Large Language Models in Medical Diagnostics: Advancing Clinical Practice, Research, and Patient Care

Dear Colleagues, 

Large Language Models (LLMs) represent a breakthrough in artificial intelligence with significant potential to facilitate clinical diagnostics and workflow. Their capacity to process, interpret, and generate natural language—particularly from unstructured clinical narratives, laboratory reports, and imaging summaries—enables novel applications in diagnostic reasoning, patient communication, workflow automation, and research that can work to improve health.

While promising, the integration of LLMs into diagnostic medicine remains at an early stage, constrained by challenges related to real-world performance, interpretability, bias mitigation, and regulatory compliance. There is a critical need for a dedicated forum to examine how LLMs can enhance diagnostic accuracy, efficiency, and accessibility while addressing inherent risks such as algorithmic errors, data privacy concerns, and model generalizability.

This Special Issue of Diagnostics will focus specifically on the role of LLMs in supporting, refining, and accelerating diagnostic processes across medical specialties. We seek contributions that demonstrate empirical advances, validate clinical utility, and engage with the practical and ethical dimensions of implementing LLM-based tools in diagnostic settings.

Potential Topics for the Special Issue

1. Diagnostic Decision Support
   • LLMs for differential diagnosis generation and case-based reasoning.
   • Integration of LLMs with clinical data (EHRs, lab results, imaging reports) to support diagnostic accuracy.
   • Comparative studies between LLM-assisted diagnostics and existing clinical decision support systems.
2. Interpretation and Reporting of Diagnostic Data
   • Automated generation of structured reports from radiology, pathology, and cardiology interpretations.
   • Summarization and translation of complex diagnostic information into clinician-friendly formats.
   • Extraction of structured findings from free-text clinical notes for diagnostic validation.
3. Patient-Centered Diagnostic Applications
   • LLM-enabled tools for explaining diagnostic results to patients in plain language.
   • Enhancing patient understanding of imaging, laboratory, and pathology reports.
   • Evaluating patient engagement and comprehension when interacting with LLM-generated diagnostic explanations.
4. Workflow Efficiency in Diagnostic Medicine
   • Reducing documentation burden through automated note-taking, coding, and preliminary summarization.
   • Applications in prior authorization, referral coordination, and test result communication.
   • Operational impacts of LLMs on diagnostic throughput and turnaround time.
5. Research and Data Curation for Diagnostics
   • LLM-assisted systematic reviews and meta-analyses related to diagnostic methods.
   • Mining biomedical literature for diagnostic biomarker discovery or test validation.
   • Synthesizing evidence to support diagnostic guideline development.
6. Validation, Ethics, and Implementation
   • Addressing biases, inequities, and fairness in diagnostic LLM applications.
   • Explainable AI approaches that ensure diagnostic model interpretability.
   • Regulatory and privacy considerations (e.g., HIPAA, GDPR) when deploying LLMs in clinical diagnostics.

Impact and Significance

This Special Issue will provide a timely and focused platform for presenting cutting-edge research on the application of LLMs in medical diagnostics. It aims to bridge the gap between computational innovation and clinical practice by highlighting studies that emphasize validation, usability, and integration into real-world diagnostic pathways. By convening research from AI experts, clinical diagnosticians, laboratory scientists, and regulatory scholars, the Special Issue will contribute to establishing best practices and guiding the responsible adoption of LLMs in diagnostic medicine.

Prof. Dr. Tim Duong
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Diagnostics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• large language models
• diagnostic
• prognosis
• decision support
• clinical practice