Are these topics actually novel?

The AI attempts to find novel intersections of your specified field and current trends. However, you MUST conduct a thorough literature search on the suggested topics to confirm true novelty before proceeding.

Why do I need to input "Available Resources"?

A great idea is useless if you cannot realistically execute it. By telling the AI what data, time, or funding you have, it will suggest projects matched to your actual capacity — for example, avoiding a 10-year prospective cohort if you only have 6 months.

Can I ask for topics in a very obscure sub-field?

Yes. The more specific your interests input, the more tailored the output will be. The AI's knowledge is limited to its training data, so for very recent sub-fields, verify that the gaps it identifies are still open.

How many topics will I receive?

The tool generates 3–5 topics by default. Quality over quantity: each topic is accompanied by a feasibility note and novelty check. If none fit, describe your constraints more precisely and run again.

My institution lacks access to large patient databases — are there topics suitable for smaller datasets?

Yes. Specify your actual data access in Available Resources (e.g., 'retrospective chart review, approximately 200 patients'). The AI will prioritize study designs — such as case-control studies, pilot feasibility studies, or secondary analyses of existing datasets — that are appropriate for smaller samples.

Can I use this for basic science (non-clinical) research?

This tool is optimized for clinical and translational research topics. For purely basic science questions, the output quality may be lower, as the AI's gap identification works best when grounded in clinical outcome data and study design constraints.

Research Topic Suggestions -- medPrompt

Usage Guide

Fill in your field, research interests, and career stage.
Click AI Run — receive a ranked list of promising research topics with feasibility notes.
Results appear in the chat — explore any topic further through follow-up questions.

Medical Research Assistant

Fill in variables and run directly with AI

Wiki

Choosing a viable research topic is one of the most consequential decisions a clinician-researcher makes. A well-chosen topic matches scientific novelty with practical feasibility: it addresses a genuine knowledge gap, fits within the investigator's available resources and timeline, and aligns with local institutional priorities. Poor topic selection — choosing ideas too broad, too dependent on unavailable infrastructure, or already addressed in recent high-powered studies — wastes years of effort and reduces the probability of publication.

This tool generates 3 to 5 tailored research topic suggestions by combining three inputs: the investigator's clinical department and sub-specialty context, their stated research interests, and a realistic account of available resources (data access, funding, staff, time horizon). The AI cross-references these constraints against known trends and identified gaps in the biomedical literature to propose projects that are both scientifically meaningful and concretely executable within the described conditions.

Each suggestion is presented with a brief rationale explaining why the topic represents a genuine opportunity, a preliminary feasibility note describing potential study designs and data sources, and a novelty check indicating how the proposed angle differs from existing published work. This structured format allows investigators to compare options objectively rather than choosing based on vague intuition.

The tool is particularly valuable for early-career clinicians preparing their first independent research project, residents or fellows drafting research proposals for fellowship applications, and established clinicians expanding into new sub-specialties who need a rapid landscape assessment of where productive work remains.

Topics generated by this tool are proposals, not guarantees. Before investing significant effort in any suggestion, investigators should conduct a targeted PubMed search to confirm that no recent high-quality study has already addressed the proposed question. The AI's training data has a knowledge cutoff, meaning very recent publications may not be reflected in its gap assessment.

Use the PICO Question Builder on this platform to convert a promising topic into a formal, structured research question once you have selected your preferred direction.

FAQ