1 Quantitative Phase Imaging: Principles and Applications
17
where the equality holds for Gaussian distributions. This uncertainty principle does
not place a bound on the ability to make a measurement, but rather on the accuracy
with which we can define the field distribution as well as its frequency representation.
Equation 1.21 has direct consequences for imaging and scattering events, as it states
that the minimum resolvable feature on the object is governed by the “spreading” in
its k-vector domain, instead of the maximum value,
R ∝
2π
k
(1.22)
Thus, a generalized resolution definition can be achieved as inverse spatial bandwidth,
in all three directions.
1.6 Summary and Outlook
As we have seen here, QPI adds essential value to microscopy. The intrinsic
phase signal measured by QPI provides novel parameters to investigate physiological processes in biological samples. As discussed in the previous sections,
tremendous progress has been made to improve the spatial and temporal resolution of QPI systems. Temporally, since no exogenous labeling is needed, QPI
enables live sample measurements, lasting from the millisecond scale to days, even
weeks. QPI could measure samples at different spatial scales, ranging from subcellular structures below the resolution limit, to thick tissues, 100s of micrometers thick. These advances in instrument development offer exciting possibilities for numerous applications in biology and medicine. Looking ahead, we foresee a few directions of potential interest. Firstly, the benefit of imaging unlabeled specimens comes at the cost of losing specificity. Most existing QPI systems rely on correlative fluorescent tagging to identify the structure of interest.
One potential solution is to replace the fluorescent markers with scattering particles. By labeling structures of interest with nanoparticles with high refractive index,
QPI would become an ideal imaging tool with high specificity, but without photobleaching. Secondly, the possibility of endoscopic QPI is another interesting avenue
[36]. While most existing QPI systems measure samples in vitro or ex vivo, the
study of in vivo or deep tissue imaging with QPI is rarely explored [143, 144].
Thirdly, label-free super-resolution imaging remains a significant technical goal.
The existing super-resolution methods are typically involving fluorophores. Finally,
we anticipate that combining QPI and artificial intelligence approaches will lead to
advances in objective diagnosis. In the years to come, we anticipate that QPI will
continue gaining popularity as it transfers from engineering to biomedical laboratories [1].
17
where the equality holds for Gaussian distributions. This uncertainty principle does
not place a bound on the ability to make a measurement, but rather on the accuracy
with which we can define the field distribution as well as its frequency representation.
Equation 1.21 has direct consequences for imaging and scattering events, as it states
that the minimum resolvable feature on the object is governed by the “spreading” in
its k-vector domain, instead of the maximum value,
R ∝
2π
k
(1.22)
Thus, a generalized resolution definition can be achieved as inverse spatial bandwidth,
in all three directions.
1.6 Summary and Outlook
As we have seen here, QPI adds essential value to microscopy. The intrinsic
phase signal measured by QPI provides novel parameters to investigate physiological processes in biological samples. As discussed in the previous sections,
tremendous progress has been made to improve the spatial and temporal resolution of QPI systems. Temporally, since no exogenous labeling is needed, QPI
enables live sample measurements, lasting from the millisecond scale to days, even
weeks. QPI could measure samples at different spatial scales, ranging from subcellular structures below the resolution limit, to thick tissues, 100s of micrometers thick. These advances in instrument development offer exciting possibilities for numerous applications in biology and medicine. Looking ahead, we foresee a few directions of potential interest. Firstly, the benefit of imaging unlabeled specimens comes at the cost of losing specificity. Most existing QPI systems rely on correlative fluorescent tagging to identify the structure of interest.
One potential solution is to replace the fluorescent markers with scattering particles. By labeling structures of interest with nanoparticles with high refractive index,
QPI would become an ideal imaging tool with high specificity, but without photobleaching. Secondly, the possibility of endoscopic QPI is another interesting avenue
[36]. While most existing QPI systems measure samples in vitro or ex vivo, the
study of in vivo or deep tissue imaging with QPI is rarely explored [143, 144].
Thirdly, label-free super-resolution imaging remains a significant technical goal.
The existing super-resolution methods are typically involving fluorophores. Finally,
we anticipate that combining QPI and artificial intelligence approaches will lead to
advances in objective diagnosis. In the years to come, we anticipate that QPI will
continue gaining popularity as it transfers from engineering to biomedical laboratories [1].