Feed it to your favorite LLM

Guide

AI visibility for heart surgeons

Short answer

The one sampled answer named no heart surgeon and no Sao Paulo hospital. It judged past-year volume, results adjusted for risk, and a heart team that weighs open surgery against a catheter procedure. We counted all 3 raw files from 2026-10-14 by hand.

The AI Visibility Index: heart surgeons

The batch asked three buyer questions on 2026-10-14 for Sao Paulo, Brazil. One question asked for the best heart surgeons in Sao Paulo. One asked how to choose. One asked how open valve replacement compares with a catheter procedure, including the hospital stay. The engines were claude-personal, twice, and codex, once. The list file and the valve file failed to authenticate after the sign-in session expired. Both failures were claude-personal. The only live answer was the codex guide on how to choose. That guide assumed a US reader, then asked which country applies. Counts below use all 3 probe files. Every source below is 1 of 3.

Who the models name

No probe file named a heart surgeon, a surgical group, or a Sao Paulo hospital. The live answer offered tests, not names.

Where the models send clients

Each source below appeared in 1 of 3 probe files, all in the live answer.

  • The American Board of Thoracic Surgery, for a US chest-surgery certificate, plus an active state license.
  • STS public reporting, for hospital or group results on common operations. Reporting is voluntary. A missing report does not prove poor quality. A group rating is not one surgeon's rating.
  • A joint ACC, STS, and SCAI statement. The link path includes 1 May 2026. The answer used that statement for a heart team that weighs surgery and a catheter for valve disease.
  • The STS risk calculator.

The live answer also asked for an active state medical license, with no state named, because the answer never reached Sao Paulo. Reviews, awards, and hospital ads appeared only as weak grounds for a choice. No file named Google, a society in Brazil, or a Sao Paulo hospital report.

What the models judge on

Every point below appeared in 1 of 3 probe files.

The live answer split the work by operation. Bypass, mitral valve repair, aorta surgery, and congenital heart surgery were different skills. Ask how many operations like this one the surgeon did in the past year, at a similar level of complexity. That recent count mattered more than a lifetime total.

For a US reader, the answer wanted a thoracic surgery certificate and an active state license. The certificate, in that answer, is a baseline, not proof of better results.

Results had to be readable. The answer listed death, stroke, kidney failure, major infection, and a repeat operation. Ask also for the patient count, the years covered, and an adjustment for how sick patients were.

A useful consult explains the reason for treatment, the alternatives, the timing, the benefit, and the personal risks. For valve disease, the answer wanted a heart team that considers open surgery and a catheter procedure. Ask who runs the cardiac ICU, who covers a complication overnight, and how follow-up and urgent calls work after discharge.

Avoid a guarantee, a zero-risk claim, and a success percent with no definition, no count, and no follow-up. Avoid a sales pitch centered on robotic or minimally invasive surgery. Ask why that approach suits the patient's anatomy, and whether it changes durability or the chance of another procedure. Pressure without a clinical reason was a reason to leave. Real urgency includes what happens if the patient waits.

The answer closed with six questions for each consult. One asks who performs the key steps, and who gives care afterward. One asks whether an STS risk calculator applies, and what the calculator misses. A second opinion at another experienced center helps when advice conflicts, if the wait is safe.

The valve file returned no text. This batch has no hospital-stay number and no Sao Paulo fee.

How to improve your AI visibility as a heart surgeon

AI access

Publish a plain text page for each surgeon with the operations, the past-year count for each, and the hospital.

AI business profile

State the board and the country of the license. The live answer checked the American Board of Thoracic Surgery for a US reader, then asked which country applies. Put the past-year count for the exact operation next to the group result, and say the group score is not your score.

Identity tags

List bypass, mitral valve repair, aorta surgery, congenital heart surgery, and catheter valve work as separate services. The live answer treated those as different skills. Name who runs the cardiac ICU.

Answer bank

Publish your own open-surgery versus catheter comparison, with the stay in days, the heart team, and the risks of death, stroke, and a second operation. Date the page.

Google profile

This sample never cited Google. The live answer said not to choose mainly from reviews, awards, or ads. Name the surgeon and the hospital. Do not write a zero-risk line.

Listing consistency

Match name, board, and hospital to any public report you cite. The live answer said STS public reporting covers the hospital or the group, not one surgeon. Label a license in Brazil and a foreign board as two facts.

Freshness and measurement

The live answer linked a statement whose address includes 1 May 2026. When that statement is revised, update your heart-team page to match.

Heart Surgeons-specific plays

  • Publish past-year volume for the exact operation, at a similar complexity. The live answer preferred that count over a lifetime total.
  • Show results adjusted for how sick the patients were, with the count and the years. The live answer said a bare success percent means little without a definition, a count, and follow-up.
  • For valve disease, name the heart team and both paths. The live answer wanted a heart team that considers both.
  • Write who does the key steps and who covers the cardiac ICU at night. The live answer told patients to ask both.
  • Drop guarantee language. If you offer robotic or minimally invasive surgery, explain why it suits a given anatomy and whether it changes durability or the chance of another procedure. The live answer told patients to ask exactly that when a pitch centers on the label.

How this index is made

We asked buyer questions to AI assistants through command-line engines on 2026-10-14. Two of 3 files failed to authenticate, both from claude-personal, so the list and the stay question produced no text. The only live answer came from codex and named no surgeon in Brazil. All 3 probe files, including the 2 failures, are stored raw in a public dataset, and only 1 holds a substantive answer. This page is one snapshot of a thin sample, and later waves will ask again. The Index methodology page describes the full method.

About Vizaeo

Vizaeo is an AI visibility service for law firms. It measures whether ChatGPT, Claude, Gemini, Grok, and Perplexity name your firm when clients ask for a lawyer, then applies the fixes that make a firm readable and quotable by AI assistants, with proof of each change. The Vizaeo sprint is $499 for 14 days of done-for-you fixes, with re-measurement at 30, 60, and 90 days against your baseline. If we miss the fix list, you get a refund.