<?xml version="1.0" encoding="UTF-8"?>



<records>

  <record>
    <language>eng</language>
          <publisher>Oriental Scientific Publishing Company</publisher>
        <journalTitle>Biomedical and Pharmacology Journal</journalTitle>
          <issn>0974-6242</issn>
            <publicationDate>2025-10-21</publicationDate>
    
        <volume>18</volume>
        <issue>October Spl Edition</issue>

 
    <startPage></startPage>
    <endPage></endPage>

	    <publisherRecordId>68436</publisherRecordId>
    <documentType>article</documentType>
    <title language="eng">An Explainable AI-Based Ensemble Framework for Brain Tumor MRI Classification and Automated Clinical Reporting Using Deep Learning and LLM Integration</title>

    <authors>
	 


      <author>
       <name>Kaliprasanna Swain</name>

 
		
	<affiliationId>1</affiliationId>
      </author>
    

	 


      <author>
       <name>Soumya Ranjan Nayak</name>


		
	<affiliationId>1</affiliationId>

      </author>
    

	 


      <author>
       <name>Santosh Kumar Swain</name>

		
	<affiliationId>1</affiliationId>
      </author>
    

	 


      <author>
       <name>Prabhishek Singh</name>

		
	<affiliationId>2</affiliationId>
      </author>
    


	


	
    </authors>
    
	    <affiliationsList>
	    
		
		<affiliationName affiliationId="1">School of Computer Engineering, Kalinga Institute of Industrial Technology (KIIT) Deemed to be University, Bhubaneswar, Odisha, India</affiliationName>
    

		
		<affiliationName affiliationId="2">School of Computer Science Engineering and Technology, Bennett University, Greater Noida, India </affiliationName>
    
		
		
		
		
	  </affiliationsList>






    <abstract language="eng">Magnetic Resonance Imaging (MRI) is invaluable for brain tumor diagnosis, but remains time-consuming and subject to inter-observer variability. This article introduces an AI-based MRI Analysis App designed to support clinicians through precise, automated brain tumor classification and clinical interpretation. The system employs an ensemble of Convolutional Neural Networks (DenseNet121, ResNet50, EfficientNet-B0, and MobileNetV3-Small) to categorize brain MRI images into Glioma, Meningioma, No Tumor, and Pituitary Tumor, using 7,023 images from Figshare, SARTAJ, and Br35H datasets, with preprocessing steps like grayscale conversion, normalization, and data augmentation. To promote transparency, Explainable AI techniques, including Grad-CAM, LIME, edge detection, and SHAP, are integrated, while a Large Language Model (LLM) via LangChain-Groq generates natural language clinical reports. The ensemble achieved high classification accuracy and consistent validation across tumor types, with XAI methods offering visual insights into model predictions and a PDF reporting system facilitating clinician use. By combining ensemble learning, interpretability, and LLM-based reporting, this AI tool delivers a real-time, end-to-end solution for automated, explainable MRI analysis, positioning itself as a valuable asset in radiological practice.</abstract>

    <fullTextUrl format="html">https://biomedpharmajournal.org/vol18octoberspledition/an-explainable-ai-based-ensemble-framework-for-brain-tumor-mri-classification-and-automated-clinical-reporting-using-deep-learning-and-llm-integration/</fullTextUrl>

<keywords language="eng">

      
        <keyword>Automated Clinical Report Generation</keyword>
      

      
        <keyword> Brain Tumor Classification</keyword>
      

      
        <keyword> Convolutional Neural Networks (CNNs)</keyword>
      

      
        <keyword> Explainable Artificial Intelligence (XAI)</keyword>
      

      
        <keyword> Large Language Models (LLMs)</keyword>
      

      
        <keyword> Magnetic Resonance Imaging (MRI)</keyword>
      
</keywords>
  </record>
</records>