Research for Pre-Med High School Students: What Future Doctors Should Do

The best research for a student aiming at medicine is a focused study of one clinical question, such as how to detect a disease earlier or predict which patients will respond to a treatment, carried out with public health data, published studies or computational methods. A first-author paper on a medical question, finished before college, shows the scientific thinking medical schools look for and gives you a head start most pre-med students do not have.

This guide is for students who already know they want to be doctors. It covers what kind of research helps, how to pick a clinical question, methods you can carry out without a hospital lab, where student medical papers get published, which competitions fit, and the medical papers YRI Fellows have written. If you are comparing programs, see our list of medical research programs for high school students. For the lab and bioinformatics side, see our biomedical research guide.

Why research matters for a future doctor

Shadowing, volunteering and hospital service show that you care about patients. Research shows something different: that you can think like a scientist about medicine.

The AAMC, the organization behind the MCAT and the main US medical school application service, lists core competencies for entering medical students. Several are thinking and reasoning skills that research builds directly:

Scientific Inquiry: applying the scientific process to solve problems and form research questions and hypotheses
Quantitative Reasoning: working with numbers and data
Critical Thinking
Written Communication

A paper you wrote at 16 is not a substitute for college research. It is a foundation. Students who start early arrive at college knowing how to read a study, design an analysis and write it up, which makes them stronger candidates for university labs.

What kind of research helps most

Research type	What it shows	Reachable before college?
Disease detection and screening	You can turn data into a tool that finds disease earlier	Yes
Outcome and risk prediction	You understand which patients face which risks	Yes
Medical genetics	You can interpret variants and their effects	Yes, with public databases
Public health	You think about populations, not just patients	Yes
Evidence synthesis	You can read and weigh many studies	Yes
Wet-lab or bedside clinical research	Hands-on lab or patient work	Rarely, and usually through a hospital program

The common thread: pick a question a doctor would care about. "Can we screen for anemia from a routine blood test?" reads as medicine. "I tested five machine learning models" does not.

Example research questions for pre-med students

Diagnosis and screening:

Can routine complete blood count values estimate anemia severity, and which values drive each prediction?
Can a model estimate endometriosis risk from symptoms and history alone, without surgery?
Can low-cost ultrasound plus deep learning screen for several conditions in clinics without specialists?

Prediction and treatment:

Can a model predict which melanoma patients will respond to immunotherapy, and flag when its own prediction is unreliable?
Do the risk factors that predict heart disease differ between women and men?

Genetics and public health:

Which missense variants in genes linked to an inherited heart disease are likely to cause disease, based on where they sit in the protein?
How do rates of a chronic condition differ across age or income groups in a national health survey?

Methods a student can actually carry out

1. Analyze public health and clinical datasets

National health surveys such as NHANES, run by the CDC, and many open research datasets let students study disease patterns and risk factors. Some clinical databases, such as MIMIC-III on PhysioNet, are restricted: users must become credentialed, complete human research protection training, and sign a data use agreement, and approval takes at least a week. Plan for that before you choose a dataset.

2. Build and test screening or prediction models

Machine learning on public clinical data is the most common route to a first medical paper. What separates strong work is honesty about failure: calibration, uncertainty, and whether the model catches the patients who matter most, not just its overall score.

3. Interpret genetic variants

Public databases such as ClinVar, from the US National Library of Medicine, collect reported genetic variants and their clinical significance. A student can study a set of variants in a disease gene and test a method for telling harmful from harmless.

4. Analyze medical images

Public imaging collections, such as The Cancer Imaging Archive, let students study CT, MRI and other scans. Projects range from texture features that predict survival to models that classify disease.

5. Synthesize the evidence

Systematic reviews and literature classification projects use PubMed and published trials to answer what the evidence says. These need care and structure, but no data collection.

6. Run a survey study

Public health surveys about sleep, nutrition or health knowledge can be done locally. Any study with human participants needs informed consent and review before data collection. ISEF publishes human participants rules that apply to affiliated fairs.

Medical research papers by YRI Fellows

These Fellows wrote first-author medical research with PhD mentors. More are on our published work page.

Nidhi K., an 11th grader, combined gut and skin microbiome data in an AI-driven multi-omics framework to find non-invasive biomarkers for PCOS and autoimmune disorders, with over 90% classification accuracy. Published at IEEE ICCCA 2025 and listed in IEEE Xplore. Read her story.
Diya A. built a digital twin of a patient from wearable sensors and health records to forecast adverse health events, evaluated on the MIMIC-III clinical dataset and a pilot study. Published in IEEE Xplore. Read her story.
Akhilesh K., an 11th grader, tested whether immunotherapy response models know when they are likely to be wrong, across two melanoma cohorts totaling 253 patients. First-author paper accepted at an IEEE conference. Read his story.
Harleen B., a 12th grader, found that a conventional endometriosis model scored a high AUC of 0.959 but caught none of the high-risk patients, then built an ensemble that reached an AUC of 0.990 and identified 4 of 5 high-risk patients in the held-out test set. Read her story.
Melissa N., a 9th grader, built an explainable anemia severity screening tool from standard complete blood count features. First-author paper accepted at an IEEE conference. Read her story.
Abhigna N. built ArogyaScan, an offline point-of-care screening system for rural clinics that reached 98.08% accuracy on lung classification and costs 70 to 80% less than conventional hospital ultrasound. Read her story.
Ajwa A., an 11th grader, interpreted missense variants in the five desmosomal genes linked to arrhythmogenic cardiomyopathy, an inherited cause of sudden cardiac death. Read her story.
Sia R. compared risk factor importance in women and men for atherosclerotic cardiovascular disease using open-access health data, accepted at an IEEE conference.
Shivi K. had an oral presentation and publication accepted at the 2026 Congress of Neurological Surgeons Annual Meeting, as the only presenter under 25, and was invited to give a research talk at Yale School of Medicine.
Sadaf S. was admitted to Brown and Johns Hopkins and was accepted to present AI-assisted ECG research at the 2027 Yale Undergraduate Research Conference. Read her story.

Where student medical papers get published

Biomedical engineering and medical AI conferences. IEEE runs many conferences on medical computing, including EMBC, and accepted papers appear in IEEE Xplore.

Medical imaging conferences. SPIE Medical Imaging accepts imaging research. YRI Fellow Ashwajit W. had a poster on CT-derived texture features in lung cancer accepted for SPIE Medical Imaging 2027.

Clinical society meetings. Specialty meetings, such as the CNS Annual Meeting, accept abstracts and presentations. These usually need a clinical angle and strong methods.

Student journals. The Journal of Emerging Investigators publishes hypothesis-driven research by students before university.

Medical research at science competitions

Regeneron ISEF has several categories for medical work:

ISEF category	Fits projects on
Translational Medical Science	Disease detection and diagnosis, prevention, treatment, drug testing
Biomedical and Health Sciences	Physiology, genetics of disease, immunology, nutrition, pathophysiology
Biomedical Engineering	Devices, sensors and imaging, biomaterials, biomechanics
Computational Biology and Bioinformatics	Computational epidemiology, pharmacology, genomics

JSHS judges a research talk followed by questions, good practice for anyone who will one day present cases.

Common mistakes pre-med students make with research

Counting shadowing as research. Shadowing is valuable, but it is observation, not research.
Choosing a cure as the goal. Aim for one clear, testable contribution.
Overclaiming. A model that works on one dataset is not a diagnostic tool. Say what it does and where it was tested.
Ignoring rare cases. In medicine, the patients a model misses often matter most.
Skipping ethics and data rules. Follow consent and data use requirements from day one.

Frequently asked questions

Do medical schools care about high school research?

Medical school applications focus on college years. High school research matters because it builds the skills medical schools list, such as scientific inquiry and quantitative reasoning, and it strengthens undergraduate applications, including to research universities.

Can I do medical research without hospital access?

Yes. Public datasets, genetic databases, imaging archives and published studies cover most student medical research. Every Fellow above worked without a hospital lab.

Is research or clinical experience more important for pre-med?

Both matter, for different reasons. Clinical experience shows commitment to patients. Research shows scientific thinking. A pre-med student benefits from having both.

What is the best pre-med research topic?

One where you can name the patient group, the question and the data. Disease screening and risk prediction are common first topics because public data exists.

Can I get a recommendation letter from research?

Yes. A research mentor who watched you work can write a specific letter. See how to get a research letter of recommendation.

How the YRI Fellowship helps future doctors publish

The YRI Fellowship pairs you 1:1 with a PhD mentor in a medical or biomedical field for 10 weeks of weekly sessions, about 5 to 8 hours of work a week. Your mentor helps you pick a clinical question, find appropriate data, and design an analysis that holds up. An editing team works through every draft, and a publishing team gets the paper into IEEE, Springer or Elsevier journals and conferences, with free help until it is published.

87% of Fellows publish and 73% place at science competitions. ISEF and JSHS preparation and a mentor letter of recommendation are included. Fellows have been admitted to Brown, Johns Hopkins, Stanford, Caltech and Harvard. No research experience is needed, and decisions come within 48 hours.

Tuition is $2,997, or 3 payments of $999, with scholarships considered case by case. See pricing and more student stories. Families focused on Ivy League admissions can also look at the Top 1% Profile, a 6 to 9 month program for 8 families a year.

Sources

SHARE THIS ARTICLE

X LINKEDIN EMAIL