How TRT Plus Analyzes Your Bloodwork: A Deterministic Engine, Then AI

Quick answer

TRT Plus splits bloodwork analysis into two jobs. First, a deterministic engine does the reasoning: it scores every marker against personalized, TRT-calibrated ranges, computes a weighted Biomarker Score, and runs your cardiovascular and metabolic markers through a cascade that sets your individual hematocrit threshold and testosterone ceiling. Then, and only then, AI takes that finished, structured result and turns it into your 8-section Bloodwork Report in plain English. The math is deterministic and repeatable. The AI writes it up. Same labs in, same analysis out.

Two jobs, done by the right tool

There are two very different kinds of work hiding inside "analyze my bloodwork," and most of the value is in not confusing them.

The first is reasoning: what does this number mean, against the right range, in the context of every other marker, and where does it sit relative to a ceiling you're managing to? That work needs to be deterministic, testable, and bounded. The same inputs should always produce the same result. That's traditional engineering.

The second is interpretation: taking that result and explaining it in language you can actually act on. That work needs to be fluent, adaptive, and readable. That's what modern AI is genuinely great at.

TRT Plus keeps these separate. The engine decides. The AI explains. You don't want the probabilistic part making the clinical judgment, and you don't want a rigid script writing your report, so each tool does the job it's actually suited for.

What the engine does before the AI sees anything

This is the part that matters most: in TRT Plus, the AI never touches your raw labs. The engine processes them first, deterministically and the same way every time, and only its structured output is handed off. Every score, range, and ratio is produced by a programmatically engineered model, not approximated by a language model. And the models and ranges themselves come from deep analysis of de-identified, real-world TRT clinic data. That's the foundation under every threshold, score, and ceiling the engine applies. That pre-processing is the product:

Personalized reference ranges. Every marker is scored against TRT-calibrated ranges built for someone actually doing the work. In-range for the general population, much of which is prediabetic, overweight, and inflamed, is the wrong target for a man on TRT who trains and eats with intent. Your ranges should reflect the room at Dick's Sporting Goods, not the checkout line at Walmart.
A weighted Biomarker Score. Markers are weighted by clinical importance and read as relationships and ratios (testosterone to estradiol, testosterone to SHBG), not isolated values, and composited into one score you can track over time.
The Cardio-Metabolic Cascade. Six cardiovascular and metabolic markers — blood pressure, ApoB, CRP, body fat, HbA1c — feed into your hematocrit threshold, which in turn sets your personalized testosterone ceiling. Lean and metabolically healthy earns a higher ceiling; rising risk tightens it.
Androgenic Load (A:E). Your androgen-to-estrogen picture accounts for DHT's outsized receptor potency and non-linear amplification, not a flat testosterone-to-estradiol number.
Zakharov Free T. Free testosterone is solved with a model that holds up at TRT and supraphysiological levels, where the standard population formula drifts.

By the time the AI is involved, the judgment is already made. Every number has been scored, every range personalized, every ceiling set, deterministically and identically on every run.

The cascade in action: hematocrit

Hematocrit is the clearest example of why cascading matters, because a single number means very different things depending on what surrounds it.

Your hematocrit isn't read alone. It's gated by the cardio-metabolic cascade. If your blood pressure, ApoB, CRP, body fat, and HbA1c are all in range, the engine recognizes a lower-risk profile and applies your earned threshold. As each of those markers slips into amber and then red, the ceiling tightens progressively. Every downgrade in the cascade pulls your hematocrit threshold, and the testosterone ceiling sitting above it, down another step. Your acceptable HCT isn't a fixed line. It's earned, and it moves with the rest of your cardiovascular picture.

The same inputs always produce the same read, which is exactly what lets you compare this quarter to last and trust that any change came from your blood, not the tool.

Then the AI: your 8-section report

Now the AI goes to work on the engine's output, not your raw labs. It takes the scored markers, the personalized ranges, the ratios, and the ceilings the engine produced and writes your 8-section AI Bloodwork Report. It opens with your recommendations: the top 3–5 highest-priority actions, synthesized from the entire analysis, with each recommendation spelling out the lifestyle, supplement, and medicinal strategies to get it done. From there it moves into the deeper sections: your Anabolic Potential Rating, Hormonal Balance, Responder Type, a Cardiovascular Risk Profile, and a Healthspan Assessment.

The AI's job here is interpretation and language: narrating the engine's math in a way you can read and act on. It doesn't invent the verdict. It explains one that's already been made.

Why this produces consistent, personal results

Because the reasoning lives in the engine, the same labs always produce the same analysis. That's what makes tracking actually work. When a marker moves from one draw to the next, you know it's your blood changing, not the model's mood. And because your ranges and ceilings are set to you rather than a population average, the read is personal: scored against the standard you're actually holding yourself to.

A deterministic foundation, an AI that speaks plain English. The complexity is ours. The clarity is yours.

Frequently asked questions

Does TRT Plus actually use AI?

Yes, for what AI is genuinely great at: interpretation and language. The scoring, the personalized ranges, and the clinical logic are the engine's job, and that part is deterministic.

How is this different from asking a chatbot to read my labs?

A general model gives you a fresh interpretation each time and leans on population reference ranges by default. TRT Plus fixes the ranges and the scoring in the engine first, so the analysis is personalized and repeatable, then uses AI to write it up. The difference isn't the AI; it's the engineered model in front of it.

Can I see how the engine works?

Yes. The schematic, the formulas, and the cited studies behind each model are laid out in the app and on the engine page. See the engine.

Does this replace my healthcare professional?

No. It gives you a consistent, contextualized read so the conversation with your healthcare professional starts from solid ground. It doesn't make the decision for you.

Do other TRT apps do this?

Hell no. Most apps advertising "AI bloodwork analysis" are wrapping the same general model you already have access to. You could dump your labs into Grok, type "analyze this," and get the same kind of answer, with the same inconsistency. The engine is the hard part, and it's the part that's missing. A deterministic framework built on real-world TRT clinic data is what actually turns your labs into a personalized, repeatable read. And it's what you're paying for here.

This article is for informational purposes only and does not constitute medical advice. Consult a qualified healthcare professional before making any changes to your protocol.

See the engine for yourself

Every score, range, and ceiling in TRT Plus comes from a deterministic model. You can read the actual schematic and formulas. See the engine · Get TRT Plus.