Elimination of virus-carrying insects, such as mosquitoes, by an efficient method is of primary importance to preventing the dissemination of infectious diseases and consequentially reducing the health and financial burden on human society. Research herein entails the design, characterization, and implementation of a structurally simple and cost-effective electro-optical system for mosquito-hunting, which is optimized with a system response time under 1.25 ms considering the average mosquito flight speed of 0.2777 m/s, capable of locating the free-flight adult mosquito and knocking it down, synchronously, without any post data-processing. Empirically, the visible-near-infrared absorption spectra of three mosquito species, namely, Culex piplens molestus, Aedes albopictus, and Armigeres subalbatus, were acquired to examine the disparity in photo-absorption property among different mosquito species and help determine an optimal wavelength for injuring the insects. Armigeres subalbatus, a natural transmission vector of filariasis to humans, was employed for the evaluation of the system's efficacy. By introducing a free-flight mosquito about 20 cm in front of an dichroic mirror that combines two optical beams for detection and eradication, a dynamically tracking photonic antenna with a maximal area of 45 mm by 39 mm can continuously track the insect and then knock it down by an instant exposure of a lethal beam with average energy ranging from 75 mJ to 155 mJ. Moreover, the dependence of the fate of the insects on the lethal beam energy dosage is statistically assessed. Overall, this research has successfully demonstrated the concept of the synchronizing scheme of identification and eradication with over 60% of mortality rate once the energy dosage is increased above 75mJ, and may be applicable to the control of other insects or avian animals.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Globally, adult mosquitoes are the primary vectors responsible for transmitting more than 100 kinds of viruses infecting different types of mammalian cells [1,2,3], which sicken at least 7 million people and claim over 1 million lives annually . Malaria and dengue, two of the major mosquito-borne infectious diseases, along, for instances, have reported annual cases of 219 million and 121.9 million in the year of 2017 and 2013, respectively [5,6]. Particularly, while the worldwide statistics indicate slight decrease in the reported cases of malaria from previous year alongside uprising trend in a 3-year period from 2015 to 2017, the cases of dengue fever have nearly doubled by a decade. Also, each of some recently well-known viruses such as ZIKA, west niles and many others, has caused several hundreds of thousands of infected cases in 2017 along [1,8].
The above-mentioned mosquito-borne diseases not only are lethal, but also have imposed tremendous economic and laborious burden on society at global scale, and thus, voraciously spur development of tools that are of vital importance for effective prevention control. Conventional methods such as the direct poisoning by a variety of insecticidal sprays and repellents [9,10], indirect trapping through the enticing to UV light  have already been utilized to dispel mosquitoes. However, long-term usage of these methods can either lead to growing insecticidal resistance or cause detrimental effects on human health and environment .
Recently, more advanced techniques of genetics-based sterilization , green synthesis of nanomaterial  and optics [15,16] have been explored for mosquito prevention. Particularly, on the front of optical techniques, observation has made marvelous advancement in extraction of RNA materials from the mosquito species of Aedes aegypti as well as in quantification of the headcount of sporozoite (spz), a viral unicellular vectors of mosquito-borne diseases, and its dynamic behavior [17,18], leading toward development of effective therapeutic medicine and methods. Likewise, through computational simulation, the flight behavior of mosquitoes including frequencies of wing flapping, aerodynamic pattern of its air-borne motion, and edge-leading fly-kinematics with small wings have been analyzed , conducive to behaviorally controlling viral vector-loaded adult mosquitoes.
Furthermore, the deployment of camera-based monitoring system in several studies reported investigation of the insects’ flight behavior and habits in hope of improving the strategies of mosquito prevention. Gibson et. al.'s study, for example, utilizing 3D video recording and illumination of white light LED revealed the alteration of anopheles coluzzi‘s flight height and its frequent ground-dropping behavior while approaching the host-odor amended bright object , and another investigation found that the majority of anopheles gambiae approached bed nets with reduced flight velocity above the occupants . Atop the functions of monitoring and identification, a system can also carry out eradication of the insects using a laser beam with an average power intensity of 440 W/cm2, and the total system response time is at least 67 ms .
Besides some sensitive CCD sensors, the above real-time studies not only necessitate complicated processing and filtering of noise and low-frequency signals but also require adequate computational capability for the storage and processing of motion video acquired at least 50 frame-per-second (fps), which may be costly and unaffordable for public utilization.
In this work, to relieve the stringent requirement for computation, storage, analysis and system power consumption while satisfying the average flight speed of mosquitoes of 0.277 m/s , we present a simpler, more efficient mosquito-hunting electro-optic system (MHEOS) centered around a central processing unit that enables electronic synchronization with a system response time well under 5 ms alongside a focused beam spot diameter less than 2 mm upon the in-flight insects. This system can identify the presence of free-flight mosquitoes within a distance of less than 300 mm using a 2 dimensional (2D) photonic antenna that has an active area of 45 cm x 39 cm, dynamically track its flying path, and injure the insects by an instant exposure of a lethal laser beam via an optical reflectance-based positive feedback system. In principle, such cascaded continuously tracking and annihilating actions until the knock-down of the free-flight mosquito is done. Experimentally, the absorption spectra of three mosquito species, Culex piplens molestus, Aedes albopictus and Armigeres Subalbatus were acquired to examine the disparity of photo-absorption property among different mosquito species, and help determine the optimal wavelength for identification and subsequent annihilation, which is followed by construction, characterization, and evaluation of the system's efficacy.
2. Method and materials
2.1 Culturing mosquitoes
The mosquito species of Culex piplens molestus, Aedes albopictus and Armigeres Subalbatus were obtained from the Southern District of Taiwan’s Center for Disease Control (CDC). All mosquitoes used in the experiment were emerged from the pupa stage about two-weeks after full-development of wings. During the period of larval stage, the larvae were fed with the congee-like mixture of porcine liver powder and yeast, in one-to-one ratio, in water, whereas, 10% sugarcane juice was used to feed the adult insects which were raised in an incubation room conditioned with temperature of 25 oC, humidity of 70% and 12:12 hours in light-and-dark cycles.
2.2 Absorbance measurement
The visible-near infrared (VIS-NIR) photo-absorption property of mosquito species including Culex piplens molestus, Aedes albopictus and Armigeres Subalbatus were characterized using a VIS-NIR spectroscopy (UV-1800, Shimadzu, Japan). The sample preparation for the absorbance measurement were carried out by packing one hundred individuals of each species in a four-side polished quartz cuvette with the dimension of 12.5 mm x 2.5 mm x 45 mm. The sample-containing cuvette, shown in the inset photo of Fig. 2, was fitted into the chamber holder of the spectrometer, illuminated by a broadband light source with the wavelength ranging from 400 nm to 1000 nm, and the measurement of spectral absorbance, A=-log (Io/I1) where Io and I are incident and transmitted intensities, respectively, was numerically extracted.
2.3 Architecture of MHEOS
Figure 1 illustrates the schematic diagram of a MHEOS system that constitutes two sub-modules synchronously linked by a positive feedback electronic circuitry for probing and knocking down mosquitoes. At the heart of the probe arm is an optical beam from laser diode that operates at the wavelength of 635 nm and with an output power of 200 mW. Specifically, the red optical beam passes through a BS, and is focused by L1 and steered by 2D GS-A onto a free-flight mosquito. The reflected beam off the body of the mosquito is collected and collimated by L1, and then gets re-routed by BS toward a BPF-shielded PD, which is used to minimize the spectral noise.
Subsequently, the PD induces MRS when the reflected beam impinges upon the surface of PD, and then gets fed to SPU unit for synchronization with the lethal beam arm via administration of PIS. The setup of the lethal beam arm that activates LBS is similar to that of the probe arm except the utilization of a laser diode operating at the wavelength of 450 nm and with an output power of 3.2 Watt, which is chosen for its significantly higher average power than the one operating at 405 nm.
2.4 Assessment of mosquito-mortality
The assessment of the MHEOS's efficacy on knocking down Armigeres Subalbatus, which is the species used throughout the shooting session, was implemented by introducing the mosquitoes, one at a time, into the backside of a cage system comprised of a cylindrical plastic transparency with a length of 100 mm and a pair of front-and-back disks with high transmittance in visible wavelength. The cage's light entrance is placed 200 mm away from to the GS-A. By manually vibrating the cage, the mosquito, either being stationary or in the state of free-flight, gets disturbed and flies over the sight of view of the photonic antenna for the system assessment, which is monitored by a camera looking from the top of the cage. A group of 10 mosquitoes for each of the following laser energy dosage, 45 mJ, 75 mJ, 115 mJ and 155 mJ was examined to garner statistical data. Possible conditions of lethal beam-illuminated mosquitoes applied to the categorization of data collection include survival with flight ability, injury with slightly impaired flight ability, and injury with impaired flight ability with/without obvious damage in appearance, which correspondingly account for the ones with unaffected flying track and pattern, disturbed flying track and pattern with the capability of only a short distance flight, and immediate knock-down.
3.1 Absorbance measurement of mosquitoes
Photo-absorption properties of biological tissues is of critical importance to MHEOS’s functions in detecting and annihilating mosquitoes, and can vary slightly from species to species due to the compositional difference among tissues and the relevant secretionary molecules. Figure 2 illustrates the absorbance spectra of three mosquito species. As can be evidently seen from the figure, the absorbance of all mosquito species rises steadily at the edge of blue wavelength while gradually getting lower as the wavelength increase toward the NIR regime, thereby helping justify the laser wavelength of 635 nm and 450 nm for identification and destruction of the insects, respectively. The species of Armigeres subalbatus was used for calibration and evaluation of the MHEOS, considering its relatively higher absorbance than the other two species.
3.2 Analysis of the system time response of MHEOS
Response time between the activation of GS-A and GS-B is crucially important to catching up with the velocity of free-flight mosquitoes. A series of time-delay experiment was implemented by setting up two signal path, a control reference in that signal are sent directly from a signal generator to oscilloscope and the processed that passes through an electronic block such as logic switch, signal amplifier or peak detector. Figure 3 depicts time-course traces of delay from one signal point to another. In Fig. 3(a), the response time of 1.114 ms comprises the activation time of GA-B and the response time of PD mimicking the impingement of the mosquito-borne back-scattered light upon the PD. Additionally, a pre-stage amplifier serving the purpose of boosting signal level is necessitated for the triggering of the logic switches, and yields 0.00103 ms in delay as presented in Fig. 3(b). To activate GS-B for a mosquito kill, passage of PIS over some logic switches and a pre-stage amplifier necessitates a total delay of 0.132 ms, which, as can be seen in Fig. 3(c), is measured from the stand-point of amplified MIS to the amplified LBS. Therefore, the estimates of the total time delay from the detection to destruction of the insect for X- and Y-scanners are 1.905 ms and 1.249 ms, correspondingly; the flow chart of system time-response is summarized as shown Fig. 3(d).
3.3 Operation of SPU
Since the operation of MEHEOS lays upon the synchronization of GS-A and GS-B through SPU with a very short (0.132 ms) time-delay as shown in Fig. 3(c). Figure 4 presents a more-detailed working principle of SPU. As shown in the left part of Fig. 4, while GS-A searches for the mosquito, GS-B is idling waiting for internal command. After an in-flight insect is identified, the mosquito-borne back-scattered light off the insect’s body is relayed to and detected by PD. With the MRS being induced by PD and amplified by a signal amplifier, a logic switch can decide on the passage of PIS depending on the single level of the amplified MRS. Once the logic of the switch is switched on to “true”, the SPU retrieves the PIS of the present X and Y position of the mosquito from the linear triangle waveforms of GS-A. The PIS is then passed on to a peak detector that sends it to GS-B when the rising edges of the triggered square pulses is detected, and each of the pulses lasts for 50 ms, corresponding to the dwelling duration of the lethal beam on the insect’s body as well as the temporal interval between the pulses. After the PIS passes through the peak detector, it is amplified and referred to as LBS that commands GS-B for positioning its mirrors, directing the focused lethal beam onto the body of the insect for a kill. Such scheme of the system architecture is adapted for x and y mirrors of both GS with the exception that an extra amplifier for a wider scanning range along y-direction is implemented.
3.4 Dynamic tracking and shooting
To demonstrate the practicality of the MHEOS as a mosquito hunting machine, a free flight Armigeres Subalbatus was introduced into the see-through cage. The free-flight path of the insect right above the threaded hole of the pedestal post can be clearly revealed by the scattered optical beam and recorded by a smartphone (iPhone 6) with a temporal resolution of 24 frames per second (fps). Figure 5(a-d) show the selected time sequence of images and Visualization 1 shows a complete video as the supplementary file. In Fig. 5 and Visualization 1, a 635 nm laser and continuous-scanning scanning mirror (2D GS-A in Fig. 1), which forms red-rectangular pattern, was used to dynamically track an Armigeres Subalbatus. Typically, the Armigeres Subalbatus are rested on the transparent cylinder. Once a mosquito was flying into the scanning area, another scanning mirror (2D GS-B) was directed to the position of the mosquito and shoot it down. In this demonstration, the power of the 445 nm damage laser was much higher than the detection 635 nm laser. Thus, a colored glass filter (FGL 515, Thorlabs) was placed in front of the CCD camera of iPhone 6 to enhance the image contrast. Figure 5 and the supporting video demonstrate that the proposed approach in this work is effective and capable of eliminating the free-flight insect.
3.5 Brightfield images of examined mosquitoes
The assessment of the system’s efficacy, including detection sensitivity, effectiveness of the lethal beam on mosquitoes, and system synchronization, were carried out by introducing a free-flight mosquito into a transparent cage situated in front of the MHEOS. Figure 6(a-d) illustrates some examples of resultant brightfield images of the mosquitoes after exposed to the lethal beam. As can be seem from the figures, mosquitoes survived the exposure of the lethal beam appear to be null apparent damage, whereas, the knocked-down mosquitoes were observed with either no obvious damage or broken body parts such as the seemingly clipped wing indicated by an red arrow.
3.5 Fatality analysis
To garner a statistical analysis on the energy dependence of lethality rate, the mosquitoes, were dosed with an array of energy, the results of which are depicted in Fig. 7 presenting the fate of mosquitoes in percentage for the designated categories. Here the morality rate accounts for the percentage of the mosquitoes that either have lost the regular flying pattern and can only fly a short distance, or are completely deprived with flying ability on the first illuminated instances. Note that while the survival rate monotonically decreases as the energy dosage increases, the percentage of injured insects with flight ability rises and plateaus beyond 110 mJ suggesting higher energy dosage enables better chance for effective damages on the insects. Likewise, the percentage of the insect with impaired flight ability falls in between 30% to 40% once the energy dosage is increased above 75mJ.
Being able to precisely monitor and track the presence of mosquitoes is critically important for an effective knock-down with proper energy dosage. The working principle behind the operation of MHEOS is laid upon the synchronization of two laser beams directed by two galvanometers, one for identification and tracking, and the other for eradication. Previous study has demonstrated that a different combination of variant laser properties such as the wavelength range from 532 nm up to 10 µm, pulse duration and energy, laser fluence, beam diameter, can lethally damage stationary mosquitoes anesthetized with CO2 gas . To properly select the wavelength for the two laser beams, photo-absorption measurement reveals that, regardless of the species of the insects, the range of blue wavelength imposes stronger absorption over other range of wavelength. One photo-absorption experiment indicates that a single exposure of blue wavelength upon the insects may impose higher mortality rate due to the trapping of the accumulated heat in the insect’s body that denaturizes the internal biological tissues , and helps establish the corresponding wavelength for detection and elimination to be 635 nm and 450 nm. While 2D photonic antenna established by GA-A is constantly in activation for sensing any free-flight mosquitoes, GA-B stands waiting for incoming commands to eliminate the insects. Although the current detection breadth of 300 mm might be considered short, the effective detection distance between the insect and the mirror set can be further extended by incorporating a lock-in amplifier that helps pick up small signal level of optical reflectance by improving the signal to noise ratio (SNR), experiment of which is still ongoing in our laboratory.
From the perspective of the flight speed of mosquitoes, which, in the case of Anopheles atropavus, one of the primary vectors of malaria, has an average flight speed of 0.2777 m/s before or after being fed with blood meals , the total response time from identifying an free-flight mosquito to activating GA-B translates into 0.529 mm and 0.346 mm in positional latency of a renewed cycle of detection for x- and y-axes, respectively. Considering the typical dimension of Armigeres subalbatus utilized the present assessment, 7.5 mm and 1.875 mm in the corresponding length and width, the positional latency of the renewed detection beam spot have 7% and 18% of the respective shift along x and y direction, suggesting the effectiveness of the temporal response of the system to tracking and eradicating the free-flight mosquitos.
Likewise, to evaluate the system’s consistency, a statistical study on the mortality rate of mosquitos was carried out by dosing free-flight Armigeres subalbatus with an array of energy. These energy were chosen to keep the duration of illumination within the time interval of pulses of the logic switch that enables the peak detector to transfer PIS signal to lethal beam unit. Based on the brightfield optical images, the results of the beam-mosquito interaction can be designated to four possible categories of outcomes shown in Fig. 5. Particularly note that the apparent damage on the insects indicated by the red arrow can be the direct evidence of lethal beam-induced injury, whereas in some cases the damages are not apparent implying that photo-thermal energy absorbed by the insects’ body parts may be accumulative, leading to the injury due to overheating.
As statistically presented in Fig. 6, the survival rate decreases monotonically as the energy of the lethal beam increases, and more than 60 percent of the mosquitoes were injured when the laser energy tops 75 mJ, which accounts for the percentage of the mosquitoes that either have lost the regular flying pattern and can only fly a short distance, or are completely deprived with flying ability on the first illuminated instances. Additionally, although the percentage of the injured ones with impaired flight ability undulates between 30% to 40% when setting the energy between 75 mJ to 155 mJ, around 50 percent had lost the normal flying track and pattern, and were capable of only a short distance fly, implying that over 60 percent of the insects can be effectively knocked down by the MHEOS.
In summary, this research demonstrates the design, implementation and real-world application of a mosquito-hunting electro-optic system that bases upon the synchronization of the sensing of free-flight object and an instant exposure of lethal beam for annihilation, and well above 60 percent of the insects were effectively knocked down either alive or dead. By customizing the speed of system response and focusing optics, this system may be applicable to the control of other insect species, just to name a few, such as grasshopper, fruit flies and ants, detrimental to agricultural products. Finally, the cost of the system can be further brought down, for example, by substituting the galvanometers with micro-electro mechanical system (MEMS) devices which can be fabricated in mass, and thus bringing down the prices. Also, fabrication of a monolithic optical system that integrates dichroic beamsplitter, lenses and mirrors is also an option to further bring down the cost.
This research proves the concept of the synchronized linking of continuous tracking and damaging upon the free-flight mosquitoes utilizing suitable wavelength which are determined by assessing mosquitoes’ photoabsorption properties. The MHEOS was constructed, characterized with its system time responses, and applied to mosquito assessment. The dependence of the mosquitoes’ immortality rate on the energy of the blue lethal beam is investigated, and it was found that the projection of optical energy dosage below and above 75 mJ upon the mosquitoes can cause erratic flying track and pattern as well as serious impairment of flight ability, respectively. Therefore, a conclusion is drawn that this system is effective in harming the insects with the success rate above 60%, which may be further improved, for instance, by the optimization of the energy density of the blue lethal beam through the shaping of focal volume.
Jian Ji Chen He (CGCH) Foundation for Education.
1. A. Sharma and S. K. Lal, “Zika Virus: transmission, detection, control, and prevention,” Front. Microbiol. 8, 110 (2017). [CrossRef]
2. R. W. Snow, C. A. Guerra, A. M. Noor, H. Y. Myint, and S. I. Hay, “The global distribution of clinical episodes of plasmodium falciparum malaria,”,” Nature 434(7030), 214–217 (2005). [CrossRef]
3. B. Wahid, A. Ali, S. Rafique, and M. Idrees, “Global expansion of chikungunya virus: mapping the 64-year history,” Int. J. Infect. Dis. 58, 69–76 (2017). [CrossRef]
4. H. Caraballo and K. King, “Emergency department management of mosquito-borne illness: malaria, dengue, and west nile virus,” Emerg. Med. Pract. 16(5), 1–23 (2014).
5. World Health Organization (WHO) world maliaria report 2018.
6. J. D. Stanaway, D. S. Shepard, E. A. Undurraga, Y. A. Halasa, L. E. Coffeng, O. J. Brady, S. I. Hay, N. Bedi, I. M. Bensenor, C. A. Castaneda-Orjuela, T. W. Chuang, K. B. Gibney, Z. A. Memish, A. Rafay, K. N. Ukwaja, N. Yonemoto, and C. J. L. Murray, “The global burden of dengue: an analysis from the global burden of disease study 2013,” Lancet Infect. Dis. 16(6), 712–723 (2016). [CrossRef]
8. N. B. DeFelice, E. Little, S. R. Campbell, and J. Shaman, “Ensemble forcast of human West Nile virus cases and mosquito infection rates,” Nat. Commun. 8(1), 14592 (2017). [CrossRef]
9. R. S. Lees, J. R. L. Gilles, J. Hendrichs, M. J. B. Vreysen, and K. Bourtzis, “Review: improving our knowledge of male mosquito biology in relation to genetic control programmes,” Acta Trop. 132, S2–S11 (2014). [CrossRef]
10. A. M. Pohlit, N. P. Lopes, R. A. Gama, W. P. Tadei, and V. F. Neto, “Patent literature on mosquito repellent inventions which contain plant essential oils-A Review,” Planta Med. 77(06), 598–617 (2011). [CrossRef]
11. Y. N. Liu, Y. J. Liu, Y. C. Chen, H. Y. Ma, and H. Y. Lee, “Enhancement of mosquito trapping efficiency by using pulse width modulated light emitting diodes,” Sci. Rep. 7(1), 40074 (2017). [CrossRef]
12. J. Hemingway and H. Ranson, “Insecticide resistance in insect vectors of human disease,” Annu. Rev. Entomol. 45(1), 371–391 (2000). [CrossRef]
13. C. F. Oliva, D. Damiens, and M. Q. Benedict, “Male reproductive biology of Aedes mosquitos,” Acta Trop. 132, S12–S19 (2014). [CrossRef]
14. B. Giovanni, “Green synthesized nanoparticles in the fight against mosquito-borne diseases and cancer- a brief review,” Enzyme Microb. Technol. 95, 58–68 (2016). [CrossRef]
15. M. D. Keller, D. J. Leahy, B. J. Norton, E. Johanson, E. R. Mullen, M. Marvit, and A. Makagon, “Laser induced mortality of Anopheles stephensi mosquitoes,” Sci. Rep. 6(1), 20936 (2016). [CrossRef]
16. E. R. Mullen, P. Rutschman, N. Pegram, J. M. Patt, J. J. Adamczyk Jr, and E. Johanson, “Laser system for identification, tracking, and control of flying insects,” Opt. Express 24(11), 11828–11838 (2016). [CrossRef]
17. J. Vanderberg, “Imaging mosquito transmission of Plasmodium soprozoites into the mammalian host: immunological implications,” Parasitol. Int. 63(1), 150–164 (2014). [CrossRef]
18. Y. S. Hong, S. Kang, M. Han, G. N. Gobert, and M. K. Jones, “High quality RNA isolation from Aedes aegypti midguts using laser microdissection microscopy,” Parasites Vectors 4(1), 83 (2011). [CrossRef]
19. R. Bomphrey, T. Nakata, N. Phillips, and S. M. Walker, “Smart wing rotation and trailing-edge vortices enable high frequency mosquito flight,” Nature 544(7648), 92–95 (2017). [CrossRef]
20. F. Hawkes and G. Gibson, “Seeing is believing: the nocturnal malarial mosquito Anopheles coluzzii responds to visual host-cues when odour indicates a host is nearby,” Parasites Vectors 9(1), 320 (2016). [CrossRef]
21. J. E. A. Parker, N. Angarita-Jaimes, M. Abe, C. E. Towers, D. Towers, and P. J. McCall, “Infrared video tracking of anopheles gambiae at insecticide-treated bed nets reveals rapid decisive impact after brief localized net contact,” Sci. Rep. 5(1), 13392 (2015). [CrossRef]
22. C. Kaufmann and H. Briegel, “Flight performance of the malaria vectors Anopheles gambiae and Anopheles atroparvus,” J. Vector Ecol. 29(1), 140–153 (2004).
24. W. F. Cheong, S. A. Prahl, and A. J. Welch, “A review of the optical properties of biological tissues,” IEEE J. Quantum Electron. 26(12), 2166–2185 (1990). [CrossRef]