A micropower adaptive linear transform vector quantiser

Micropower signal classification and compression are becoming important req uirements for implantable cardioverter defibrillators although they are cur rently limited by power and computational constraints. This paper describes an integrated circuit that facilitates Intracardiac Electrogram (ICEG) cla ssification and compression of 30 dimensional analogue vectors while consum ing a maximum of 2.5 mu W power for a heart rate of 60 beats per minute (1 vector per second) from a 3.3 V supply. This represents a significant advan ce on previous work which achieved ultra low power supervised morphology cl assification [6] since the templated matching scheme used in this chip enab les unsupervised blind classification of abnormal rhythms and the computati onal support for low bit rate date compression. The adaptive template match ing scheme used is to tolerant to amplitude variations, and inter- and intr a-sample time shifts. Micropower performance is achieved using CMOS analogu e circuits biased in weak inversion in order to minimise energy per computa tion. Results from the fabricated chip demonstrate the impact of the amplit ude and shift tolerance on ICEG data and performance for blind classificati on of an abnormal rhythm in five heart patients. For four out of the five p atients, no false negative classifications and a worst case of 11% false po sitive classifications were made.