Open Access
Add to BibSonomy
Abstract
Lissajous micro scanners are very attractive in compact laser scanning applications for biomedical endoscopic imaging, such as confocal microscopy, endomicroscopy or optical coherence tomography. The scanning frequencies have a very important effect on the quality of the resulting Lissajous scanning imaging. In this paper, we propose a frequency selection rule for high definition and high frame-rate (HDHF) Lissajous scanning imaging, by deriving the relationship among the scanning field of view (FOV), actuation frequencies and pixel size based on the characteristics of the scanning trajectory. The minimum sampling rate based on the proposed frequency selection rule is further discussed. We report a lead zirconate titanate piezoelectric ceramic (PZT) based Lissajous fiber scanner to achieve HDHF Lissajous scanning imaging. Based on the frequency selection rule, different frequency combinations are calculated, under which the Lissajous fiber scanner can work at the frame rate (FR) of 10 Hz, 20 Hz, 40 Hz and 52 Hz. The trajectory evolution of the Lissajous scanning at the frame rate of 10 Hz has been obtained to verify the applicability of the proposed rule. The measured resolution of the scanner is 50.8 lp/mm at the unit optical magnification, and the measured FOV at the FR of 10 Hz and 40 Hz are 1.620 mm ×1.095 mm and 0.405 mm ×0.27 mm, respectively. HDHF Lissajous scanning images of the customized spatial varying binary pattern are obtained and reconstructed at the FR of 10 Hz and 40 Hz, demonstrating the practicability of the frequency selection rule.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Micro-scanner is playing an important role in many modern technologies such as the medical imaging [1,2] and the laser projector [3–5]. Typically, the micro-scanners [6] have been used in the medical imaging techniques such as confocal microscopy [7,8], two-photon microscopy [9,10], and optical coherence tomography (OCT) [11,12] to examine the early lesions and tumors of organs.
The actuation mechanism is important for the micro-scanner. A main micro-scanner is based on the microelectromechanical systems (MEMS) [13,14], which has been intensively investigated due to the advantages of the MEMS such as low power consumption, light weight and high performance. However, the MEMS-based micro-scanners still have the limitation of the relatively large dimension, complicated circuit and difficulty in the fabrication. Compared with the MEMS based micro-scanner, the lead zirconate titanate piezoelectric ceramics (PZT) based micro-scanner [9,15,16] has the advantages of the simplicity, compactness, convenience in the fabrication, low power consumption and large deformation [17]. The PZT is becoming the popular material to actuate the micro-scanner. In general, the PZT based micro-scanners drive the optical fiber fixing on it to form a 2D scanning trajectory.
There are three commonly applied scanning patterns in the PZT based micro-scanners, i.e., the raster scanning, spiral scanning and Lissajous scanning. The raster scanning [18] has a fast axis motion and a slow axis motion. The frame rate (FR) of raster scanning equals to the frequency of the slow axis motion. The spiral scanning [19] can be formed by driving the PZT tube with two driving signals with the same resonant frequency and a phase-shift of 90°. However, the strong illumination density of the center region of the spiral scanning restricts its further application [20]. The Lissajous scan [21] has been proposed to form the scanning pattern with the uniform illumination and high frame rate, simultaneously. Thus, Lissajous scanning is the preferred candidate for the micro-scanners. Tekpınar et al. present a mode-optimized two-dimensional piezoelectric fiber actuator and a robust actuation scheme to establish three different scan patterns with the same device [22]. They offered an actuation scheme to achieve different scan patterns, and successfully generated raster, spiral and Lissajous patterns at ≥ 20 frames per second (FPS).
The quality evaluation criterions of the Lissajous scanning imaging include the definition, resolution, and field of view (FOV). All the evaluation criterions are closely related to the scanning pattern of the Lissajous scanning, which is determined by the parameters of the two orthogonal drive signals, such as the amplitude, initial phase, and frequency. To form a high-definition high frame rate (HDHF) Lissajous scanning pattern all the parameters must be determined appropriately. Some research groups proposed some strategy to realize the HDHF Lissajous scanning imaging. Zhang et al. provided a general calculation method for the fill factor (FF) values of the Lissajous scanning, and further redefine the center calculation (Δah and Δav) based on the vertical network pixels [23]. Hwang et al. reported a selection rule for HDHF Lissajous scanning by choosing the maximum greatest common divisor (GCD) among various sets of the scanning frequencies satisfying the total lobe number for a target FF [24]. Unfortunately, they did not establish the mathematical relationship among the parameters. Also, the effect of the limited sampling rate to realize the HDHF Lissajous scanning imaging was not considered. Brunner et al. proposed a phase modulation method for Lissajous scanning systems, which provided adaptive scan pattern design without changing the FR or the FOV [25]. They derived phase modulation constrains and a method for pixel calculation to achieve adaptive scan. Tanguy et al. reported a MEMS based micro-scanner used for the medical diagnosis [26]. They proposed a strategy for the frequency selection to realize the HDHF Lissajous scanning imaging. However, their MEMS based micro-scanner has an upper limit of the resonant frequency, which will limit the FR as well as the FF. He et al. presents the most simplified fiber scanner to date [27]. The scanner they proposed contains only a single piezoelectric bimorph and a single non-symmetrical fiber with a 1D actuator for 2D scanning. The scanner work at 5 FPS with a scanning range of >300 µm.
In this paper, we present a quantitative frequency selection rule considering the parameters including the pixel size, driving frequencies and scanning field of view based on the characteristics of the Lissajous trajectory. We utilize a PZT-based fiber scanner with an asymmetric fiber cantilever, whose driving frequencies can reach 1720Hz and 2550 Hz. For the spacing between the driving frequencies is large, the fiber scanner can realize an ideal mechanical coupling-free Lissajous scanning imaging. Additionally, the high driving frequencies enable the Lissajous scanning imaging with higher FR, high definition and wide FOV. In the following section we mathematically describe Lissajous trajectory and specifications and matching principle of pixel size and Lissajous trajectory, after that we describe the design and fabrication of the micro-scanner and the optical setup. Finally, we report the experimental results and discussion.
2. Methods
2.1 Frequency selection rule for HDHF Lissajous scanning
The imaging data corresponding to the Lissajous trajectory projected on the target area is obtained for the Lissajous scanning imaging. The Lissajous trajectory formed by scanning the laser spot can be mathematically expressed as
To achieve the HDHF Lissajous scanning imaging, the projected trajectory should cover the target area as densely and uniformly as possible. The frequencies of the sinusoidal movements in the orthogonal directions affect the density and uniformity, as well as the FR of the Lissajous trajectory. Hence, the frequencies should be selected appropriately. To characterize the coverage density and uniformity of the Lissajous scanning trajectory we introduce the term of the trajectory fill factor (TFF), i.e., the ratio of the pixel number covered by the trajectory NP to the total pixel number NT, which can express as
For HDHF Lissajous scanning imaging three requirements must be met. Firstly, the TFF must be as high as possible. Secondly, the total pixel number must be as much as possible. Thirdly, the FR must be as high as possible.
In order to make the TFF as high as possible, the pixel size should match the density and uniformity of the trajectory. As shown in the Fig. 1, in the case of the green pixel the TFF can achieve 100%, but the green pixel size will decrease the total pixel number and the image definition. In the case of the black pixel the total pixel number and the image definition will be improved, but the TFF of the Lissajous trajectory will be reduced. In the case of the red pixel the total pixel number and the image definition will be improved while ensuring the 100% TFF.
Fig. 1. Lissajous scanning trajectory and pixel matching relationship. The red pixel size is set to be psx, the green pixel size is set to be 2psx, and the black pixel size is set to be 0.5psx.
To make the TFF as high as possible, it is necessary to ensure that there is only one scanning trajectory passing through the pixels in the center FOV, and the side of the diamond in the center FOV is used as the diagonal of the pixel, as indicated by the red pixel in Fig. 1. To quantitatively determine the optimal pixel size, take half of the value of the first intersection point between the trajectory and the positive half axis of the x-axis and y-axis as the length and width of the pixel, respectively. According to the Eq. (1), the pixel size can be express as
As shown in the Eq. (5), the definition of the Lissajous scanning is related to the two driving frequencies fx, fy, and the FR f0. Using the Eq. (5) to determine the pixel size and driving frequencies, we can achieve the Lissajous trajectory with the 100% TFF and the least redundant scanning at the edge of the FOV.
In order to maximize the total pixel number and the FR, the relationship between the pixel number and the FR is derived from Eq. (5), usually, the FR is much lower than the driving frequencies, so this relationship can be expressed as
The optical resolution of the pixel is also the limit factor for the HDHF Lissajous scanning imaging. The pixel resolution should be matched with the spatial resolution of the optical system. If the pixel resolution is lower than the spatial resolution, the image resolution will be degraded compared with the spatial resolution due to the insufficient pixel number. If the pixel resolution is higher than the spatial resolution, there will be many pixels not sampled by the Lissajous trajectory, and the TFF will fall down. This also induces the resolution degradation. In a word, the pixel resolution should be matched with the spatial resolution. The proposed frequency selection rule ensures the matching between the pixel resolution and spatial resolution.
2.2 Influences of the sampling rate on the HDHF Lissajous scanning imaging
Since the image data is acquired by the data acquisition (DAQ) card, the imaging quality of the Lissajous scanning imaging is affected by the sampling rate of the DAQ card. We define the parameter of pixel fill factor (PFF) to quantitatively evaluate the influence of the sampling rate. The PFF is defined as the ratio of the sampled pixel number NS to the total pixel number NT in the reconstructed Lissajous scanning image, which can be expressed as
In order to demonstrate the effect of the sampling rate on the Lissajous scanning imaging we simulate the Lissajous scan trajectory under the same driving frequencies fx, fy, but with different sampling rates. The two orthogonal driving frequencies are set to 9 Hz and 7 Hz, respectively. According to Eq. (6) Nx and Ny are set to 12 px and 9 px, respectively for achieving the 100% TFF. The Lissajous scan trajectory are sampled at the sampling rates of 300 Samples/s, 1300 Samples/s and 2300 Samples/s.
The reconstructed results are shown in the Fig. 2. The red dots on the trajectory represent the sampled data points, the white pixels represent the sampled pixels, and the gray pixels represent the pixels haven’t been sampled. In Fig. 2(a) some pixels are not sampled due to the low sampling rate of 300 Samples/s. In Fig. 2(b) as the sampling rate increasing to 1300 Samples/s all the pixels are sampled. In Fig. 2(c) the sampling rate is further increased to 2300 Samples/s. Each pixel is sampled, but the pixels at the edge region is sampled multiple times. The PFFs for Fig. 2(a), (b) and (c) are 72.2%, 100% and 100%, respectively. Therefore, even if the TFF is 100%, the imaging data corresponding to some pixels still may not be sampled due to the low sampling rate. On the other hand, if the sampling rate is too high, the pixel information of the edge region will be sampled multiple times, increasing the data redundancy and the burden of the DAQ card.
Fig. 2. A brief schematic diagram of the matching relationship between the pixels and the Lissajous trajectory. In all the subgraphs, the ratio of the pixels along horizontal and vertical directions is 4:3. The PFF is related to the sampling rate, which can be calculated by the ratio of effective pixels (white grid with at least one red dot) and the total pixels. The sampling rate of the subgraphs from (a) to (c) is set to be 300 Samples/s,1300 Samples/s and 2300 Samples/s, respectively. The subgraphs have a FOV of 12 px × 9 px, the corresponding scanning frequencies are fx = 7 Hz, fy = 9 Hz, the calculated PFF is 72.2%, 100% and 100%, respectively.
As shown in Fig. 2, the decrease of PFF is mainly occurred in the central area of the scanning FOV. In the central area of the scanning FOV, the scanning trajectory is distributed along the diagonal of the square pixels. Thus, it is needed that the diagonal of each square in the central scanning FOV have at least one sampling point. For making the most use of the sampling rate, there should be one and only one sample point on the pixel diagonal. To meet the condition, the scanning time when the scanning trajectory passes through the diagonal of the pixel grid in the central FOV must be calculated. We derive the time of the scanning trajectory through the diagonal of the pixel grid in the center area as
Considering the scanning trajectory first passes through the nearest central origin of the scanning FOV, the t takes the minimum positive value which corresponds to the minimum sampling rate. Thus, the threshold sampling rate for a HDHF Lissajous scanning imaging can be expressed as
The Eq. (9) combines factors that affect PFF to provide a sampling rate selection basis for HDHF Lissajous scanning imaging.
To sum up, regardless of the considered FOV, the PFF of 100% will be achieved by the help of Eq. (9). Combing the proposed frequency selection rule and the threshold sampling rate the HDHF Lissajous scan can be realized.
3. Experiment
To verify the proposed frequency selection rule a PZT based Lissajous fiber scanner is fabricated and tested. The scanner utilizes the asymmetric fiber cantilever [27] to generate the orthogonal resonant frequencies. The basic structure of the asymmetric fiber cantilever is shown in Fig. 3(a). A light transmitting fiber L1 is glued to the upper surface of the PZT, while a protruding fiber L2 is glued to the lower surface of the PZT. Then a connecting fiber is glued to the light transmitting fiber and protruding fiber to form a rigid plane. The optical magnifier is used during the fabrication to control the lengths of L1 and L2 accurately. For the purpose of endoscopic imaging and a higher resonant frequency the rigid length of the asymmetric fiber cantilever should not be longer than 10 mm. Thus, the fabricated asymmetric fiber cantilever has the L1 length of 8.327 mm and L2 length of 2.242 mm. As shown in Fig. 3(b) the experimentally measured orthogonal resonant frequencies at the maximum amplitudes are 1720Hz and 2550 Hz, and the corresponding bandwidths are 36 Hz and 22 Hz, respectively. The difference between the resonant frequencies is greater than 800 Hz, predicting the ideal mechanical coupling performance of the asymmetric fiber cantilever-based scanner.
Fig. 3. (a) The structure of the asymmetric fiber cantilever. (b) The experimental measured scanning amplitude under different driving frequencies. (c) The calculated FR under different driving frequency combinations. (d) The total pixel numbers under different driving frequency combinations.
According to the proposed frequency selection rule the appropriate frequency combinations should be selected from the two bandwidths to achieve HDHF Lissajous scanning. By using the Eq. (5) and Eq. (6) to determine the pixel numbers, the TFF can achieve 100%. The FR and total pixel numbers under different frequency combinations are shown in Fig. 3(c) and (d). As shown in Fig. 3(c) and (d), we can see the trade-off relationship between FR and total pixel number. The higher the FR, the lower the total pixel number, vice versa. Therefore, we selected 4 frequency combinations with FR ranges from 10 Hz to 52 Hz. The corresponding frequency combinations are (1720Hz, 2550 Hz), (1720Hz, 2540 Hz), (1720Hz, 2560 Hz) and (1716Hz, 2548 Hz). Providing the pixel shape is square, the pixel size psx × psy is set to be 5 µm × 5 µm, thus, the corresponding total pixel numbers are 324 × 219, 161 × 109, 81 × 54 and 62 × 42. The linear FOV can reach 1.620 × 1.095 mm2, 0.805 × 0.545 mm2, 0.405 × 0.27 mm2 and 0.31 × 0.21 mm2. The above discussed parameters are listed in Table 1.
Table 1. Parameters of the representative frequency combinations for the higher FR and larger FOV imaging
Figure 4 shows the schematic of the optical setup. Using a near-infrared laser (S1FC1310, Thorlabs Inc.) with the operating wavelength of 1310 nm, the output light is coupled into the scanner. The light deflects by the scanner is focused by a single lens (Lens1) with the focal length of 35 mm on the CCD camera (acA2000-340kmNIR, BASLER Inc.) or the tested sample. The distance between the scanner and Lens1 is twice the focal length, i.e., the optical magnification of the lens system is 1x. The ideal imaging spot size is 10 µm according to the fiber core diameter of 10 µm. A waveform generation card (PCIe-6738, NI Inc.) is used for generating the drive signals to the fiber scanner, as well as the synchronous signal for the data acquisition. The detected electrical signal is digitized by a DAQ card (PCIe-9350, Alazar Inc.). A lookup table of the position coordinate corresponding to each sampling point is firstly generated based on the theoretical Lissajous scanning trajectory. Then the Lissajous scanning image can be reconstructed by the pre-calibration-based method using the digitized electrical signal [28].
Fig. 4. The schematic of the fiber-based microscope system for the real-time imaging, the blue lines represent the optical fiber, red solid and dotted lines represent the light paths, the black lines represent the electrical signals, respectively. The part in the dashed box is interchangeable with the CCD camera for different experiments. SYNC represents synchronizing trigger signal. DAQ represents data acquisition card.
In order to verify that the developed Lissajous fiber scanner can realize the HDHF Lissajous scanning imaging, the CCD camera is used to record the Lissajous scanning trajectory of the fiber end. The Lissajous fiber scanner operates at the FR of 10 Hz with the driving frequencies of 2550 Hz and 1720Hz, i.e., the scanning period is 100 ms. The actual scanning range of the vibrating fiber end can be recorded by the CCD camera due to the unit optical magnification. The image detected by the CCD camera is obtained by an image acquisition (IMAQ) card. To observe the evolution of the scanning trajectory the exposure time of the camera is set to be 1 ms, 10 ms, 20 ms, 50 ms, 75 ms and 100 ms, i.e. the 1%, 10%, 20%, 50%, 75% and 100% of the scanning period.
To investigate the relationship between the fill factor and the sampling rate a negative USAF-1951 USAF resolution target is placed at the focal plane of the Lens1, i.e., the CCD camera in Fig. 4 is replaced by the optical path in the dashed box. The Lissajous fiber scanner operates at the FR of 10 Hz with the driving frequencies of 2550 Hz and 1720Hz. The light transmitted through the resolution target is collected by the Lens2 (f = 50 mm), and focused on a biased detector (DET10N2, Thorlabs Inc.). The sampling rate of the DAQ card is set to 200 KSamples/s, 400 KSamples/s, 1 MSamples/s and 2 MSamples/s to obtain the image in the experiment. The PFFs of the obtained images are calculated according to the Eq. (7). To measure the practical resolution of the developed fiber scanner the smallest discernable line pairs on the resolution target is experimentally determined and analyzed using the same optical path in Fig. 4.
A spatially varying binary pattern is imaged to verify the real-time imaging performance of the HDHF Lissajous scanner. The binary pattern is formed on a Cr-coated glass substrate. The pattern is imaged at the FR of 10 Hz and 40 Hz, the corresponding FOV is 1.620 × 1.095 mm2 and 0.405 × 0.27 mm2. The sampling rate is set to be 2 MSamples/s for achieving the 100% PFF. The image of the resolution target and the binary pattern are reconstructed by the pre-calibration-based method.
4. Results
The scanning trajectories with the FR of 10 Hz obtained at the different exposure time are shown in Fig. 5. As we can see, with the increase of exposure time the coverage of the trajectory on the target gradually increases. When the exposure time reaches 100 ms, that is, one scanning period, the entire target FOV can be completely covered by the trajectory. It verifies that the proposed frequency selection rule can be used to realize 100% TFF. Observing the evolution of the rectangular trajectory, the scanner is not influenced by the mechanical coupling, which agrees well with the measured orthogonal resonant frequencies.
Fig. 5. Lissajous trajectory captured at 1 ms, 10 ms, 20 ms, 50 ms, 75 ms and 100 ms over a full scan period.
Figure 6 is the reconstructed images of the resolution target tested by the fiber scanner at the different sampling rates varied from 200 KSamples/s to 2 MSamples/s. The PFF of the Lissajous scanning images shown in Fig. 6(a) to (d) is 22.2%, 45.7%, 71.3% and 100%, respectively. According to the Eq. (9) the threshold sampling rate is 1.75 MSamples/s for the experimental setup. As we can see from Fig. 6(a) to (c) the PPF cannot reach 100% due to the lower sampling rate relative to the threshold, and the PFF increases with the increasing of the sampling rate. As shown in Fig. 6(d) the PFF reaches 100%, for the sampling rate is higher than the threshold. From the figures it demonstrates that the Eq. (9) can be used to determine the threshold sampling rate that is necessary to realize the HDHF Lissajous scanning imaging.
Fig. 6. The reconstructed images of the resolution target tested by the fiber scanner at the different sampling rates of (a) 200 KSamples/s, (b) 400 KSamples/s, (c) 1 MSamples/s and (d) 2 MSamples/s.
Figure 7(a) is a detailed photograph of the negative 1951 USAF resolution target downloaded from the internet. It shows the enlarged drawings of the fourth, fifth, sixth and seventh groups of elements. We imaged the interested area marked with the red box with the developed fiber scanner. The reconstructed image is shown in Fig. 7(b), the line pairs are the 4th∼6th elements of the group 5. Both the vertical and horizontal lines of the 5th element of the group 5 are the smallest discernable line pairs. The number of line pairs per millimeter of that element is 50.8 lp/mm, i.e., the resolution is higher than 50.8 lp/mm. This means that the experimentally measured spot size is smaller than 19.69 µm, which is the spacing between the two adjacent lines of that element. The discrepancy between the experimentally measured spot size and the ideal imaging spot size is mainly due to diffraction effect, aberration effect and the misalignment of the line path.
Fig. 7. (a) The photograph of negative 1951 USAF resolution target with the interest area marked with a red box. (b) The reconstructed image of the 4th∼6th elements of group 5. The scale bar is 50µm.
The photograph of the spatially varying binary pattern is shown in the Fig. 8(a) and (b), which is obtained by an optical magnifier. The binary pattern is the abbreviated name ‘NUAA’ and the logo of the Nanjing University of Aeronautics and Astronautics. The dark part of the pattern is the region coated with the metal Cr for blocking light from passing through, while the white part is Cr-free region. Figure 8(c) shows the reconstructed Lissajous scanning image of the ‘NUAA’ at the FR of 10 Hz. The physical FOV is 1.620 mm × 1.095 mm. Under the fixed resolution, to realize large field of view the FR is needed to be lowered down. In order to perform high FR scanning imaging, the FOV can be downsized to 0.405 mm × 0.27 mm. Thus, the FR can be set to 40 Hz. The reconstructed image of the logo pattern is shown in Fig. 8(d). The short lines at the bottom left of the image in Fig. 8(d) have the linewidth of 15 µm. The ‘1952’ in the logo can be differentiated clearly, as well as the triangle-like pattern. It suggests that the scanner can realize the HDHF Lissajous scanning imaging with the FR of 40 Hz and resolution better than 15 µm. Based on the results shown in Fig. 8(c) and (d), there is no missing pixel information in the two images, which proves the validity of the proposed frequency selection rule.
Fig. 8. (a) The magnified photograph of the pattern ‘NUAA’ with the linewidth of 50 µm. (b) The magnified logo pattern with the size of 1 mm × 1 mm. (c) Reconstructed scanning image of ‘NUAA’ obtained at the FR of 10 Hz and the FOV of 1.620 mm × 1.095 mm. Scale bar = 200 µm. (d) Reconstructed image of a part of the logo pattern at the FR of 40 Hz and the FOV of 0.405 mm × 0.27 mm. Scale bar = 50 µm.
5. Conclusion and discussion
In summary, we present a frequency selection rule for HDHF Lissajous scanning imaging method. A quantitative frequency selection rule considering the parameters including the pixel size, driving frequencies and scanning field of view based on the characteristics of the Lissajous trajectory has been proposed. Based on the rule a PZT-based fiber scanner with an asymmetric fiber cantilever was designed, fabricated and tested. The driving frequencies of the fiber scanner reached 1720Hz and 2550 Hz. Typically, the developed fiber scanner can operate with the FR of 40 Hz, the FOV of 0.405 mm ×0.27 mm and the resolution of 50.8 lp/mm. The experimental results demonstrate the validity of the proposed frequency selection rule to realize the HDHF Lissajous scanning imaging.
Compared with the HDHF frequency selection rule proposed in the prior work [24,26], we proposed a quantitative frequency selection rule considering the parameters of the Lissajous scanning imaging. We utilized a PZT-based fiber scanner with an asymmetric fiber cantilever to verify the proposed rule. The driving frequencies of the developed fiber scanner can reach 1.7 kHz and 2.5 kHz, higher than that of any previously reported fiber scanner including the MEMS based scanner. The higher resonant frequency of the PZT-based fiber scanner enables the Lissajous scanning imaging with the higher frame rate, higher definition or wider FOV. Meanwhile, the presented asymmetric fiber cantilever based scanner has much larger spacing between the driving frequencies, realizing an ideal mechanical coupling-free Lissajous scanning imaging. Additionally, the presented PZT-based fiber scanner can switch between two different imaging modes according to the clinical need by varying the drive frequency conveniently. For the wide range screening of a certain tissue, the scanner can operate in its large FOV imaging mode by lower the FR. On the other hand, for the fast screening of some dynamic process in the tissue, the scanner can operate in its high FR imaging mode by downsizing the FOV.
However, due to the limitation of the material properties the resonant frequency of the fiber scanner has an upper limit, which cannot be increased infinitely. The mechanical structure of the currently developed coating-stripped fiber scanner is fragile, and this limits the maximal scanning FOV it can achieve. Additionally, the aberration of the imaging optics limits its practical resolution. In future a HDHF Lissajous fiber scanner designed with the proposed frequency selection rule with the higher structural strength and better optical resolution could be further developed and applied in the practical clinical scenarios.
Funding
National Natural Science Foundation of China (62105146, 62175106).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. G. J. Tearney, S. A. Boppart, B. E. Bouma, et al., “Scanning single-mode fiber optic catheter–endoscope for optical coherence tomography,” Opt. Lett. 21(7), 543–545 (1996). [CrossRef]
2. S. A. Boppart, B. E. Bouma, C. Pitris, et al., “Forward-imaging instruments for optical coherence tomography,” Opt. Lett. 22(21), 1618–1620 (1997). [CrossRef]
3. Y. H. Seo, K. Hwang, H. Kim, et al., “Scanning MEMS Mirror for High Definition and High Frame Rate Lissajous Patterns,” Micromachines 10(1), 67 (2019). [CrossRef]
4. U. Hofmann, F. Senger, J. Janes, et al., Wafer-level vacuum-packaged two-axis MEMS scanning mirror for pico-projector application, SPIE2014.
5. C. D. Chen, Y. J. Wang, and P. Chang, “A novel two-axis MEMS scanning mirror with a PZT actuator for laser scanning projection,” Opt. Express 20(24), 27003–27017 (2012). [CrossRef]
6. Mandeep Kaur, Pierre M Lane, and Carlo Menon, “Scanning and Actuation Techniques for Cantilever-Based Fiber Optic Endoscopic Scanners—A Review,” Sensors 21(1), 251 (2021). [CrossRef]
7. H. J. Shin, M. C. Pierce, D. Lee, et al., “Fiber-optic confocal microscope using a MEMS scanner and miniature objective lens,” Opt. Express 15(15), 9113–9122 (2007). [CrossRef]
8. L. Liu, L. Wu, P. Zory, et al., “Fiber-Optic Confocal Microscope with an Electrothermally-Actuated, Large-Tunable-Range Microlens Scanner for Depth Scanning,” Proc Ieee Micr Elect 827–830 (2010).
9. R. Sharma, L. Yin, Y. Geng, et al., “In vivo two-photon imaging of the mouse retina,” Biomed. Opt. Express 4(8), 1285–1293 (2013). [CrossRef]
10. R. M. Boutilier, J. S. Park, and H. Lee, “High-speed Two-photon Laser Scanning Microscopy Imaging of in vivo Blood Cells in Rapid Circulation at Velocities of Up to 1.2 Millimeters per Second,” Curr Opt Photonics 2, 595–605 (2018). [CrossRef]
11. D. Huang, E. A. Swanson, C. P. Lin, et al., “Optical coherence tomography,” Science 254(5035), 1178–1181 (1991). [CrossRef]
12. H.-C. Park, Y.-H. Seo, and K.-H. Jeong, “Lissajous fiber scanning for forward viewing optical endomicroscopy using asymmetric stiffness modulation,” Opt. Express 22(5), 5818–5825 (2014). [CrossRef]
13. H. C. Park, C. Song, M. Kang, et al., “Forward imaging OCT endoscopic catheter based on MEMS lens scanning,” Opt. Lett. 37(13), 2673–2675 (2012). [CrossRef]
14. G. Zhou, Z. H. Lim, Y. Qi, et al., “Single-Pixel MEMS Imaging Systems,” Micromachines 11(2), 219 (2020). [CrossRef]
15. T. Wu, Z. H. Ding, K. Wang, et al., “Two-dimensional scanning realized by an asymmetry fiber cantilever driven by single piezo bender actuator for optical coherence tomography,” Opt. Express 17(16), 13819–13829 (2009). [CrossRef]
16. N. Zhang, T. H. Tsai, O. O. Ahsen, et al., “Compact piezoelectric transducer fiber scanning probe for optical coherence tomography,” Opt. Lett. 39(2), 186–188 (2014). [CrossRef]
17. T. Naono, T. Fujii, M. Esashi, et al., “Non-resonant 2-D piezoelectric MEMS optical scanner actuated by Nb doped PZT thin film,” Sens. Actuators, A 233, 147–157 (2015). [CrossRef]
18. Y. R. Teo, Y. Yong, and A. J. Fleming, “A comparison of scanning methods and the vertical control implications for scanning probe microscopy,” Asian Journal of Control 20(4), 1352–1366 (2018). [CrossRef]
19. H. Schulz-Hildebrandt, T. Pfeiffer, T. Eixmann, et al., “High-speed fiber scanning endoscope for volumetric multi-megahertz optical coherence tomography,” Opt. Lett. 43(18), 4386–4389 (2018). [CrossRef]
20. W. Liang, K. Murari, Y. Zhang, et al., “Increased illumination uniformity and reduced photodamage offered by the Lissajous scanning in fiber-optic two-photon endomicroscopy,” J. Biomed. Opt. 17(2), 021108 (2012). [CrossRef]
21. J. Wang, G. Zhang, and Z. You, “Design rules for dense and rapid Lissajous scanning,” Microsyst. Nanoeng. 6(1), 101 (2020). [CrossRef]
22. M. Tekpınar, R. Khayatzadeh, and O. Ferhanoğlu, “Multiple-pattern generating piezoelectric fiber scanner toward endoscopic applications,” Opt. Eng. 58(02), 1 (2019). [CrossRef]
23. Y. Zhang, Y. Liu, L. Wang, et al., “Resolution adjustable Lissajous scanning with piezoelectric MEMS mirrors,” Opt. Express 31(2), 2846–2859 (2023). [CrossRef]
24. K. Hwang, Y. H. Seo, J. Ahn, et al., “Frequency selection rule for high definition and high frame rate Lissajous scanning,” Sci. Rep. 7(1), 14075 (2017). [CrossRef]
25. D. Brunner, H. W. Yoo, R. Schroedter, et al., “Adaptive Lissajous scanning pattern design by phase modulation,” Opt. Express 29(18), 27989–28004 (2021). [CrossRef]
26. Q. A. A. Tanguy, O. Gaiffe, N. Passilly, et al., “Real-time Lissajous imaging with a low-voltage 2-axis MEMS scanner based on electrothermal actuation,” Opt. Express 28(6), 8512–8527 (2020). [CrossRef]
27. L. He, X. Li, J. Yang, et al., “Super-simplified fiber scanner for cellularresolution endoscopic imaging,” Photon. Res. 11(12), 2020–2032 (2023). [CrossRef]
28. T. Wu, L. Zhang, W. Q. Huo, et al., “Miniaturized precalibration-based Lissajous scanning fiber probe for high speed endoscopic optical coherence tomography [J],” Opt. Lett. 45(8), 2470–2473 (2020). [CrossRef]
