Now in private beta
Turn audiograms into
structured data, instantly.
The open infrastructure layer for hearing loss research. Upload audiogram images, get clean structured JSON. Power the pipelines accelerating the cure.
POST https://api.auriflow.io/v1/digitize
Now in private beta
Accepting design partners in hearing loss research. Gene therapy trials. Cochlear implant outcome registries. Cisplatin ototoxicity monitoring programs.
The problem
Hearing research is
stuck in 1998.
The biology has found its footing. Gene therapy trials are showing real results. But the data infrastructure hasn't kept up.
01 —
Audiograms are images, not data
Most audiograms still live as image files in EMRs. Research pipelines start with slow manual transcription.
02 —
No standard data model
Genomic, audiology, and treatment data live in separate systems with no shared schema or interoperability.
03 —
Cross-study signal detection is manual
Teams still email spreadsheets. Promising patterns across studies are missed because infrastructure is fragmented.
04 —
Gene therapy trials lack real-time pipelines
OTOF and GJB2 trials generate real data, but signal gets buried in generic clinical trial tooling.
05 —
Compliance slows every data handoff
PHI-safe workflows and auditability are mandatory, yet most teams rebuild compliance controls for each study.
06 —
Time-to-analysis is measured in weeks
By the time data is cleaned and structured, early insights are delayed. Research needs machine-ready output fast.
How it works
One API call.
Structured hearing data.
Upload an audiogram image — scanned, photographed, or PDF — and receive clean, FHIR-compatible JSON in seconds.
01
Upload audiogram image
POST any audiogram — scanned paper, photograph, PDF report, or S3 URI. We handle every format and orientation automatically.
02
ML pipeline extracts thresholds
Our YOLO-based symbol detection identifies every audiological marker — air conduction, bone conduction, masked and unmasked — with 96%+ accuracy.
03
Receive FHIR-ready structured JSON
Poll for results or receive a webhook. Clean structured data with frequency, threshold, ear, and classification — ready for your pipeline.
04
Stream into your research pipeline
Integrate with your existing stack. Export to Kafka, push to your data warehouse, or query longitudinally across patient cohorts.
import auriflow client = auriflow.Client("coch_sk_...") job = client.digitize.create( file=open("audiogram.png", "rb") ) result = client.digitize.wait(job.id)
→ response (illustrative — IDs/URLs redacted)
{
"job_id": "aud_01…",
"status": "complete",
"confidence_score": 0.97,
"processed_at": "2026-05-01T14:32:38Z",
"audiogram": {
"right_ear": {
"air_conduction_unmasked": [
{ "frequency_hz": 500, "threshold_db_hl": 25.0 }, …
], …
},
"left_ear": { … }
},
"classification": {
"hearing_loss_type": "sensorineural",
"severity": "moderate_to_severe",
"configuration": "sloping"
},
"warnings": [],
"metadata": { "run_id": "example-001" },
"raw_image_uri": null,
"error": null
}
Features
Everything your research
pipeline actually needs.
Built by engineers who've run mission-critical data pipelines at scale. Production-hardened from day one.
96%+ extraction accuracy
YOLO-based symbol detection handles every audiological marker across scanned, photographed, and digital formats.
FHIR R4 native output
Every result maps directly to FHIR R4 DiagnosticReport resources. Drop into any EHR or research data warehouse without custom ETL.
Kafka-ready streaming
Results publish to your Kafka topics in real time. Avro schemas included. Plug directly into your existing data pipeline without modification.
HIPAA-compliant by default
Encryption at rest and in transit. Immutable audit logs. BAA available. Built for institutions where compliance isn't optional.
Processing in seconds
17x faster than manual transcription. Submit a batch of 1,000 legacy audiograms and have structured data before lunch.
Longitudinal cohort queries
Query across patient timelines. Track threshold progression. Correlate with genomic and treatment data. The cross-study API your research needs.
Pricing
Start free.
Scale as you discover.
Usage-based pricing that grows with your research program. No seats. No surprises.
Research
$0
/ month, forever
- 100 audiograms / month
- REST API access
- JSON output
- Community support
- Python SDK
Most popular
Growth
$299
/ month
- 10,000 audiograms / month
- FHIR R4 native output
- Webhook delivery
- Longitudinal cohort API
- Kafka streaming integration
- Email support
Enterprise
Custom
volume pricing
- Unlimited audiograms
- HIPAA BAA included
- Dedicated infrastructure
- 99.9% SLA
- On-premise deployment
- Dedicated engineer support
Get started
The data infrastructure
the cure deserves.
Join researchers and biotech teams using Auriflow to accelerate hearing loss research.