Homeo AI vs Isabel DDx: LLM Differential or Case Copilot?

Quick answer

Choose Isabel DDx Companion when a registered healthcare professional needs a conventional differential-diagnosis safety net that expands possible diseases from presenting features and, on the Premium plan, compares selected diagnoses with an LLM layer. Choose Homeo AI when a qualified homoeopathic practitioner already has an approved case narrative and needs to review how it becomes symptoms, proposed rubrics, source passages, a remedy differential, and a follow-up rationale.

As reviewed on 2 August 2026, Isabel lists Standard at $149 per year and Premium at $199 per year, with unlimited access across multiple devices. The official page also lists a $49 yearly student or resident offer, an emerging-economy discount, and a free 30-day trial. Confirm the live country, tax, eligibility, renewal, and checkout terms because the dollar and euro offers are not interchangeable.

Premium adds Isabel's LLM interface for diagnosis explanation, comparison, and quick summaries. The important boundary is that Isabel's established engine generates the differential first. The LLM is then used to explain and compare selected items. Homeo AI serves a different specialist reasoning job and should not be evaluated as if it were a conventional disease generator.

What changed in Isabel's Premium workflow

Isabel's dated 22 October 2025 announcement introduced a ChatGPT integration for clinicians. The practitioner enters age, sex at birth, travel history, and presenting clinical features. Isabel produces its differential, after which the user can select between two and five diagnoses for AI-generated explanations and a side-by-side comparison of symptoms, investigations, treatments, and prognosis.

The release said the ChatGPT features were desktop-only and English-only at launch. The current pricing page now says the LLM features are available on mobile, while still describing the interface as English-only. That is a meaningful product change. Buyers should follow the current page for availability but preserve both the dated release and the accepted checkout terms, then verify mobile access during the trial.

The Standard plan links diagnoses to sources such as Merck Manual and PubMed and includes mobile access without a separate app. Premium keeps those links and adds the explain-and-compare functions. Isabel also lists a free 30-day trial without presenting it as clinical validation for a specific practice.

These are vendor claims and current commercial facts. They do not establish diagnostic accuracy, safety, or benefit for a particular clinic. A current product page also contains a 96 percent top-ten claim. Before using that number, inspect the cited validation, population, outcome definition, product version, and whether the tested workflow matches the clinic's setting. A marketing percentage is not a local acceptance test.

Where Isabel is genuinely strong

Isabel is designed to broaden a conventional differential when a presentation is unclear, atypical, or at risk of premature closure. Its product page describes more than 10,000 conditions, all ages and specialties, red-flag or don't-miss highlighting, medication-cause lookup, multiple languages, and links to evidence resources. It can accept free text or selected clinical features and can be integrated into an electronic medical record through an API.

That workflow has a clear purpose. A clinician who has considered only two common causes can use a structured disease generator to ask whether another plausible or dangerous alternative deserves examination. The Premium comparison can help organize the differences between a small set of candidates before the clinician opens the cited source or local guideline.

The output should be treated as a prompt for verification, not a diagnosis. Isabel's terms describe the service as decision support and state that it does not provide medical advice, diagnosis, or treatment. The clinician still decides which possibilities fit the patient, which source applies, which investigation or referral is required, and what belongs in the record.

Where Homeo AI fits differently

Homeo AI should preserve the words and structure of a homoeopathic case, propose symptoms and rubrics with visible support, let the practitioner correct them, compare a remedy differential, and keep the reviewed rationale available at follow-up. Its useful output is a traceable specialist reasoning record, not a disease list.

The error checks are therefore different. In Isabel, ask whether the entered conventional features are accurate, whether a must-not-miss alternative appears, whether the evidence link supports the comparison, and whether a local guideline changes the next step. In Homeo AI, ask whether chronology, causation, modalities, negation, and practitioner corrections remain visible from narrative to rubric and remedy comparison.

Neither product should silently supply inputs to the other. A disease on an Isabel list is not a homoeopathic rubric. A Homeo AI remedy differential cannot exclude a conventional diagnosis or delay investigation, referral, or emergency assessment. If both are used, keep the source records and review responsibilities explicit.

Price is only one procurement layer

The individual pricing page makes Isabel easier to budget than enterprise products that require a sales call. Standard is currently $149 yearly and Premium is $199 yearly. A practitioner should still calculate the time needed to enter features, review the list, open sources, correct the record, train staff, and document the final reasoning. A low annual fee is poor value if the workflow adds unchecked suggestions or is used outside its intended scope.

Check whether the account is for an individual or an organization, who can access saved queries, what data is retained, which subprocessors receive it, whether submitted information is used for model improvement, how deletion and export work, and which agreement applies in the clinic's jurisdiction. The public terms were last updated on 25 March 2024. Ask for the controlling current privacy, security, and contractual material before any identifiable information is entered.

Homeo AI requires the same procurement discipline. Use the minimum necessary data, approve the account and data path before clinical use, restrict access, and preserve the authoritative record in the clinic's approved system. A statement that a tool is built for healthcare does not complete this review.

Run a reproducible clinic test

Prepare twelve completed, de-identified cases with conventional and homoeopathic answer keys written before testing. Include common and uncommon presentations, a follow-up, an explicit negative, a medication effect, contradictory timing, incomplete information, a case where no narrow diagnosis is justified, and a case that requires urgent conventional assessment. Remove names, exact dates, addresses, contact details, identifiers, and rare combinations that could identify anyone.

For Isabel Standard, score feature-entry fidelity, presence of the answer-key differential, useful don't-miss prompts, irrelevant suggestions, evidence-link quality, time to independent verification, mobile consistency, and whether a second clinician can reproduce the review. For Premium, select the same two to five diagnoses every time and score whether the comparison preserves uncertainty, supports each statement, and changes appropriately when one input is corrected. Do not score fluency as accuracy.

For Homeo AI, use only the checked case narrative. Score preservation of wording, symptom extraction, rubric mapping, source traceability, correction effort, remedy-differential clarity, and follow-up continuity. Change one disputed rubric and verify that the new result remains understandable.

Keep separate scorecards. The tools answer different questions, and combining their results into one accuracy percentage would hide the failure mode of each. Neither tool replaces diagnosis, referral, emergency care, practitioner judgment, prescription, potency, dosage, or follow-up responsibility. Neither tool should make an emergency-care decision.

Conditional verdict

Choose Isabel Standard when the measurable need is fast conventional differential expansion with source links and the clinic does not need LLM comparisons. Choose Premium when a controlled trial proves that comparing two to five diagnoses saves verified review time, the mobile behavior matches the current product page, and the account terms fit the data path.

Choose Homeo AI when the clinic already has an approved narrative and needs inspectable homoeopathic case reasoning with practitioner-controlled mappings and a follow-up trail. Judge it on narrative fidelity, sources, corrections, and reproducibility.

Use both only with a documented boundary. Isabel can support conventional diagnostic breadth and safety checking. Homeo AI can support specialist case analysis. The practitioner remains responsible for conventional assessment, source verification, referral, prescribing, and the final record.

Official sources checked

• Isabel official individual pricing
• Isabel official October 2025 LLM release
• Isabel official DDx Companion product page
• Isabel official terms of use

Product facts reviewed 2026-08-02. Confirm current features, prices, source libraries, and terms with the vendor.