emtelligent’s™ proprietary, enterprise-scale emtelliPro™ NLP engine processes all types of medical text with high precision and recall, aided by our deep learning models that parse out the often-confusing text of medical reports.
Our deep learning-based emtelliPro™ NLP engine is custom-built for clinical use by our CEO, a practicing radiologist, our CTO, a PhD and full professor in NLP, and a team of PhDs and Masters graduates.
emtelliPro extracts medical entities using multiple ontologies (e.g. MEDCIN, SNOMED, RadLex, RxNorm, HGNC, and the NCI Thesaurus), and custom ontologies can also be used
Feature extraction includes report sentences, segments, experiencers, negation, uncertainty, measurements, time expressions, and higher-level features such as follow-up
recommendations. Additional features, assertions, and relations are undercontinuous development
emtelliPro is offered as either a HIPAA-compliant cloud solution, or an appliance for a healthcare institution’s data centre
SDKs and clients are available in C#, PHP, Python, Java, and custom clients can be easily created to access the emtelliPro API
For text-based reports with a limited number of features, millions of reports/day can be processed on a single server instance. Cluster processing is available for rapid processing of tens-to-hundreds-of-millions of reports/day