Patents, on the record.
I am a co-inventor on a system that helps AI detect disease in 3D medical images while protecting privacy and accountability. The lead application, No. 19/200,539, was allowed by the USPTO. A second application, No. 19/199,034, is in examination.
What it does
The system takes in real medical imaging across MRI, CT, PET, ultrasound, and histopathology, cleans and standardizes it, then uses what we call probabilistic masking to focus on the regions most likely to be medically meaningful, the way an experienced radiologist knows where to look first.
Those regions are turned into compact, privacy-preserving representations rather than raw identifiable images, and can be combined with the text of the associated report into a single richer view. A radiologist stays in command, and their feedback is used to improve the system over time. Every step is encrypted, access-controlled, and written to a tamper-evident record, so the system can be audited.
Why it matters
The hard part of medical AI is rarely the model. It is doing the work without exposing patients and without creating a black box no one can explain later. This invention is a blueprint for doing it responsibly: focused, private, improvable, and accountable. That is the combination that lets this kind of technology near a real patient, and it sits exactly where my two fields meet, healthcare AI and cybersecurity. There is a full, plain-English walkthrough in what the patent actually does.
U.S. Patent App. No. 19/200,539
AI-Driven System and Methods for 3D Medical Image Disease Detection. Allowed by the USPTO, 12 claims. Co-inventor. Grant pending.
U.S. Patent App. No. 19/199,034
Related AI medical-imaging methods. In examination.
The story behind the patents.
What the Patent Actually Does
People ask me what the patent is "for." The short version is that it is a way to let AI help read 3D medical scans without giving up privacy, accountability, or the doctor's judgment. The longer version is worth a few minutes, because the meaning is in how the pieces fit together.
Working With a Patent Attorney for the First Time
I have spent my career making technical calls under pressure. Co-inventing a patent taught me a different discipline: saying exactly what you mean, and being able to prove it.
From CTO to Co-Inventor
I have built and secured a lot of systems. Co-inventing one was a different thing entirely, and it pulled together everything I had learned in two careers I used to think of as separate.
Probabilistic Masking, Explained Without the Jargon
The simplest way I can describe the core idea behind our patent is this: we taught the system where to look before we asked it what it saw.
Privacy-Preserving Diagnostics: AI on Medical Images Without Exposing the Patient
The hardest problem in medical AI is not the AI. It is the data. This is the part of our patent that comes straight out of my security career.
What "Allowed" Actually Means
People hear "patent" and assume there are two states: you have one or you do not. There is an in-between, and it is the moment that actually matters.
Why a Human Stays in the Loop
The most important design decision in our system is not a clever algorithm. It is the choice to keep a person in charge and to let the machine learn from them.
Building AI You Can Audit
There is a question every organization using AI in a serious setting will eventually face, usually at the worst possible moment. When a decision is challenged, can you reconstruct exactly what happened and who shaped it? If the answer is no, you do not really have a system. You have a liability.
How AI Is Changing Radiology
Radiology is where AI in medicine has gone furthest, and where the hype and the reality are easiest to confuse. Here is what is actually happening, from someone building in the space.
Can You Patent an AI Algorithm?
This is one of the most common questions I get now that I have a patent in AI. The short answer is that you usually cannot patent the math itself, but you can patent a specific, novel system that puts it to work. The distinction is the whole game.
AI and HIPAA: Using Patient Data Without Breaking the Rules
You cannot build useful medical AI without medical data, and medical data is some of the most regulated information there is. Here is how those two facts are reconciled, from a security perspective.
Why Medical AI Needs Security, Not Just Accuracy
Every medical AI pitch leads with accuracy. Almost none of them lead with security. That is exactly backwards if the goal is to put the thing near a real patient.
3D Medical Imaging, Explained
Before you can understand why AI in medical imaging is hard, you have to understand what a 3D scan actually is. It is not a picture. It is a volume, and that changes everything.