Title :
Training Probabilistic VLSI models On-chip to Recognise Biomedical Signals under Hardware Nonidealities
Author :
Jiang, P.C. ; Chen, H.
Author_Institution :
Inst. of Electron. Eng., Nat. Tsing Hua Univ., Hsinchu
fDate :
Aug. 30 2006-Sept. 3 2006
Abstract :
VLSI implementation of probabilistic models is attractive for many biomedical applications. However, hardware non-idealities can prevent probabilistic VLSI models from modelling data optimally through on-chip learning. This paper investigates the maximum computational errors that a probabilistic VLSI model can tolerate when modelling real biomedical data. VLSI circuits capable of achieving the required precision are also proposed
Keywords :
VLSI; electrocardiography; feature extraction; medical signal processing; probability; system-on-chip; ECG; biomedical signal recognition; computational errors; feature extraction; hardware nonidealities; on-chip learning; probabilistic VLSI model; Bioinformatics; Biomedical computing; Circuit noise; Cities and towns; Hardware; Hydrogen; Neurons; Training data; USA Councils; Very large scale integration;
Conference_Titel :
Engineering in Medicine and Biology Society, 2006. EMBS '06. 28th Annual International Conference of the IEEE
Conference_Location :
New York, NY
Print_ISBN :
1-4244-0032-5
Electronic_ISBN :
1557-170X
DOI :
10.1109/IEMBS.2006.260401
