Systems and methods for simultaneous single particle tracking, phase retrieval and PSF reconstruction
View Patent ↗3D particle tracking and localization provide direct means to monitor details within nano-scale environments. However, a major shortcoming of 3D techniques is the sample induced aberrations due to inhomogeneous refractive index, resulting in distortion of point spread functions (PSFs), which are an important measurement tool required for these tasks in the field. This issue is particularly important when using pre-calibrated PSFs that do not take into account the sample induced aberrations. A system incorporates a Bayesian framework for simultaneous particle tracking and PSF inference directly from a given data. The system is data efficient by taking into account existing sources of uncertainty, such as uncertainty in the shape of the PSF, which is often ignored. The system is benchmarked using a wide range of synthetic and experimental data.
1 . A system, comprising:
a processor in communication with a memory, the memory including instructions executable by the processor to:
access observation data including brightness data indicative of one or more light-emitting particles captured across a plurality of frames and across a plurality of planes by an imaging device, the plurality of frames having an aberration profile observable across each frame of the plurality of frames;
sample a set of joint probability values associated with observing the observation data for values of each respective parameter of a plurality of parameters of a measurement model, the plurality of parameters including:
a particle trajectory for each respective light-emitting particle of the one or more light-emitting particles across the plurality of frames; and
a set of point spread function parameters of a point spread function of the imaging device;
and
jointly infer, based on the set of joint probability values and the observation data, a set of most probable values of the plurality of parameters; and
sample probabilities associated with a background photon count per pixel for each frame of the plurality of frames using a Metropolis-Hasting procedure at each iteration of a Markov Chain Monte Carlo procedure.
2 . The system of claim 1 , the set of point spread function parameters including:
an amplitude and a phase of a pupil function associated with the aberration profile and the point spread function of the imaging device.
3 . The system of claim 1 , the plurality of parameters further including one or more of:
a diffusion coefficient;
a particle photon emission rate; and
a background photon count per pixel.
4 . The system of claim 1 , the memory including instructions executable by the processor to:
apply the Markov Chain Monte Carlo procedure to iteratively sample probability values associated with values of each respective parameter of the measurement model over a plurality of iterations.
5 . The system of claim 1 , the memory including instructions executable by the processor to:
sample probabilities associated with an amplitude and a phase of a pupil function over the plurality of frames from respective amplitude and phase posterior probability distributions using the Metropolis-Hasting procedure at each iteration of the Markov Chain Monte Carlo procedure.
6 . The system of claim 5 , wherein the amplitude and phase posterior probability distributions are respectively obtained through application of Gaussian priors on the amplitude and phase of the pupil function.
7 . The system of claim 1 , the memory including instructions executable by the processor to:
sample probabilities associated with particle trajectories over the plurality of frames from a particle trajectory posterior probability distribution using a hit-and-run sampler at each iteration of the Markov Chain Monte Carlo procedure.
8 . The system of claim 1 , the memory including instructions executable by the processor to:
sample a probability associated with a diffusion coefficient directly from a posterior probability distribution of the measurement model at each iteration of the Markov Chain Monte Carlo procedure.
9 . The system of claim 1 , the memory including instructions executable by the processor to:
sample probabilities associated with a particle photon emission rate from a light-emitting particle of the one or more light-emitting particles for each frame of the plurality of frames using the Metropolis-Hasting procedure at each iteration of the Markov Chain Monte Carlo procedure.
10 . The system of claim 1 , where the measurement model simultaneously considers each frame of the plurality of frames.
11 . A method, comprising:
accessing observation data including brightness data indicative of one or more light-emitting particles captured across a plurality of frames and across a plurality of planes by an imaging device, the plurality of frames having an aberration profile observable across each frame of the plurality of frames;
sampling a set of joint probability values associated with observing the observation data for values of each respective parameter of a plurality of parameters of a measurement model, the plurality of parameters including:
a particle trajectory for each respective light-emitting particle of the one or more light-emitting particles across the plurality of frames;
a set of point spread function parameters of a point spread function of the imaging device; and
jointly inferring, based on the set of joint probability values and the observation data, a set of most probable values of the plurality of parameters; and
sampling probabilities associated with a background photon count per pixel for each frame of the plurality of frames using a Metropolis- Hasting procedure at each iteration of a Markov Chain Monte Carlo procedure.
12 . The method of claim 11 , the plurality of parameters further including one or more of:
a diffusion coefficient;
a particle photon emission rate; and
a background photon count per pixel.
13 . The method of claim 11 , further comprising:
applying the Markov Chain Monte Carlo procedure to iteratively sample probability values associated with values of each respective parameter of the measurement model over a plurality of iterations.
14 . The method of claim 11 , further comprising:
sampling probabilities associated with an amplitude and a phase of a pupil function over the plurality of frames from respective amplitude and phase posterior probability distributions using the Metropolis-Hasting procedure at each iteration of the Markov Chain Monte Carlo procedure.
15 . The method of claim 14 , the amplitude and phase posterior probability distributions being respectively obtained through application of Gaussian priors on the amplitude and phase of the pupil function.
16 . The method of claim 11 , further comprising:
sampling probabilities associated with particle trajectories over the plurality of frames from a particle trajectory posterior probability distribution using a hit-and-run sampler at each iteration of the Markov Chain Monte Carlo procedure.
17 . The method of claim 11 , further comprising:
sampling a probability associated with a diffusion coefficient directly from a posterior probability distribution of the measurement model at each iteration of the Markov Chain Monte Carlo procedure.
18 . The method of claim 11 , further comprising:
sampling probabilities associated with a particle photon emission rate from a light-emitting particle of the one or more light-emitting particles for each frame of the plurality of frames using the Metropolis-Hasting procedure at each iteration of the Markov Chain Monte Carlo procedure.