50, DenseNet121, and EfficientNetB4. In addition to conventional performance metrics, explainability is integrated using Gradient-weighted Class Activation Mapping (Grad-CAM) and SHapley Additive exPlanations (SHAP) to provide both region-level and pixel-level interpretability. Furthermore, quantitative explainability metrics including Insertion Test, Deletion Test, Sensitivity-N, Average Confidence Drop, and Average Confidence Gain are employed to objectively evaluate explanation faithfulness. Experimental results on MRI datasets demonstrate that Xception and EfficientNet models achieve superior classification performance, while the explainability analysis offers deeper insights into model reliability and clinical trustworthiness. The proposed framework enhances transparency, robustness, and clinical applicability of AI-based brain tumor detection systems.
Keywords	Brain Tumor Detection, Deep Learning, MRI Images, Explainable AI, Grad-CAM, Quantitative Explainability Metrics
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 1, January-February 2026
Published On	2026-02-18

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 1 Cover Page Vol 7 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Explainable Brain Tumor Detection Using Deep Learning Models with Quantitative Explainability Metrics

Share this