EarlyTracks is a university spin-off (CENTAL, UCL) incorporated in 2009. Our first efforts were driven by R&D needs in NLP and semantic technologies. In 2014, we started our commercial developments in Enterprise Search and purpose-built application for the media and financial sectors.
In early 2016, we decided to focus our efforts on only 1 single sector: healthcare. This decision was mainly based on the potential added-value of our past & current efforts and the purpose we found in improving care through technology.
Over the last decade, Medical ICT has turned paper records into digital contents. EPRs gained in maturity and information became more accessible.
Unfortunately, EPR are too frequently perceived as administrative / compulsory tasks by practitioners and medical information quality remains sub-optimal. This is a major issue for care institutions as the promise of automation in this domain can bring disruptive innovation: data is the new oil.
Fundamental trends emerge to address this concern but there is still a long way to go... A care institution cannot achieve this on its own.
EarlyTracks is a technology company: we develop solutions to address specific issues we face to achieve our mission of organizing medical records. Over the years, EarlyTracks (and its sister organization, CENTAL) has built powerful solutions and expertise in the fields of Medical Text Mining and Semantic Technologies. We use a mix of advanced technologies in AI (machine learning, ao.) and formal knowledge representation (semantic web technologies) mixed with domain standards for interoperability (SNOMED CT and HL7 in particular).
WHERE DO WE START?
Having a technological edge is one thing.... Effectively addressing customers concerns is another. Over the years, we have built (and continue to build with our partners) an approach to address the foundational issues in knowledge management.
Medical records (document) base
Quality is fundamental to build strong business cases. This is why EarlyTracks has focused most of its expertise on beyond state-of-the-art NLP capabilites, Having a full control on these technologies enable us to adapt our performance to the specific needs of a customer.
Text Mining To perform our medical records management, we start with the main medical assets available: patient reports. From these reports, we extract useful information using NLP...
Patient (data) graphs
Natural Language Processing
WHY DOES MEDICAL RECORDS MANAGEMENT MATTER?
It is critical for many aspects of hospital management and objectives. At EarlyTracks, we have listed and organised more than 40 different types of applications that globally address 4 types of aims (some inspired of IHI triple aim)...