21 Clinical logic and reasoning algorithm (CLARA): a whatsapp-based AI chatbot using artificial intelligence for precision anticoagulation in atrial fibrillation using a simulated and validated dataset
Bibliographic record
Abstract
Introduction The use of oral anticoagulation (OAC) in atrial fibrillation (AF) is often complicated by comorbidities such as renal dysfunction and bleeding risk. We developed and tested CLARA (Clinical Logic and Reasoning Algorithm), an AI-driven clinical decision support tool that uses a WhatsApp chatbot interface with a real-time Excel database (figure 1). Methods A simulated dataset of 150,000 AF patients was generated, each with full clinical profiles including age, sex, CHA2DS2-VASc components, HAS-BLED criteria, renal function (eGFR), weight, and previous anticoagulant exposure. A WhatsApp chatbot API was built to interact with healthcare providers, capturing structured inputs and relaying them in real time to a backend Excel-based clinical engine. The system matched responses with validated dosing algorithms (DOAC adjustment criteria) to recommend the precise agent and dose. To evaluate accuracy, CLARA was tested on a real-world dataset of 100 complex AF patients—preselected for challenging clinical profiles including renal impairment (eGFR <60) and high bleeding risk. The system’s OAC recommendations were compared against expert cardiologist consensus. Results CLARA achieved a sensitivity of 96% and a specificity of 94% in prescribing the correct agent and dose of OAC, based on gold-standard clinician judgement. Amongst patients with renal dysfunction (n=41), the algorithm correctly adjusted DOAC dosing in 39/41 cases. In patients with HAS-BLED ≥3 (n=36), CLARA accurately flagged bleeding risk and maintained guideline-appropriate prescribing. No inappropriate full-dose DOAC recommendations were issued in patients meeting criteria for dose reduction. The chatbot interaction averaged under 90 seconds per user. Conclusions/Implications CLARA demonstrates high sensitivity and specificity in providing individualized OAC recommendations in AF, including complex patients with renal impairment and bleeding risk. Its WhatsApp-based interface makes it highly accessible and adaptable in real-time clinical workflows. Further prospective validation is planned, but initial results support its potential as a scalable decision support tool for AF.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.010 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".