Frequently Asked
Questions.

MINA is a clinical decision support platform built from scratch for Turkish clinical practice. On top of an open-weight large language model, the Turkish clinical language was internalized via Continued Pre-Training (CPT), specialty expertise was added through LoRA adapters, and the model was connected to live clinical literature through a GraphRAG architecture.

No. MINA was not built with classical instruction fine-tuning but with a three-layer process: first CPT (Continued Pre-Training) instilled the Turkish clinical language into the model; then a dedicated LoRA adapter was trained for each specialty and task; finally GraphRAG added live information retrieval. Compared to classical fine-tuning, this approach yields more durable language mastery and a more modular expertise layer.

Both are Opinion AI platforms and both run on the same MINA. TIS (hospital platform) is built for the hospital operator and physician; it covers pre-authorization, the patient digital twin, ICD-10, discharge summaries, MMIS inventory and audit simulation. SIT (insurance platform) is built for the insurance specialist and manager; it offers file assessment, reimbursement compliance, policy coverage checks, fraud risk screening and portfolio-level actuarial audit. Data remains within each institution; the two platforms are connected via a shared bridge protocol.

MINA is the clinical decision support agent that works alongside the user (ICD-10, reimbursement compliance, discharge summaries, file evaluation, etc.). MINA Cockpit is the management/portfolio-level simulation layer of the same MINA: for hospital managers, it simulates Ministry of Health audits via BH-13/32/33/34/35 and ADSH-13/14 indicators; for insurance managers, it produces an actuarial portfolio view and above-threshold alerts.

MINA is a decision supporter, not a decision maker. All critical actions (approve, missing, reject, prescribe) are taken by the user. AI outputs are presented with rationale and source; for each prediction, source text and confidence score are visible. Thanks to the GraphRAG architecture, answers are always tied to reference sources; unverifiable claims are not produced.

Opinion AI does not collect, process or store any patient or clinical data. All data remains on the institution's own servers, under its own control; no data flows to our environment. Under KVKK, the role of "data controller" stays with the institution — we are a technology provider. MINA computations run on the institution's infrastructure (on-premise or in an isolated cloud). All connections are HTTPS, storage is AES-256 encrypted. The platform holds KVKK, GDPR and ISO 27001, 27701, 9001, 14001, 45001 certifications.

No. TIS is an intelligent layer that runs on top of your existing HBIS. It reads from the HBIS, does not write back to it; the integration is non-disruptive. Connection is via XML/CSV export or a direct database connection.

Standard integration (HBIS export + data normalization + secure publishing) is completed within 1-2 weeks. The first MINA Cockpit report arrives on day one; sensitivity tuning matures over a few weeks.

No. The first phase is operator- and finance-team-oriented. Physicians start using MINA for discharge summaries, ICD-10 suggestions and clinical decision support whenever they want, integrating into their own workflows.

Because the system works on a document and process basis, it is compatible with all private health insurance companies and SGK. Insurance-specific rule sets can be defined.

Yes. SIT works independently as well; even if the hospital sends documents to SIT by PDF or email, MINA reads the documents and produces the same synthesis. If TIS is also in use, the integration eliminates manual steps entirely.

No. MINA only produces suggestions; rejection, approval and requests for missing information are always made by the specialist. The rationale and references (clinical guidelines, reimbursement clauses, policy terms) produced by the agent are presented to the specialist as decision support.

Yes. Per-policy benefit scope, waiting periods, contracted institution lists and custom rule sets are defined per company. MINA applies these rules automatically in every file evaluation.

Only the data specific to a given application is exposed to the specialist who can evaluate that application. KVKK-compliant role-based access control is applied; not everyone sees everything. Access to other patient files held in the hospital's TIS is not possible from the insurer side.

MINA's "Fraud Risk Analysis" query works not only at the single-file level but also at the portfolio level: recurring diagnosis patterns, hospital–physician pairings, atypical procedure–policy relationships and duplicate billing signals are systematically scanned. Suspicious patterns surface in MINA Cockpit's daily report as above-threshold alerts.