January 13, 2026 Proposal Closed Access

Agegnehu Teshome

DataCite XML Export

<?xml version='1.0' encoding='utf-8'?>
<resource xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://datacite.org/schema/kernel-4" xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4.1/metadata.xsd">
  <identifier identifierType="DOI">10.20372/nadre:23965</identifier>
  <creators>
    <creator>
      <creatorName>Agegnehu Teshome</creatorName>
      <affiliation>Mekdela Amba University</affiliation>
    </creator>
  </creators>
  <titles>
    <title>Medical Image Captioning Using Deep Learning</title>
  </titles>
  <publisher>Zenodo</publisher>
  <publicationYear>2026</publicationYear>
  <subjects>
    <subject>Medical Image Captioning</subject>
    <subject>Deep Learning</subject>
    <subject>Chest X-ray Analysis</subject>
    <subject>Faster R-CNN</subject>
  </subjects>
  <contributors>
    <contributor contributorType="ResearchGroup">
      <contributorName>Habtamu Shiferaw</contributorName>
      <affiliation>Mekdela Amba University</affiliation>
    </contributor>
    <contributor contributorType="ResearchGroup">
      <contributorName>Getie Balew</contributorName>
      <affiliation>Mekdela Amba University</affiliation>
    </contributor>
    <contributor contributorType="ResearchGroup">
      <contributorName>Melese Alemante</contributorName>
      <affiliation>Mekdela Amba University</affiliation>
    </contributor>
  </contributors>
  <dates>
    <date dateType="Issued">2026-01-13</date>
  </dates>
  <resourceType resourceTypeGeneral="Text">Proposal</resourceType>
  <alternateIdentifiers>
    <alternateIdentifier alternateIdentifierType="url">https://nadre.ethernet.edu.et/record/23965</alternateIdentifier>
  </alternateIdentifiers>
  <relatedIdentifiers>
    <relatedIdentifier relatedIdentifierType="DOI" relationType="IsVersionOf">10.20372/nadre:23964</relatedIdentifier>
    <relatedIdentifier relatedIdentifierType="URL" relationType="IsPartOf">https://nadre.ethernet.edu.et/communities/mau-community</relatedIdentifier>
  </relatedIdentifiers>
  <rightsList>
    <rights rightsURI="info:eu-repo/semantics/closedAccess">Closed Access</rights>
  </rightsList>
  <descriptions>
    <description descriptionType="Abstract">&lt;p&gt;This research proposal presents a novel deep learning framework for automated medical image captioning, focusing specifically on chest X-ray analysis. The study aims to develop a robust and interpretable system that automatically generates accurate, clinically relevant textual descriptions by integrating object detection and caption generation. The proposed methodology optimizes the Faster R-CNN architecture through advanced techniques&amp;mdash;including Feature Pyramid Network (FPN), Feature Reuse, and Optimized Anchor Generation&amp;mdash;to enhance the detection of small and overlapping anatomical structures. Furthermore, an end-to-end learning approach is adopted to improve contextual understanding and caption quality directly from raw image data. By addressing key limitations in current methods, such as spatial context neglect and sequential processing bottlenecks, this research seeks to contribute to diagnostic accuracy, workflow efficiency, and patient communication in clinical practice.&lt;/p&gt;</description>
    <description descriptionType="Other">On Progress</description>
  </descriptions>
</resource>