AI-Powered Intelligent Assistant for Healthcare Report Analysis

DOI

10.2991/978-94-6239-789-7_5How to use a DOI?

Keywords

Artificial Intelligence; Healthcare Reports; MedLinker; Health Insights Agent; Large Language Models; Natural Language Processing; Medical Data Interpretation; Hybrid AI; Clinical Decision Support

Abstract

Modern health care systems generate large scale diagnostic data in the form of lab reports, blood test results, and electronic medical records which patients have difficulty in making sense of and which doctors have trouble processing. This issue of translation of medical info sees delay in diagnosis, miscommunication, increased health care costs, and more anxious patients. We present MedLinker (Health Insights Agent HIA) a hybrid AI which we designed out of rule based medical logic and large language models (LLM) to put together diagnostic parameters, identify issues and present structured bilingual reports for the clinicians and patients. We tested MedLinker on a set of 550 diagnostic reports which included 380 from MIMIC-III and 170 from in house and synthetic sources. The system reported an accuracy of 91.4% for parameter extraction, 88.7% for abnormality detection, 90.2% for recall, and 28.3 seconds average report processing time. Also, we saw that clinical doctors found the summary very reliable at 87%. That which is proposed is that hybrid AI models do in fact present a great solution which makes the analysis of medical reports faster, better and more accessible. As for what is left, we have the issues of multi modal data integration, language accessibility, better explanation of results, protection of data privacy, and easy EMR system integration. We present a proposed framework which addresses these issues.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Ashi Verma
AU  - Riya Aggarwal
AU  - Aastha Panwar
AU  - Vani Rastogi
AU  - Anushka Pundir
PY  - 2026
DA  - 2026/10/08
TI  - AI-Powered Intelligent Assistant for Healthcare Report Analysis
BT  - Proceedings of theInternational Conference on Innovation and Practices in Artificial Intelligence (ICIPAI 2026)
PB  - Atlantis Press
SP  - 45
EP  - 56
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-789-7_5
DO  - 10.2991/978-94-6239-789-7_5
ID  - Verma2026
ER  -