I use machine learning techniques in decision exploration systems. I am particularly interested in using rule-based association mining techniques to allow visual exploration of risk and certainty in decision making systems. Application areas are varied, but much of my work has been biophysical: electromyographic (EMG) based muscular disease characterization, EMG and postural fatigue and pain prediction, and EMG and force based characterization of sleep apnea. By combining rule based systems with statistical certainty, models of decision confidence can be created to allow contingent decision planning and exploration. I am primarily interested in exploring and visualizing these sorts of data domains. In addition, I have an interest in physiological models, such as my EMG simulator, for use in providing gold standard data for validation of EMG based techniques.
high-risk decision making
J. Yousefi and A. Hamilton-Wright (2014), “Characterizing EMG Data using Machine-Learning Tools,” Computers in Biology and Medicine, 51C, pp. 1–13. DOI: 10.1016/j.compbiomed.2014.04.018
R. Varga, S. M. Matheson and A. Hamilton-Wright (2014), “Aggregate Features in Multi-Sample Classification Problems,” IEEE Journal of Biomedical and Health Informatics, Open Access. DOI: 10.1109/JBHI.2014.2314856
J. P. Saboisky, D. W. Stashuk, A. Hamilton-Wright, A. L. Carusona, L. M. Campana, J. Trinder, D. J. Eckert, A. S. Jordan, D. G.McSharry, D. P. White, S. Nandedkar,W. S. David and A.Malhotra (2012), “Neurogenic Changes in the Upper Airway of Obstructive Sleep Apnea Patients” American Journal of Respiratory and Critical CareMedicine, 185(3), pp. 322–329. DOI: 10.1164/rccm.201106-1058OC PMID: 22016445
A. Hamilton-Wright, L. McLean, D. W. Stashuk and K. M. Calder (2010), “Bayesian aggregation versus majority vote in the characterization of non-specific arm pain based on quantitative needle electromyography,” Journal of NeuroEngineering and Rehabilitation, 7(8), Open Access. DOI: 10.1186/1743-0003-7-8
A. Hamilton-Wright and D.W. Stashuk (2005), “Physiologically based simulation of clinical EMG signals,” IEEE Transactions on Biomedical Engineering, 52(2), pp. 171–183. DOI: 10.1109/TBME.2004.840501