Dr Nigel Brown wrote: > The real test is, of course, not case notes but how it interfaces with a > real-time consultation process in a busy practice. It may be that a system > that learns each doctor's quirks and helps them with the case note process > while deducing the SNOMED-CT coding as an epiphenomenon will be the final > goal.
If mobile phones can do it (predictive text matching when entering SMS text in T9 mode), why can't medical record systems? There is sufficient processing power available on desktop computers these days to have very powerful predictive systems which can make seemingly instantaneous guesses on a keystroke by keystroke basis about what you actually mean as you type, automatically completing the word or concept (and storing the corresponding SNOMED CT concept ID behind the scenes), using the context (including the current patient's past history if that is known to the computer) offer the word or SCT concept most likely to be correct from a list of possibilities, and which can learn all sorts of personalised and idiosyncratic abbreviations and modes of expression. For voice recognition, substitute phonemes for keystrokes, but the principles are the same. Tim C _______________________________________________ Gpcg_talk mailing list [email protected] http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
