© 2022 All Rights Reserved



1Key Laboratory of Biomedical Engineering, School of Biomedical Engineering, Hainan University, Hainan, China
2Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan, China


Over the past decade, modern advanced optical imaging technologies change people’s ways of understanding the world in many aspects, including clinical application, industrial measurement, and fundamental research [1-3]. For all-optical imaging methods, there always exists a trade-off between high resolution and deep penetration depth [4]. Among them, a group of imaging methods, which collect ballistic or single scattering photons [5,6]. They commonly provide high-resolution imaging power but the imaging depth limit to a superficial layer. For those methods that are deep imaging availably, they generally rely on diffuse photons [7]. But the imaging resolution is degraded a lot. The fundamental limitation of this phenomenon is the multiple scattering of light. The primary cause is the anisotropy of biological tissues and the impact of photons on the micro-particles. To solve this problem, pioneers’ have done significant work in improving society. Reflection matrix measurement and decomposition of the time-reversal method have shown their abilities to tell the single scattering photons from the multiple scattering [8,9]. By shaping the wavefront of light, scattering events become controllable so that light is focused even transmits through a highly scattering medium [10].
Here, we present our work for the following two purposes--- deep imaging and focusing. In the aspect of deep imaging purpose, we combine optical heterodyne detection and lock-in amplitude to detect those weak signals drowned in the background noise [11]. After reconstructing the reflection matrix that describes the light propagation process in the sample, and applying a singular value decomposition of the matrix. We succeed in imaging through a sample with 15.2 times scattering means free path (SMFP) without compromising to resolution decline. It is a significant improvement in imaging depth when compared to confocal microscopy ~ 2-3 SMFP or optical coherence tomography ~ 6-7 SMFP.
In the aspect of focusing light deep within media purposes, we have done two types of work. The first work is a high-speed wavefront determination for millisecond beam focusing through a scattering medium. For a very long time, one of the major challenges to apply wavefront shaping methods in practical applications is the required long optimization time. In this work, we propose an in-&-out light field analysis method to calculator the matched wavefront. The advantage of this technology is that the speed of the whole beam focusing process is ~113ms [12]. When compared to the conventional iterative feedback wavefront of transmission matrix measurement methods, it has shortened the time consumption for 2~3 orders. We believe this work shows the potential to enable the translation of the wavefront shaping method to increase image depth in a highly scattering medium such as biological tissues. The second work is to focus the beam inside the scattering medium without any guide-star assistant. Firstly, we applied time-reversal operation to the reflection matrix to filter out those multiple scattering photons that arrived at the target position. Then, by shaping the incident light according to the optimal wavefronts, the above-mentioned multiple scattering photons would travel in the “open channels” this time. Light travel in the open channel will experience zero-energy loss and the shortest flight of time towards the target. After optimal, the intensity of light distribution, at the imaging depth of 9.6 times scattering mean free path, becomes focusing again. At the same time, the energy has also been improved by one order.
In conclusion, we have combined decomposition of time-reversal operation, reflection matrix measurement, and wavefront shaping to imaging and focusing ultra-deep inside the complex medium. Based on the above research, we plan to apply this technology to a broader range of fields in the coming work. Such as brain science research, the skull layer is considered a typical highly scattering layer. We want to provide a new in-vivo, noninvasive, and high-resolution optical imaging method to imaging the nerve and blood vessels beneath mice skulls. In the field of fundamental research, we show the potential of this method in studying the light propagation process, separating single from multiple scattering photons, and the ability to control photons' travel path within the scattering medium.

File with abstract


Jing Cao
Hainan University


Ask question