Precision OCR

The out of the box, cloud-native solution that takes even the messiest and illegible documents such as faxes, pdfs, and clinical notes and extracts medical meaning

Structured data enables more informed decision making

Start your free trial today.

Leveraging the cloud, computer vision, machine learning and natural language processing to efficiently abstract and structure data from faxes and clinical notes.

Built for Clinical Data

Clinical Data Extractor


PrecisionOCR Workflow

From Bulk Documents to Relevant Medical Meaning Within Minutes

PrecisionOCR is end-user configurable to support disease and use case-specific entity recognition with an output to FHIR. Just add documents. Once the data is structured, it is converted to FHIR resources that can be used in the Precision Health Cloud.


Superior OCR/NLP Service

PrecisionOCR leverages proprietary computer vision denoising algorithms to reduce visual artifacts and improve OCR. It provides auto rotation, date and timeline analysis and machine learning based spellcheck using MIMIC-III data set. The contents of the document can easily be searched to identify, understand and extract data from forms and tables.

Built-In Ontologies

Quickly surface a list of medications, conditions, procedures and observations within a large document. Once detected, PrecisionOCR can turn these observations and procedures into medical codes using public ontologies like LOINC and SNOMED that can be exported or used within the Precision Health Cloud.

PHI Redaction

PrecisionOCR can mask all HIPAA protected Patient Health Information. This allows you to focus on the information you need most while protecting subject privacy.

No Up Front Costs

Create an account today that includes a BAA (Business Associate Agreement) with no upfront costs. With PrecisionOCR you can simply upload documents via your web browser for a low per-page price. No software to install.

Innovative Search Capability

Using a hybrid machine learning and rules-based system, search and filter across all documents by keyword as well as define and maintain domain specific keywords to be surfaced within the document or clinical note. FHIR resources will be intelligently suggested based on keywords for indexing data from the document, making search functionality more robust and flexible.

Document Classification

Organize documents in folders by patient (name or MRN) or by custom category. PrecisionOCR is machine learning based document classifier trained using the MIMIC-III data set. With this function you can get insight into what each page consists of without looking at it yourself and sort to your preference.

Automated Lab Report Extraction

Harness the power of the PrecisionOCR library of pre-built data extractors to expedite repetitive data extraction from lab reports.

Powerful Analytics

Combine your data from stacks of paper with other data sources to create beautiful visualizations in just minutes leveraging the integration with our Precision Health Cloud.

Held To The Highest Security Standards


Create Your Account Today

Note: All data is synthetic and does not contain protected health data