We developed a new technique to mathematically transform a peripheral artery pressure (PAP) waveform distorted by wave reflections into the physiologically more relevant aortic pressure (AP) waveform. First, a transfer function relating PAP to AP is defined in terms of the unknown parameters of a parallel tube model of pressure and flow in the arterial tree. The parameters are then estimated from the measured PAP waveform along with a one-time measurement of the wave propagation delay time between the aorta and peripheral artery measurement site (which may be accomplished noninvasively) by exploiting preknowledge of aortic flow. Finally, the transfer function with its estimated parameters is applied to the measured waveform so as to derive the AP waveform. Thus, in contrast to the conventional generalized transfer function, the transfer function is able to adapt to the intersubject and temporal variability of the arterial tree. To demonstrate the feasibility of this adaptive transfer function technique, we performed experiments in 6 healthy dogs in which PAP and reference AP waveforms were simultaneously recorded during 12 different hemodynamic interventions. The AP waveforms derived by the technique showed agreement with the measured AP waveforms (overall total waveform, systolic pressure, and pulse pressure root mean square errors of 3.7, 4.3, and 3.4 mmHg, respectively) statistically superior to the unprocessed PAP waveforms (corresponding errors of 8.6, 17.1, and 20.3 mmHg) and the AP waveforms derived by two previously proposed transfer functions developed with a subset of the same canine data (corresponding errors of, on average, 5.0, 6.3, and 6.7 mmHg). arterial tree; blood pressure; generalized transfer function; model; wave reflection SINCE ITS INTRODUCTION by O'Rourke and coworkers in 1993 (3), the generalized transfer function has received attention for providing a convenient and safe means for monitoring central aortic pressure (AP) by mathematical transformation of a peripheral artery pressure (PAP) waveform. The basic premise of the transformation is that a single, universal transfer function exists that can faithfully relate the PAP waveform to the AP waveform of all individuals for all time. However, the transfer function linking PAP to AP would ideally be able to adapt to the intersubject and temporal variability of the arterial tree due to, for example, age-related arterial compliance differences, disease-induced peripheral resistance variations, baro-and thermoregulatory modulation of peripheral resistance in response to physiological perturbations, and therapeutic administration of vasoactive agents. To this end, Sugimachi et al. (12) and Westerhof et al. (18) previously proposed a technique to partially adapt the transfer function by defining it through an arterial tube model with a personalized value for a model parameter reflecting the wave propagation delay time and population averages for the remaining parameters. We recently introduced (13, 14) perhaps the first entirely adaptive techn...

Fundamental to robotics is the debate between model-based and model-free learning: should the robot build an explicit model of the world, or learn a policy directly? In the context of HRI, part of the world to be modeled is the human. One option is for the robot to treat the human as a black box and learn a policy for how they act directly. But it can also model the human as an agent, and rely on a "theory of mind" to guide or bias the learning (grey box). We contribute a characterization of the performance of these methods under the optimistic case of having an ideal theory of mind, as well as under different scenarios in which the assumptions behind the robot's theory of mind for the human are wrong, as they inevitably will be in practice. We find that there is a significant sample complexity advantage to theory of mind methods and that they are more robust to covariate shift, but that when enough interaction data is available, black box approaches eventually dominate.Index Terms-theory of mind, inverse RL, model-based RL, model-free RL, sample complexity

We have developed a new technique to estimate the clinically relevant aortic pressure waveform from multiple, less invasively measured peripheral artery pressure waveforms. The technique is based on multichannel blind system identification in which two or more measured outputs (peripheral artery pressure waveforms) of a single-input, multi-output system (arterial tree) are mathematically analyzed so as to reconstruct the common unobserved input (aortic pressure waveform) to within an arbitrary scale factor. The technique then invokes Poiseuille's law to calibrate the reconstructed waveform to absolute pressure. Consequently, in contrast to previous related efforts, the technique does not utilize a generalized transfer function or any training data and is therefore entirely patient and time specific. To demonstrate proof of concept, we have evaluated the technique with respect to four swine in which peripheral artery pressure waveforms from the femoral and radial arteries and a reference aortic pressure waveform from the descending thoracic aorta were simultaneously measured during diverse hemodynamic interventions. We report that the technique reliably estimated the entire aortic pressure waveform with an overall root mean squared error (RMSE) of 4.6 mmHg. For comparison, the average overall RMSE between the peripheral artery pressure and reference aortic pressure waveforms was 8.6 mmHg. Thus the technique reduced the RMSE by 47%. As a result, the technique also provided similar improvements in the estimation of systolic pressure, pulse pressure, and the ejection interval. With further successful testing, the technique may ultimately be employed for more precise monitoring and titration of therapy in, for example, critically ill and hypertension patients.

We developed a technique to calculate forward and backward arterial waves from proximal and distal pressure waveforms. First, the relationship between the waveforms is represented with an arterial tube model. Then, the model parameters are estimated via least-squares fitting. Finally, the forward and backward waves are calculated using the parameter estimates. Thus, unlike most techniques, the arterial waves are determined without a more difficult flow measurement or an experimental perturbation. We applied the technique to central aortic and femoral artery pressure waveforms from anesthetized dogs during drug infusions, volume changes, and cardiac pacing. The calculated waves predicted an abdominal aortic pressure waveform measurement more accurately (2.4 mmHg error) than the analyzed waveforms (5.3 mmHg average error); reliably predicted relative changes in a femoral artery flow measurement (14.7% error); and changed as expected with selective vasoactive drugs. The ratio of the backward- to forward-wave magnitudes was 0.37 ± 0.05 during baseline. This index increased by ∼50% with phenylephrine and norepinephrine, decreased by ∼60% with dobutamine and nitroglycerin, and changed little otherwise. The time delay between the waves in the central aorta was 175 ± 14 ms during baseline. This delay varied by ±∼25% and was inversely related to mean pressure.