AI in Healthcare : The Perfect Passion Project for Gonzaga University Applicants
- BetterMind Labs
- 2 hours ago
- 7 min read
Most high school students applying to Gonzaga University have good grades. Many volunteer at hospitals. A surprising number shadow physicians. So what actually separates the applicants who get in from the ones who do not?
Here is the thing about elite admissions that nobody says out loud: the students who stand out are not the ones with the longest list of activities. They are the ones who built something real, something that shows they can think like a practitioner, not just a student. In 2026, for pre-med and health-science applicants, that something is an AI healthcare project. And once you understand why, you will not want to spend your summer any other way.
Why Gonzaga Specifically Rewards This Kind of Project
Gonzaga is a Jesuit institution. That matters more than people realize. Jesuit education is built on a specific idea: knowledge exists to serve others. Every applicant essay, every interview, every extracurricular is evaluated through that lens. What have you done with what you know? Who did it serve?
An AI project in healthcare answers that question directly. You are not checking a volunteering box. You are building a tool that could, in theory, help a real patient make a better decision about their health. That framing aligns perfectly with what Gonzaga's admissions team wants to see.
According to Gonzaga's own mission statement, the university seeks students committed to "using their talents in service of others." A machine learning model that predicts disease risk or flags early health warning signs is exactly that: talent, applied in service.
Beyond mission fit, Gonzaga's pre-med program is competitive. The Princeton Review consistently ranks it among the top pre-med programs in the Pacific Northwest. Students accepted into that track are expected to show scientific thinking before they arrive on campus, not after.
An AI healthcare project is proof that you already think that way.
What Makes a Healthcare AI Passion Project Actually Good to get in Gonzaga University
Not every project is equal. A lot of students "do AI" by running someone else's code on a Kaggle dataset, changing a few parameters, and calling it a project. Admissions officers are not naive. They have read enough essays to spot the difference between someone who built something and someone who copied something.
A strong AI healthcare project has a few specific qualities.
It starts with a real question. Not "can I train a model" but "what would actually help a patient or clinician?" That shift in framing changes everything about what you build.
It uses real or realistic data. Public datasets from the NIH, the CDC, or UCI Machine Learning Repository are perfectly legitimate. What matters is that you can explain where the data came from and why it is appropriate for the problem.
It produces something explainable. A model that spits out a prediction without any reasoning is a black box. A project worth writing about can explain its outputs. Why did the model flag this user? What features drove the prediction? That explainability is what makes the project clinically relevant, not just technically interesting.
It connects technical work to human impact. The best projects show you understand the stakes. Early detection of a chronic disease changes treatment options. A lifestyle recommendation engine, used consistently, could reduce hospital readmission rates. If your write-up does not connect the model to the patient, you have done only half the work.
Here is a list of beginner-friendly AI healthcare project ideas if you are still figuring out which direction to go.
The Admissions Math on AI Projects
Let's be direct about something. A passion project only helps you if you can talk about it coherently. And you can only talk about it coherently if you actually built it, understood it, and iterated on it.
This is where most self-directed projects fall apart. A student reads a tutorial over spring break, builds a basic classifier, then tries to write a 650-word Common App essay about it. The essay comes out thin. The supplemental short answers are vague. An interviewer asks one follow-up question and the whole thing unravels.
The students who do this well spent real time on their projects. They hit roadblocks and debugged them. They made design decisions and can explain why. They know what the model is good at and where it breaks down. That depth shows in every part of the application, not just one essay.
This is also why the recommendation letter matters so much. A mentor who supervised your project can speak to your process, your intellectual curiosity, your persistence when things did not work. That is a qualitatively different letter than "Priya volunteered at our clinic and was always punctual."
Siddhant Iyer: What This Looks Like in Practice
Abstract advice is easy to give. So let's look at what one student actually built.
Siddhant Iyer, a high school student, spent his summer building an AI-powered disease predictor and lifestyle suggestion tool through BetterMind Labs. The goal was straightforward: help users identify potential health risks early, before symptoms become serious.
The system predicts the likelihood of five conditions based on user input: arrhythmia, diabetes, cancer, asthma, and obesity. Under the hood, it uses machine learning models trained on health data. On the surface, users get something simple and useful: a risk assessment and a set of personalized recommendations generated through Google's Gemini AI.
What makes this project strong from an admissions standpoint is not the technical stack. It is the framing. Siddhant understood the clinical problem before he wrote a line of code. Chronic conditions like diabetes and asthma often go undiagnosed until they have already caused significant damage. Early detection is not a nice-to-have. It is, in some cases, the difference between manageable treatment and irreversible progression.
His documentation captures that clearly. The project explanation does not just describe what the model does. It explains why early detection matters, who benefits, and what limitations exist. That is the writing of someone who thought carefully about the problem, not someone who stitched together a tutorial.
That project is now portfolio-ready. It can anchor a Gonzaga supplement response. It gives his recommenders something specific to point to. And it gives him a story that is genuinely his, not a version of what ten thousand other applicants are submitting.
Where to Build This Kind of Project
Most students who want to do this well need structured support. Not because they lack ability, but because building a real project requires feedback loops that do not exist in self-directed learning. You need someone to tell you when your model is overfitting. You need a deadline that forces you to finish, not just tinker indefinitely. You need a mentor who has seen what actually works in admissions documentation and can help you frame your project appropriately.
BetterMind Labs runs four-week online cohorts specifically designed for this. The mentorship ratio is 1:3, which means you get real feedback, not a weekly group webinar where you are one of sixty students. Students come out with working projects, documented capstones, and recommendation letters from mentors who supervised the actual work.
The projects students build include healthcare prediction systems, machine learning pipelines, and AI dashboards that are deployment-ready, not just proof-of-concept demos. That difference matters for college applications and, more importantly, for what you actually learn.
If you are serious about going into medicine, public health, or bioinformatics, this is the kind of summer that moves the needle.
Frequently Asked Questions
Can I build an AI healthcare project without coding experience? Yes, with the right support structure. Most students start with basic Python, and a mentor-guided program will take you from fundamentals to a working model inside four weeks. The key is having someone to remove blockers quickly, so you are not spending three days stuck on a data preprocessing error.
Will this kind of project actually help my Gonzaga application? Gonzaga's admissions process rewards intellectual depth and demonstrated service orientation. An AI healthcare project checks both boxes if it is well-documented and you can speak to it fluently. The strongest applications show a through-line between a student's interests and their actual work, and this kind of project creates exactly that.
Can I do this on my own, or do I need a program? You can start on your own, but the students who produce the strongest work almost always had structured mentorship. Building alone means no feedback on your approach, no accountability milestones, and no one to help you frame the project for admissions purposes. A program with a low mentorship ratio gives you the advantages of a supervised research experience without requiring a university affiliation.
What makes BetterMind Labs different from other summer programs? Most programs expose students to AI concepts at a survey level. BetterMind Labs is built around individual project completion, which means you leave with a deployable, documented system, not just a certificate. The 1:3 mentor ratio and capstone documentation process are specifically designed to produce work that holds up in college applications and beyond.
The Honest Case for Doing This
Here is what I keep coming back to. The students who struggle in college admissions are rarely the ones who were not smart enough. They are the ones who stayed safe. They took the expected classes, did the expected activities, and wrote essays about the expected experiences.
Gonzaga, like most selective universities, has read thousands of versions of those applications. They are not bad applications. They are just forgettable.
An AI healthcare project is memorable because it is specific. It has a technical approach, a dataset, a set of outputs, a set of limitations, and a human story behind why it mattered to build it. That specificity is what makes essays vivid, interviews engaging, and recommendation letters credible.
If you are applying to Gonzaga with pre-med ambitions, the question is not whether you should build this kind of project. It is whether you are going to put in the work to build it well.
The students who do that, the ones who can explain their model, defend their design choices, and connect their technical work to patient outcomes, those are the applicants that admissions teams remember.
Explore more on what makes a strong AI project at bettermindlabs.org.
Comments