Extract hidden text information and build precise patient registries — quickly and easily! Regenstrief Institute has developed data analytics tools for population health that enable deep exploration of both unstructured textual data and structured discrete data to create precise patient cohorts.
nDepth™ incorporates natural language processing (NLP) capabilities and advanced text mining to query vast collections of free text data in order to qualify patients based on concepts derived from notes in the medical record. Paired with the Regenstrief Patient-List Generator (RPG), which searches limitless structured data points to identify populations that match specific criteria, the ability to query across the entire electronic health record significantly improves the accuracy and effectiveness of patient identification for research, quality assessment and improvement initiatives.
Why the rising interest in NLP?
“Up to 80% of clinical data is locked away in unstructured documents”
A wealth of useful data is buried in free text within electronic health records and other clinical documents. Unlocking the value of this information is key to quality improvement, research and outcomes analysis. Typically, highly qualified clinical staff must extract discrete information from patient charts manually, which is time-consuming and expensive. Getting this data efficiently and effectively requires faster and more scalable alternatives.
Built on the foundations of natural language processing and text mining, nDepth™ unifies vast collections of free-text data sources, creating a sophisticated clinical index. Built and fine-tuned by clinicians using the oldest and largest health information exchange in the country, nDepth renders complex search patterns as simple recipes in order to find more patients with higher accuracy and faster than other methods.
What is nDepth?
nDepth™ is a powerful tool to explore and derive meaningful insights from clinical text.
Using nDepth, the textual information buried within electronic health records is delivered to your fingertips.
Free text documents (such as physician notes, operative reports, pathology and radiology reports) contain a vast amount of important information that is inadequately captured in structured data (e.g., ICD-9 codes).
nDepth users can quickly and easily extract this data and other hard-to-find patient characteristics such as social behaviors, symptoms and family history from deep below the surface of electronic health data.
The goal of nDepth™ is to bring meaningful clinical content to your fingertips through simple exploration of rich data in order to improve patient care, clinical research, quality assessment and improvement initiatives.
To do this, nDepth indexes and searches vast collections of data for indicators hidden in free-text. These indicators, called phenotypes, are a set of characteristics that identify a specific condition or population.
Creating phenotypes starts with using clinical text search to easily locate rich data.
Expanding the results using synonyms quickly expands query results.
Advanced Text Analytics breaks down the textual data into usable insights. NLP identifies key information critical for context, like negation. nDepth helps to discover not only the key concepts within the sentence structure, but who it’s about (experiencer), when it happened (temporality), and whether it happened to the experiencer (negation).
So does accuracy.
nDepth uses a built-in validation module to create surveys for assessing your queries. Custom questions can be built in minutes and randomized reports are assigned to team members to validate accuracy. This is a key step in the iterative process of developing and refining phenotypes and nDepth makes this simple, and easy for all users.