Quantitative evaluation of the supercontinuum laser eye dazzling effect: in vivo experimental research

To quantitatively evaluate the dazzling effect of each spectrum band of the supercontinuum laser, we conducted experimental research to explore the safety and dazzling of animal eyes. Under the condition of dark adaptation, the rabbit eyes were irradiated with different power densities and spectral bands by frontal incident mode for 0.25 s, which was repeated ten times. The fundus of the rabbit eyes was examined using an ophthalmoscope, and the upper limit of safe power density was explored. Rabbit eyes were irradiated with different doses of dazzling light for 0.1 s. Visual electrophysiological signals were collected dynamically, and the recovery time of the electroretinogram (ERG)-b wave amplitude of the rabbit eyes was recorded and analyzed after laser irradiation. When the power density was 8.0 mW cm−2 in visible spectrum (vs.), the recovery time of the ERG-b wave in the rabbit eye was 4.11 ± 0.67 s. When the power density was 12.0 mW cm−2 in the full spectrum (FS), the recovery time of the ERG-b wave in the rabbit eye was 4.16 ± 0.55 s. The recovery time of the ERG-b wave was 4.50 ± 0.94 s at a power density of 4.6 mW cm−2 in FS-1 and 3.81 ± 0.11 s at a power density of 5.0 mW cm−2 in the FS-2. When the power density was 628.00 mW cm−2 in infrared spectrum (IS), the recovery time of the ERG-b wave was only 0.84 ± 0.09 s. The reference values for the upper limit of the safe irradiation power density of the supercontinuum laser are set as follows: 25.2 mW cm−2 in vs., 118.4 mW cm−2 in IS and 105.0 mW cm−2 in FS. The vs., FS, FS-1 and FS-2 of the supercontinuum laser had a good dazzling effect on rabbit eyes, and the dazzling effect was enhanced with increasing radiation power density, but the IS had little dazzling effect.


Introduction
Laser dazzling interacts with the visual function of normal eyes through the stimulation of instantaneous strong light, * Author to whom any correspondence should be addressed. which makes people experience vertigo, vision decline and even temporary blindness. The principle is that strong light stimulation of the fundus of the eye will lead to the isomerization of the photosensitive chemical component of the retinal rhodesin chromophores, which is bleaching and decomposition, thus causing the human eye to lose or reduce the ability to respond to external light stimulation and cause a temporary loss of visual function, accompanied by psychological tension and even panic. In turn, the target's ability to act and react to the outside world is lost or reduced transiently [1,2]. In addition to the physical process of retina bleaching in response to laser dazzling may also cause visual transmission channel interference, visual synthesis dysfunction of the brain and psychological deterrence. Therefore, laser dazzling technology has wide application prospects in the public security field. Laser dazzling devices can cause temporary vision loss of the target without causing physical damage, which has become an important direction in the current research of high-tech nonlethal laser antiterrorism devices [3], laser radar or illuminations of car drivers [4][5][6][7]. At present, there are some reports about the development of laser dazzling devices and car driver illumination, but no specific parameters are involved. Therefore, it is of great significance to carry out in-depth research on the biological effects of laser dazzling.
A supercontinuum laser source not only has high intensity, high brightness and good directivity of the laser source but it also has the characteristics of the wide spectrum of the light source [8][9][10]. It has been used in basic science, industry, communication and medicine [11][12][13][14]. A supercontinuum laser has two advantages: first, the spectrum distribution is very wide, and traditional laser protection equipment can only filter a single or a few wavelengths of the laser and cannot achieve an effect of complete protection [6]. Second, the use of a low peak-power continuous or quasi-continuous laser, can effectively reduce the damage to the human eyes, not only increasing their safety but also appropriately increasing the amount of laser radiation to achieve a stronger dazzling effect. Therefore, supercontinuum lasers can be used in the research and development of laser dazzling devices, and they have important research value.
Current research mainly focuses on the dazzling effect of a single laser wavelength. Furthermore, charge-coupled device detectors are widely used to evaluate the laser dazzling of some special lasers [15][16][17][18] but these methods are undirected for evaluating the performance of laser dazzling. An electroretinogram (ERG) [19] was also used to reflect the biocurrent change under the high-intensity laser stimulus. It is a functional measurement tool for visual function changes [20,21]. Furthermore, it is also a useful method of laser eye treatment for panretinal photocoagulation [22][23][24]. Hence, it would be a useful tool for evaluating the biological effect of laser dazzling of different kinds of lasers.
This research focused on the quantitative evaluation of the biological effects of supercontinuum laser-induced dazzling. Two important parameters of the laser-induced dazzling were investigated, namely, the minimum radiation power density and maximum safe radiation power density. The former determines the lower limit of radiation power density that can cause the dazzling effect, and the latter determines the upper limit of radiation power density that will not cause permanent damage to the target retina. Moreover, the recovery time of animal retinal current signals after laser radiation was used to quantitatively evaluate the dazzling efficiency of each segment of the supercontinuum laser and explore the dose-effect relationship between the power density of each segment of the supercontinuum laser and the dazzling effect of the rabbit eye to provide a biological reference for the development of the supercontinuum laser dazzling device.

Experimental animals
A total of 60 rabbits were used in this research. Their body weight was approximately 5.0 kg. There was no abnormality in the fundus observed by ophthalmoscopic observation. These experimental animals were kept in single cages by the Experimental Animal Center of Beijing Institute of Radiation Medicine. These animals were allowed to adapt and maintained routinely for 3-7 d before the experiment. The experiment was carried out after no abnormalities were observed. The animal experiments with rabbits were performed in accordance with the Beijing Institute of Radiation Medicine Experiment Animal Center-approved animal protocols. All experiments were performed in accordance with the guidelines of the IACUC-DWZX-2019-502.

Laser radiation method
A supercontinuum laser (SuperK EXTREME EXW-12, NKT Photonics, Denmark) was used to irradiate the rabbit eyes. The spectral bands were divided into five bands: (a) visible spectrum (vs.): 550 nm-760 nm, which can simulate the laser that passes through a laser protective goggle, which blocks green and infrared lasers; (b) infrared spectrum (IS): 900 nm-2400 nm, which was the middle infrared light spectrum; (c) full spectrum (FS): 400 nm-2400 nm, which was the FS of the supercontinuum laser; (d) full spectrum 1 (FS-1): 400 nm-2400 nm, whose frequency was the same as the FS, but the weight of each spectrum is not the same, which can simulate a laser that passes through locomotive glass and driver's glasses; (e) full spectrum 2 (FS-2): 400 nm-2400 nm, whose frequencies were the same as the FS and FS-1, but the weights of each spectrum were not the same, which can simulate the laser that passes through locomotive glass.
The laser was transmitted by an optical fiber and then expanded laser beam and collimated through a lens, which was limited by a diaphragm. The shutter was used to control the radiation time, and LABMAX TOP (Coherent Inc., Santa Clara, CA USA) was applied to monitor the power density of the supercontinuum laser. A TES-1339 illuminometer was used to detect the ambient light illumination ( figure 1(a)). The experiment was carried out under dark adaptation conditions, and the ambient background illumination was set to be no higher than 0.01 Lux. The beam-limiting diaphragm was used to control the size of the laser spots that reached the pupil of the experimental animal; that is, the diameter of the aperture that was illuminated was set to be approximately 7.3 mm. During the experiment, the laser energy in the pupil of the animal was measured and calculated with a laser power meter before each radiation. The calculation formula was Q = Pt, H = 4Q πd 2 , in which the laser energy Q, power P, irradiation time t, laser irradiation power density H and laser irradiation spot diameter d.

Study on the safety of supercontinuum laser-induced dazzling
After the rabbit eyes were dilated by compound topical eye drops, the rabbits were fixed in the rabbit box, and the rabbit eyes were placed in front of the laser source. The dazzling laser was radiated directly into the rabbit eyes by frontal incident mode, and the power density increased gradually from small to large. Each power density was used to irradiate the rabbit eyes ten consecutive times with a radiation duration of 0.25 s. The fundus of the rabbit eyes was examined with ophthalmoscopes immediately after irradiation. The laser power density was adjusted, and the maximum dose without fundus injury in more than ten rabbit eyes was set as the upper limit of the macroscopic safe power density.

Evaluation of visual ERG-b wave for dazzling efficiency
Rabbit eyes were dilated by compound topical eye drops. The rabbits were anesthetized with Sumianxin injection and fixed in an electrostatic shielding box. The rabbit eyes were placed in front of the laser. The related electrode and electrode were fixed on the center of the auricle and forehead, respectively. The corneal electrode was placed in the eye in the conjunctival sac, and the electrode was confirmed to have good surface contact. The rabbit eye was adapted to the environment with an illuminance less than 0.01 Lux for 30 min. Then, the rabbit eyes were radiated with a white reference flash of 2 Hz at a frequency of one time per second before, during and after laser radiation. The flash and the laser were as close to the same axis as possible. After 0.1 s of dazzling light irradiation, the dynamic ERG (figure 1(b)) waves were collected according to the international standard method of ERG, and the amplitude changes of the ERG-b wave before and after laser radiation were measured. The transient disappearance or decrease in the amplitude of the ERG-b wave indicates the occurrence of the glare phenomenon, and the recovery time of the ERG-b wave is the time taken from the moment when the dazzling laser illuminates until the amplitude of the ERG-b wave returns to the preillumination level. The degree of laser dazzling is usually measured by the recovery time and amplitude of the ERG-b wave. The longer the recovery time of the ERG-b wave was, the lower the amplitude was and the more serious the dazzling effect was.

Statistical analysis
SPSS 13.0 software was used to analyze the data results, which are expressed as x ± s. One-way analysis of variance was used for comparisons between groups, and the LSD-t-test was used for further comparisons. The difference was statistically significant at P < 0.05.

Safety evaluation of eye injury caused by a supercontinuum laser under dark adaptation
Under dark adaptation, namely, the ambient background illumination was less than 0.01 Lux, each spectrum band of the supercontinuum laser was radiated to the rabbit eye ten consecutive times with each radiation duration of 0.25 s. In the range of different radiation doses, ocular fundus injury was observed by ophthalmoscopy immediately after the laser dose was radiated from low to high. When the laser power density of the vs. was 252 mW cm −2 , ten rabbit eyes were irradiated in total, and no damage was found in the fundus examination. Therefore, the dose was set as the upper limit of the macroscopic safe power density of the vs.. When the laser power density of the IS was 1184 mW cm −2 , 12 rabbit eyes were irradiated, and no damage was found in the fundus examination. Therefore, the dose was set as the upper limit of the macroscopic safe power density of the IS. When the laser power density of the whole spectrum segment was 1050 mW cm −2 , 14 rabbit eyes were irradiated, and no damage was found in the fundus examination. Therefore, the dose was set as the upper limit of the macroscopic safe power density of the whole spectrum segment. Higher than the above power densities of laser radiation in rabbit eyes could cause light white lesions or bleeding lesions.  figure 2. When the power density was between 0.20-1.60 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was less than 3 s. When the power density was 15.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 5.19 ± 0.40 s. When the power density was 60.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 6.65 ± 0.38 s. When the power density reached 180.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 10.80 ± 0.29 s, which was significantly higher than that in groups with a power density lower than 160.00 mW cm −2 , and the difference was statistically significant (P < 0.01). The above results indicated that the recovery time of the ERG-b wave was positively correlated with the increase in power density when the vs. of the supercontinuum laser was irradiated to rabbit eyes under dark adaptation. The power density of the dazzling laser was higher than 8.00 mW cm −2 , and the recovery time of the ERG-b wave was higher than 4 s; hence, it had a good dazzling effect.

Effect of the recovery time of the rabbit eye ERG-b wave
with the IS under dark adaptation. Under dark adaptation, the supercontinuum laser IS was radiated for 0.1 s. The recovery time of the ERG-b wave in rabbit eyes was less than 0.25 s when the power density was between 71.70 mW cm −2 and 521.00 mW cm −2 , which indicated that no obvious dazzling occurred. When the power density was 628.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 0.84 ± 0.09 s. These results showed that when the IS of the supercontinuum laser affected rabbit eyes, the ERG-b wave backed quickly after interference, and there was no obvious dazzle. We conclude that when the power density of the IS of the laser reached a very high level, visible laser filter residue after visible laser power density also reached a certain level, thus causing a very weak dazzling effect.

Effect of the recovery time of the ERG-b wave in rabbit
eyes with FS under dark adaptation. Under dark adaptation, the FS of the supercontinuum laser radiated the rabbit eye for 0.1 s, and the change in the ERG-b wave recovery time with power density in the rabbit eye is shown in figure 3. When the power density was between 0.02 and 0.98 mW cm −2 , the recovery time of the ERG-b wave was less than 2 s. When the power density was 2.50 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 2.84 ± 0.56 s. When the power density was 12.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 4.16 ± 0.55 s. When the power density was 117.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 7.54 ± 0.46 s. When the power density reached 1000.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 19.76 ± 0.90 s, which was significantly higher than that in the other groups (P < 0.01). The above results indicated that the recovery time of the ERG-b wave is positively correlated with the increase in power density when the FS of the supercontinuum laser was radiated to rabbit eyes under dark adaptation. However, when the power density increased from 2.50 mW cm −2 to 1000.00 mW cm −2 , the power density increased by 400 times, while the rabbit eye ERG-b wave recovery time only increased  from 2.84 s to 19.76 s, which was an increase of approximately seven times. This indicated that with the significant increase in the power density of the dazzling laser, the recovery time of the ERG-b wave in rabbit eyes showed a trend of slow increase. When the power density of the dazzling laser was higher than 12.00 mW cm −2 and the recovery time of the ERG-b wave was higher than 4 s, it also had a good dazzling effect.

Effect of the recovery time of the ERG-b wave in rabbit eyes with FS-1 and FS-2 under dark adaptation.
Under dark adaptation, when the FS-1 of the supercontinuum laser radiated the rabbit eye for 0.1 s, the change in ERG-b wave recovery time with power density in the rabbit eye is shown in figure 4. When the power density was between 0.06-0.24 mW cm −2 , the recovery time of the ERG-b wave was less than 1 s. When the power density was 4.60 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 4.50 ± 0.94 s. When the power density reached 93.30 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 9.82 ± 1.53 s, which was significantly higher than that in other groups, and the difference was statistically significant (P < 0.01).
Under dark adaptation, when the FS-2 of the supercontinuum laser radiated the rabbit eye for 0.1 s, the change in ERG b-wave recovery time with power density in the rabbit eye is shown in figure 5. When the power density was between 0.03 and 0.70 mW cm −2 , the recovery time of the ERG-b wave was less than 2 s. In the range of 1.10-3.00 mW cm −2 , the average recovery time of the ERG b-wave in rabbit eyes was 2.26 s, and there was no significant difference among all groups. When the power density was 5.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 3.81 ± 0.11 s. When the power density was 23.40 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 5.27 ± 0.10 s. When the power density reached 40.00 mW cm −2 , the recovery time of the ERG-b wave in rabbit eyes was 6.06 ± 0.19 s, which was significantly higher than that in the other groups, and the difference was statistically significant (P < 0.01).
The above results indicated that the recovery time of the ERG-b wave was positively correlated with the increase in power density when the FS-1 and FS-2 of the supercontinuum laser irradiated the rabbit eye under dark adaptation. The power density of the dazzling laser was higher than 4.6 mW cm −2 in the FS-1 and 5.50 mW cm −2 in the FS-2, and the recovery time of the ERG-b wave was higher than 4 s; therefore, they both had a good dazzling effect.

Discussion
Currently, ERG is the most commonly used method to study the biological effects of laser dazzling. The experimental animals, whose eyes are similar to human eyes, are rabbits and rhesus monkeys. It is objective, quantitative and simple to evaluate the effect of laser dazzling by recording the time required for the recovery of the ERG-b wave of the retinal response to the flash stimulus after laser radiation [25]. Here, we conducted in vivo experimental research to quantitatively evaluate the supercontinuum laser eye dazzling effect, which was all our work [26]. The results demonstrated that the recovery time was positively correlated with the radiation dose in a certain range, and the amplitude gradually returned to the preradiation level. The dazzling effect caused by laser flash did not cause organic damage to the eye tissue. Research on the biological effects of laser dazzling provides a biological experimental basis of equipment finalization and efficiency evaluation of laser dazzling devices or some other applications of lasers, and the direct evaluation method based on retinal ERG can provide direct and effective information about eyes' dazzling effect.
In this experiment, the vs., IS and FS of a supercontinuum laser were used to irradiate the eyes of rabbits, and the damage to the fundus was examined through an ophthalmoscope to analyze the upper limit of the safe power density of each spectral band of the laser without causing damage to the rabbit eyes. The experimental results showed that under the condition of dark adaptation, the upper limit of the macroscopic safe power density without damage in fundus examination was 252.00 mW cm −2 for the vs., 1184.00 mW cm −2 for the IS, and 1050.00 mW cm −2 for the FS. When the power density of the vs. was higher than 289.00 mW cm −2 , light white lesions or bleeding lesions were found in the fundus examination of rabbits, and when the power density of the FS was up to 1530 mW cm −2 , there were bleeding lesions in the fundus examination of rabbits. Accurate security dazzling power density limits also need further study, which did not lead to retinal pathological microstructure changes, such as local bulges, uneven thickness and disorders of the inner and outer nuclear layers, local inflammatory exudation and so on. However, when the maximum power density decreased from 0.1 to 0.5 times, according to the conventional evaluation methods, pathological changes in the microstructure were not observed. Therefore, as a preliminary study result, before further experimental studies, the reference value of the upper limit of safe power density under dark adaptation on the premise of giving priority to the principle of safety can be set as 25.20 mW cm −2 in the vs., 118.40 mW cm −2 in the IS and 105.00 mW cm −2 in the FS.
The results of visual electrophysiological evaluation showed that under the condition of dark adaptation, and the recovery time of the ERG-b wave was only approximately 0.8 s of the supercontinuum laser IS. Therefore, the dazzling effect was very weak, so the spectrum band could not be used as a dazzling spectrum segment. When the power density of the vs. was 17.00 mW cm −2 , it was close to the upper limit of the safe power density of 25.20 mW cm −2 , and the recovery time of the ERG-b wave was approximately 5.6 s. Therefore, this spectrum band had a good dazzling effect. When the power density of the FS was 117.00 mW cm −2 , it was slightly higher than the upper limit reference value of the safe radiation power density of 105.00 mW cm −2 , and the recovery time of the ERG-b wave was approximately 7.5 s, which is longer than that of the vs. segment. Therefore, the spectrum band had a better dazzling effect than the visible spectral segment. When the power density of FS-1 was 4.60 mW cm −2 , the recovery time of the ERG-b wave was approximately 4.2 s, and when the power density of FS-2 was 5.50 mW cm −2 , the recovery time of the ERG-b wave was approximately 4.5 s. Therefore, both spectral bands had a good dazzling effect. If the recovery time of the ERG-b wave was up to 4 s, the power density of the dazzling laser in the vs. of the supercontinuum laser was approximately 8.00 mW cm −2 , which is approximately 12.00 mW cm −2 in the FS, approximately 4.60 mW cm −2 in the FS-1, and between 5.00-5.50 mW cm −2 in the FS-2. According to the dazzling effect, under dark adaptation, the dazzling effect of FS-1 and FS-2 was stronger, and the dazzling effect of FS-1 was slightly stronger than that of FS-2, which was obviously stronger than that of FS. The dazzling effect of the vs. was between the three, while the dazzling effect of the IS was basically non-existent.

Conclusions
In summary, we investigated and evaluated the laser dazzling effect with the ERG-b wave. The results suggested that under dark adaptation, when the vs., FS, FS-1 and FS-2 of the supercontinuum laser irradiated rabbit eyes, the recovery time of the rabbit eye ERG-b wave was more than 4 s. All of them had a good dazzling effect, and the dazzling effect strengthened and increased with the radiation power density, while when the IS of the supercontinuum laser irradiated rabbit eye ERG-b came back quickly after the wave was interfered with, and there was no dazzling effect. The reference value of the upper limit of safe power density under dark adaptation on the premise of giving priority to the principle of safety can be set as follows: 25.20 mW cm −2 in the vs., 118.40 mW cm −2 in the IS, and 105.00 mW cm −2 in the FS. Both the FS and the vs. showed a good dazzling effect at the above radiation doses. Supercontinuum lasers not only achieve good dazzling effects but also have good safety. It has great potential to be developed into anti-terrorism and anti-riot nonlethal equipment. The results of this paper can provide a biological reference for the development and application of related technologies.