Lung Function

We are working with the Woolcock Institute to develop improved anomaly detectors for breathing cycles using the forced oscillatory technique [1]. This improves the quality of the data collected and hence the utility of the technique.

[1] Thuy T. Pham, Cindy Thamrin, Paul D. Robinson, Alistair L. McEwan, and Philip H.W. Leong. Respiratory artefact removal in forced oscillation measurements: A machine learning approach. IEEE Transactions on Biomedical Engineering, 2016. In press.


We developed analog VLSI chips for the low power classification of abnormal heart rhythms from intracardiac leads [1]. The scheme combined morphology and timing information [2] to achieve better accuracy than previous techniques. The feasibility of this approach for implantable defibrillator devices was demonstrated.

[1] P. H. W. Leong and M.A. Jabri. A low power VLSI arrhythmia classifier. IEEE Transactions on Neural Networks, 6(6):1435–1445, November 1995.

[2] P.H.W. Leong and M. Jabri. Matic – an intracardiac tachycardia classification system. Pacing and Clinical Electrophysiology (PACE), 15:1317–1331, September 1992.

[3] P.H.W. Leong and M.A. Jabri. A method and system for automatically classifying intracardiac electrograms. US Patent 5,280,792, January 25, 1994. University of Sydney.