Health Language Analytics (HLA) is about to complete the first stage of a contract to use its natural language processing (NLP) technology, Horizon, to extract and codify cancer data from pathology text reports for the California Cancer Registry (CCR) in a project that CEO Jon Patrick says Australian cancer registries should follow to improve their cancer surveillance programs.

The Sydney-based firm is also using its scanning technology as a service with the NSW Cancer Registry to identify tumour streams in cancer records that have been faxed, using a mixture of optical character recognition (OCR) and a spelling correction engine to return performance figures of over 92 per cent.