Automatic Indexing of Handwritten Medical Forms for Search Engines

A new paradigm, which models the relationships between handwriting and topic categories (denoted as ‘concepts’), in the context of medical forms, is presented. The ultimate goals are (i) the recognition of medical handwriting, and (ii) the use of such information for a medical form search engine. Medical forms have diverse, complex and large lexicons consisting of English, Medical and Pharmacology corpus. This technique shows that a handwriting recognition engine, with just a few recognized characters, can be used to represent a medical concept. This allows (i) a reduced lexicon to be constructed, thereby improving the performance of handwriting recognition engines [6][21], and (ii) unseen PCR forms to be tagged with a concept and later searched. Both practical and theoretical numbers are reported. This research builds the notion of a ‘computational semantic lexicon’ which was vaguely introduced in our IWFHR 2002 paper [15] and incorporates other research in the area of call-routing [2][3].