Multimodal Biomedical Imaging Systems - Biomedical Optical Imaging Technologies: Design and Applications - page 312

or sequentially, while each modality works independently. In spectral integration,
different imaging modalities are integrated through the dichroic combiner. The
optical axes may or may not overlap, and different modalities may share some
components, such as the objective lens and scanner. One requirement for this
configuration is that the each modality should have a distinct working spectrum
with minimal overlap. For example, fluorescence imaging and optical coherence
tomography can be spectrally incorporated because most endogenous components
do not absorb or emit within the OCT spectral range. In temporal integration, each
modality is basically independent without common component. Multimodal images
are captured by moving the sample to the object plane of each modality or moving
each modality to the sample sequentially. This configuration is time-consuming
and the sample or the system should be stable enough so that multimodal images
can be correlated accurately. Therefore, temporal integration is typically used to
demonstrate the concept and is not commonly used in clinical applications.
Multimodal images can be acquired either sequentially or simultaneously. If
the spectra of different modalities overlap, the system should be time multiplexed,
which means that images are captured sequentially. When the spectra do not overlap,
it is preferable to capture images simultaneously for better registration and faster
image capture. Sometimes, it is difficult to image biological events that occur over
a short interval by using a multimodal imaging system which requires switching
between light sources and associated optical paths. For such applications, it is
necessary to image all the details contained in the specimen at the same time.
In certain situations, spectral overlap is advantageous because more than one
image can be captured with only one illumination beam. One example is dual-
channel optical coherence tomography and indocyanine green dye (OCT-ICG)
fluorescence system for investigation of the eye fundus [ 14 ]. The same superlu-
minescent diode is used to simultaneously generate the OCT image and excite the
ICG fluorescence. Pixel-to-pixel correspondence between the OCT and fluorescence
images allows the user to obtain OCT B-scans precisely at selected points on the
ICG images.
The effectiveness of a multimodal system relies on the combination of the various
imaging techniques but not on the extreme diagnostic capabilities of each technique.
It is always challenging to integrate different imaging techniques into one system
because specific hardware is typically required for various imaging modalities.
The clinical environment also does not allow for very complex instrumentation.
Therefore, short-scale, simple, and reliable configurations must be considered for a
multimodal system.
To develop an effective multimodal imaging system, one needs to understand the
different types of information needed to obtain, principles and design considerations
for each imaging modality, photon paths in the tissue, depth-resolved requirements,
and signal separation. The design also needs to consider the resolution, material
constraints, FOV, light source, illumination system design, and imaging system
design.