Product Summaries

The natural language engine for clinical information has a number of features which are available in different combinations.

• SNOMED-CT with or without ICD/10 mapping
• User interface to assist in term selection and optimization of ICD-10 coding
• Post coordination of concepts - mappable to established reference sets or coding systems
• Bucket - collection of your specific terms, spellings and their integration into our natural language processor to meet your local situation

Smart Termer A clinical terminology engine (able to use any clinical terminology but currently established to operate with SNOMED-CT) providing context based natural language translation. Designed to give clinicians an easier option...

Smart Coder Built to support Inpatient morbidity coding, this is an ICD-10-AM implementation engine providing context based natural language translations to ICD-10-AM including coding algorithms and options, can easily be interfaced to a DRG Grouper and local systems.  Designed to support your coders and make their workload easier to handle...

SmartTermer/Coder Using the features of both systems you can collect clinical data using natural language and store snomed and icd codes into appropriate fields of your system. designed for you to reduce data coding time ...

Special Requests

If you have a terminology problemor need automated coding or terming contact us to discuss options on how we might be able to help -we are happy to assist. 

Key Features

This software was developed to support clinical care by a 'plug in' to existing source software to automate coding and reporting requirements without clinical input. Although there are many natural language processors ,this NLP is different due to it's intended design to fit a single purpose – support of clinical data capture and clinical care. Items in bold are unusual NLP features that, when taken together offer a unique solution.

• Allows clinicians to enter the information they need to record about the individual into the legacy or clinical information system.
• Supporting misspelling, personal and local abbreviations.
• Active learning – bucket of new/unrecognised words and terms for regular update and improvement.
• Will produce automatic post-coordination of qualifiers and additional clinical content eg: Left/Right, suspected.
• Reporting SNOMED-CT. Where a specific, limited SNOMED-CT reference term set has been established, the clinician can still enter the details they require and the system will return only those concepts acceptable to the term set. e.g. will remove left/right/suspected
• Collects information on the patient, data context (field/speciality of clinician) to improve word recognition and semantic processing and identification of the concept required.
• Dynamically Returns
• SNOMED-CT concept ID's into context. This feature allows the clinician to make a statement in detail and send it to the relevant field/s in the source system for storage / processing.
   Problem/disease/diagnosis
   Treatment
   Cause of Injury
   Medication (in development)
   Location of Event (injury)
   Activity
  The effect of this is that the clinician may mix concepts in one statement and the system will return them appropriately to the relevant field/s. eg: orif # humerus returns '# humerus' as a diagnosis/problem/disorder and the 'Open reduction and internal fixation of the humerus' as a treatment.
• ICD-10-AM (or other) codes into context.
   Identifies the ICD-10-AM code through application of age/sex and other coding rules
   Returns code/s to appropriate fields:
  
  • Diagnosis/Morphology
  • Cause of injury
  • Location of event (injury)
  • Activity
  • Procedure
• Removes codes not required for reporting in the context required.
• Assistance
• The system can be established to return a range of alternative codes/IDs or a place in the SNOMED-CT or ICD structure from which children can be selected to support user authentication and quality improvement. Used in initial testing and implementation, but rarely when established.