Method for smart sex-sorting of immature-stage mosquitoes and apparatus for implementing same

Abstract

Within the framework of health policies for monitoring human and animal epidemiological diseases in environments where different mosquito species proliferate as endemic species or appear as non-native invasive species, the smart mosquito sex-sorting device, which uses the sexual dimorphism of pupa size of certain mosquito species, automatically separates mosquito pupae by sex using artificial intelligence (AI) by analysing images with algorithms to identify the sex, thus optimising the volume of males recovered. For this purpose, successive steps are performed, using a tank of pupae and larvae to be sexed, a batch dosing device, a pupae dispenser with a flow divider shaft (12), a pupae calibrating device containing an optical camera for acquiring a digital image, and a distributor arm for depositing the sorted insects into containers.

Description

[0001] METHOD FOR THE INTELLIGENT SEX CLASSIFICATION OF IMMATURE STAGES OF MOSQUITOES AND APPARATUS FOR ITS APPLICATION

[0002] DESCRIPTION

[0003] TECHNICAL SECTOR

[0004] This patent relates to a method for sexing immature mosquito stages, as well as the apparatus for its application, which we call a sexer. It is, therefore, an invention in the field of biotechnology related to the sexing or screening of invertebrate animals, such as insects of the order Diptera and suborder Nematocera, including those belonging to the family Culicidae, commonly known as mosquitoes.

[0005] Mosquitoes go through a life cycle of egg, larva, pupa, and adult. Specifically, this invention relates to a method for identifying or characterizing the sex of each individual before it reaches adulthood, specifically during the pupal stage. This is done to separate males and females, an essential step in raising large numbers of mosquitoes for the application of various genetic control techniques. These vector insects are targeted within the framework of public health policies aimed at controlling human and animal diseases in environments where they proliferate as endemic species or appear as invasive, non-native species.

[0006] In summary, the method and its implementing apparatus are based on exploiting the sexual dimorphism in pupal size of certain mosquito species, enhanced through image analysis with developed algorithms based on the application of artificial intelligence techniques. The use of this artificial intelligence (AI) allows for greater efficiency and accuracy in mosquito sexing. BACKGROUND OF THE INVENTION

[0007] We are dealing with an emerging and interesting technical development due to its relevance to epidemiology and the control and eradication of harmful insects, particularly certain mosquito species that are vectors for transmitting diseases to humans and animals. Therefore, we will explore various methods for classifying insects according to their evolutionary stages and genus, whether by mechanical means or more sophisticated methods using optics, computer science, and other emerging technologies.

[0008] It is particularly interesting to note how the state of the art is predominantly based on US inventions, such as the worldwide patent WO18134829, which took precedence over several US patents, granted by SENECIO LTD (2018), for a method and apparatus for sexing mosquitoes. Sexing is performed on adult insects, identifying them through image analysis and then using a robotic arm to separate the specimens. Also noteworthy is the US patent US9538739 from INNOVATION CONSULTANTS LLC (2017), which employs a laser-based device for controlling specimens to damage and reduce the population of virus-carrying mosquitoes. And the US patent US11744235 from VERILY LIFE SCIENCES LLC (2023) describes a sieving device with openings for mosquito pupae based on their size.The same patent holder has published US patent 10178857 for a camera with an inclined ramp and openings (2019), and the more sophisticated US patent 11793172 for detecting one or more insects when present within a test chamber using reflected signals indicative of Doppler changes caused by the movement of one or more insects (2023). His prior patent, US patent 10798924, describes an apparatus with a sensor that can determine the sex of an adult insect and a shutter to control access to a container (2020). Regarding an automated mass rearing system that can facilitate egg hatching, feeding, and monitoring of larval growth and pupal emergence, the same holder has published US patent 2023210097, also in 2023.

[0009] Also of interest is US patent OXITEC LTD (2015), which describes a device for classifying arthropod larvae and pupae based on size, particularly in mosquitoes, using two parallel bars pivotally mounted on a frame where the insect larvae and pupae are separated by the grid between the bars. Also of interest is US patent MOSQUITOMATE INC (2023), which describes a method for sexing mosquitoes using a plate with multiple water-filled wells. An image of each mosquito is captured in each well, and each image is analyzed to determine its sex. Mosquitoes have distinct anatomical areas for identification, which are captured in the images and used to determine whether a mosquito is male or female.

[0010] Another characterization and classification system results from the North American patent US2021329903 of ISCA TECH INC and the UNIVERSITY OF CALIFORNIA (2021) relating to an apparatus and method for classifying flying insects by recording the frequency of wing flapping and secondary and complementary factors such as the precise circadian rhythm for the time of year, current environmental conditions and the speed or direction of flight.

[0011] In other regions, we find the developments of European patent EP4018372 by DIPTERA Al LTD (2022) with an electro-optical module for identifying specimens passing through a tube. EXPLANATION OF THE INVENTION

[0012] According to scientific reports published and disseminated by the World Health Organization (WHO) of the United Nations, major vector-borne diseases collectively account for approximately 17% of the estimated global burden of communicable diseases and claim more than 700,000 lives each year. Vector-borne diseases, primarily those transmitted by mosquitoes, impose an immense social and economic burden on the economies of many countries and hinder both rural and urban development.

[0013] Some mosquito species of the genus Aedes are invasive species and vectors of human diseases caused by arboviruses. The Aedes aegypti and Aedes albopictus mosquitoes are the main vectors of the viruses that cause dengue, Zika, and chikungunya, while the Aedes polynesiensis mosquito is the main vector of lymphatic filariasis.

[0014] In a context of increasingly stringent regulation of approved insecticide molecules and the spread of insecticide resistance among mosquito populations, new, environmentally friendly technologies are needed to control invasive mosquitoes. Many of these new mosquito control technologies are genetic control techniques, including the Sterile Insect Technique (SIT), the Incompatible Insect Technique (IIT), the replacement of natural mosquito populations with mosquitoes carrying the Wolbachia bacteria, which can block virus transmission, and genetic forcing techniques. All of these require the laboratory rearing of large numbers of mosquitoes and their subsequent sexing, either to release sterile male mosquitoes into the field or to facilitate rearing procedures.When sterile / incompatible male mosquitoes are released as a method of controlling wild populations, it is vitally important to ensure that there are no females among the released insects, as it is the females that bite and are therefore potentially disease carriers.

[0015] The intelligent mosquito sexer described in this patent allows for the automated separation of mosquito pupae by sex, using artificial intelligence (AI) to identify the sex, optimizing the volume of males recovered and minimizing the contamination of females.

[0016] The proposed invention, the intelligent mosquito sexer, is composed of the following stages, as can be seen in Figure 1:

[0017] Stage. A.- Tank containing the larvae and pupae to be sexed. This is a water container (1) with an overflow (2) and a bubbling system (3) at the base regulated by a solenoid valve (4) that keeps the pupae and larvae moving and ensures the homogeneity of insects throughout the water volume, thus avoiding surface agglomerations where pupae and larvae tend to concentrate to breathe.

[0018] Stage B. Batch Dosing: The purpose of this stage is to create batches containing a number of insects that can be processed in a single operation by the pupa size separator in the subsequent stage. The pupa size separator will be referred to hereafter as the “pupa calibrator” or “calibrator.” The pupae and larvae (18) mixed in water are drawn into a smaller capacity regulating vessel (10) in which negative pressure has been created by a vacuum pump (9). The solenoid valves (5, 6, 7, 8, 11) of the pipes connected to this regulating vessel control the inflow of pupae, air, water, vacuum, and outflow of pupae, allowing the filling of predetermined volumes of water and subsequent emptying to the next stage. Stage C. Pupa Dispenser: The purpose of this stage is to introduce pupae into the upper part of the main body of the calibrator, the next stage.The pupae must be distributed evenly throughout the entire intake area of ​​the calibrator. To achieve this, the proposed system uses a tree-shaped distributor (12), which eliminates the need for additional moving mechanical systems. This distributor allows for the even distribution of the pupae and larvae by dividing the water flow into numerous channels, as many as necessary, which flow into the inlet points of the calibrator's pupa intake area. These channels cover the entire width of the calibrator and distribute the pupae and larvae uniformly along its length. Figure 2 shows several possible configurations of the tree-shaped dispenser.

[0019] Although a uniform distribution of pupae across the width of the sorting surface is generally desirable, sometimes it may be more advantageous to concentrate a greater number of pupae in the central area rather than along the edges, following a normal distribution. In these cases, a dispenser based on the Galton grid concept can be used, as shown in Figure 3.

[0020] Stage D.- Pupa Calibrator: This is a mosquito pupa separator by size based on the classic Fay-Morlan glass plate separator first described by RW Fay and HB Morlan (1959), which takes advantage of the sexual dimorphism of the size of the pupae of various mosquito species, including Aedes aegypti, Ae. albopictus, Ae. polynesiensis, Culex quinquefasciatus and others.

[0021] In the Fay-Morlan mechanical device, mosquito pupae are introduced from above into the groove formed between two vertical glass plates. The plates are nearly parallel, but the position of one can be manually adjusted by the lab technician using four screws fixed to the corners of the plate. This changes the angle between the two plates and alters the gap between them. Consequently, the pupae are sorted vertically according to their size and the angle formed by the two glass plates. Smaller pupae penetrate further into the gap, while larger ones are trapped at a higher point. On average, male pupae are smaller than female pupae, resulting in a horizontal cloud of female pupae wedged at the top of the glass plate and a second horizontal cloud of male pupae wedged at the bottom.Furthermore, there is often an overlap in the size distribution of the populations of both sexes, resulting in an intermediate zone where smaller-than-average female pupae and larger-than-average male pupae can be found simultaneously. The objective of the operation is to classify the pupae into two batches. One batch contains almost exclusively males with a residual number of females, referred to as the female contamination rate. The second batch contains mainly females and a certain percentage of males of a similar size to the females that have been discarded to maintain the female contamination rate at the desired level.

[0022] The new smart mosquito sexer also takes advantage of the size difference between mosquito pupae for separation, but incorporates several components that allow the application of artificial intelligence to optimize the process. Figure 4 shows a diagram of the different components of the pupa calibrator. Unlike the Fay-Morlan device, the main body of the pupa calibrator forms an internal cavity closed by a front cover, through which water flows. This cavity houses an internal plate that moves freely in three-dimensional space, which we will henceforth refer to as the "floating plate" or "moving plate." The front of this calibrator cavity is closed by a transparent plate fixed to the main body with screws or another type of fastener and sealed with a waterproof gasket to prevent water leakage.Between the fixed front plate and the floating plate, the separation and angle necessary for the sex separation mechanism of the pupae, described above, are formed.

[0023] The position of the floating plate in three-dimensional space is electronically controlled by a set of levers and linear or angular actuators, which may be electric, hydraulic, pneumatic, or any other type that allows precise control of the linear position from a digital or analog signal. The actuators are located outside the main body of the gauge, at the rear. They have an axial spacer that passes through the gauge housing via a hole sealed with a gasket, or similar, acting as a retainer. The tip of the spacer is in contact with the rear surface of the floating plate and can push it forward with great precision. In turn, the floating plate has levers embedded in it that tend to push it backward, attracted in turn by another set of levers embedded in the fixed housing.Other types of mechanisms can be implemented to control the position of the floating plate, such as actuators directly attached to the plate via a ball joint, universal joint, or similar connection. Alternatively, mechanisms based on the use of springs can be implemented to create the same push / pull effect of the floating plate against the actuators.

[0024] The pupae / larvae enter the sealed cavity through one or more entry points on the top of the grader body. They exit through one or more discharge points located at the opposite end. This creates a uniform water flow between the inlet and outlet, allowing the pupae / larvae to move freely within the space formed by the grading surfaces: the front cover and the floating plate.

[0025] At the front, the system features a high-resolution optical camera facing the transparent front plate, which captures an image of the batch of pupae / larvae that have been introduced into the main body of the grader. An image analysis algorithm, based on artificial intelligence, individually identifies the pupae, analyzes them, and assigns a sex to each one, initially based on its position on the sorting surfaces. Alternatively, the sex of the pupae can be determined based on differentiating morphological characteristics, regardless of their position on the sorting surface, allowing for a potential dual sexing method.Similarly, a larval analysis algorithm can be implemented to identify and sex larvae based on any distinguishing morphological characteristics between males and females, including in the case of genetically sexed mosquito strains that can be developed in the laboratory. The use of an artificial intelligence image analysis system can also be useful in detecting anomalous cases and / or situations (recombinant individuals, inappropriate pupal age, etc.). In general, the AI-based vision and image analysis system allows for the identification of individuals, pupae, and larvae, the determination of their relative position on the classification surfaces, and the identification of any detectable characteristics for each individual based on image analysis (color, morphological details, etc.).The system can incorporate additional lighting subsystems, at different wavelengths, and the optical camera can incorporate filters that allow algorithms to identify features that are even invisible to the human eye.

[0026] Image analysis and floating plate positioning are performed for each batch of pupae that is introduced into the classifier, adjusting the classification threshold to the actual size of the pupae in the specific batch being classified and thus optimizing the recovery of males and the contamination rate of females.

[0027] In addition to mechanical size sorting, a device for discarding individuals (either pupae or larvae) can be added. This device consists of a laser beam or any other electromagnetic or sound wave capable of generating directional energy beams that cause lethal damage to insects through the front plate. The beam head can be a galvanometer system and / or mounted on a Cartesian displacer located on the outside of the sorter at the front to precisely direct the beam toward the pupa or larva to be discarded. This option allows for the elimination of unwanted individuals based on a characteristic other than size, such as morphology or coloration, thus enabling the sexing of individuals belonging to genetically sexed strains that can be developed in the laboratory.

[0028] Stage E.- Pupa distribution arm: at the exit of the pupa calibrator, there is a mechanical arm (16) that directs the individuals (pupas or larvae) collected in a flexible collection tube to the corresponding container (17), according to their classification.

[0029] The main advantage is that the proposed method and device allow for the optimization of mosquito sexing operations, thanks to their self-calibration and self-adjustment capabilities for each batch of pupae processed. The proposed device can vary the size threshold used for sex separation based on the specific characteristics of the batch of pupae being classified, thanks to image analysis and recognition of individuals and their characteristics through the application of real-time artificial intelligence techniques.

[0030] This is a fundamental improvement, as the average pupal size, dispersion, and overlap between males and females vary significantly depending on the larval rearing conditions (feeding regime, feeding schedule, and environmental conditions). Within the same rearing batch, males are on average smaller than females; however, the average sizes of males and females from different batches can overlap due to slight variations in larval rearing conditions. The self-adjusting capability for each batch of pupae analyzed allows the intelligent sexer to identify the threshold value for pupal size that optimizes sex separation, recovering a greater number of male pupae for a given female contamination rate in each of the processed pupal batches. Furthermore, it considerably reduces the operator's labor time required when using a non-self-adjusting classifier.

[0031] The intelligent classifier allows the user to predefine an acceptable rate of female contamination. After image analysis, the intelligent sexer's software / firmware identifies individual pupae, performs a statistical analysis of their properties, such as position and size, and automatically generates histograms and Gaussian curves of pupal size for the male and female populations. The user has previously defined the desired percentage of female contamination along with its statistical probability, knowing that the female contamination rate is inversely correlated with the male recovery rate (a lower female contamination rate will require accepting a lower male recovery rate).The software / firmware will then calculate the cutoff threshold that corresponds to the desired female contamination rate on the Gaussian curve and accurately position the floating plate to perform pupal separation at the desired female contamination value.

[0032] The intelligent classifier features a sealed cavity through which water containing mosquito pupae and larvae flows from inlet holes at the top to outlet holes at the bottom. The floating plate used for classification is housed within this cavity. Unlike other sexing devices, the water is contained within the sealed cavity, increasing cleanliness and ease of use. The absence of external moving parts also contributes to enhanced safety. The floating plate, via integrated actuators, has three-dimensional free movement capabilities.This feature allows it to be arranged in any position and orientation within the space limits of the sealed cavity, enabling multiple combinations between the fixed plane formed by the back face of the front cover of the cavity and the front movable plane of the floating plate; both planes can be parallel at any distance, can form any angle between them while maintaining a parallel axis, or can establish any angle between them without any axis being parallel.

[0033] This feature is particularly relevant to the device's self-calibration and self-adjustment mechanism. In the sexing process, which is based on the size difference between female and male pupae, as well as the removal of larvae, it is essential that the populations of individuals form clouds with a horizontal topology. This allows for the existence of separating areas between the different populations, facilitating their classification through an orderly sliding motion towards the bottom of the calibrator, via the controlled action of the actuators. To achieve this, it is essential that the horizontal axes defining the fixed plane of the front cover and the moving plane of the floating plate are parallel. The described system allows for the automatic, real-time correction of any deviations of the clouds from the horizontal.By detecting and positioning individuals using image analysis and artificial intelligence techniques, the system can extract the characteristic regression line that describes each cloud and apply the necessary corrections to the actuators in real time. This allows for modifying the position and orientation of the moving plane in three-dimensional space, thus correcting any deviation from the horizontal.

[0034] Furthermore, this feature gives the device the ability to count individuals and dose batches with acceptable accuracy and resolution for most processes involved in mass rearing programs, making it a versatile device. Counting and dosing pupal batches are necessary and especially important steps in these types of programs. The described device allows these tasks to be performed innovatively, combining sexing and batch counting and dosing capabilities in a single unit, which represents a significant advantage. The mechanism takes advantage of the three-dimensional control capabilities of the floating plate. Through the controlled action of the actuators, the moving plane can be positioned so that the axes present the necessary angles for the cluster of individuals to form a specific angle with respect to the horizontal.The greater this angle, the smaller the cross-section, thus allowing sufficiently precise dosing by causing, through the actuators, the cloud to slide towards the bottom of the sexer, maintaining the appropriate angle.

[0035] Additionally, the smart mosquito sexer records information on all processed pupae (or larvae, as applicable) in a database, ensuring traceability and enabling statistical analysis to evaluate the work procedures used in larval rearing. The user has access to a wealth of information and analysis to evaluate and improve their larval rearing procedures. This is made possible by analysis using computer vision and artificial intelligence techniques, which allow for the identification of individuals, whether pupae or larvae, their counting with a high degree of accuracy, and the extraction of morphological characteristics for each one, all in real time.

[0036] As an added advantage, it is worth noting that all components of the smart sorter can be implemented so that they require only low-voltage direct current power (typically 24 volts), which eliminates the risk of electrocution, complying with mandatory legislation on essential safety requirements.

[0037] In summary, the proposed innovative method and sexing device are as follows: The method involves the following phases or procedural steps:

[0038] 1 o ). - Creation of batches of pupae from the initial deposit containing all the insects in a volume of water by absorbing predefined volumes of water thanks to a vacuum system generated by a pump.

[0039] 2 o). - Distributed dispensing of pupae and larvae from a batch to the admission zone of the pupa grader thanks to the division of the initial flow of water with pupae into several channels, as many as necessary and at several levels if necessary to distribute the pupae homogeneously along the entire length of the pupa admission zone of the grader, distributing the pupae uniformly without the need for any motorized element for distribution.

[0040] 3°). - Acquisition of digital frontal images of the set of pupae and larvae inside the body of the calibrator, using a high-resolution optical camera located on the outside of the main body of the classifier.

[0041] 4°) - Analysis of the captured images using artificial intelligence algorithms that identify individual pupae and classify them by sex. The image analysis can identify larval stages and other morphological characteristics of potential interest for classification. The software also performs a statistical analysis of the pupal population and determines the optimal position of the internal movable plate for pupal separation, providing a higher recovery rate of male pupae for a user-defined level of female pupal contamination.

[0042] 5°) - Elimination of unwanted individuals by firing a laser beam or any other electromagnetic or sound wave capable of causing lethal damage to insects through the front plate from outside the main body of the classifier. Unwanted individuals are identified by image analysis software based on characteristics other than size, such as morphology or coloration, as is the case with individuals belonging to genetically sexed strains developed in the laboratory. This stage of the process is optional.

[0043] 6 o- Precise three-dimensional positioning of the movable inner plate is achieved thanks to actuators electronically controlled by the classifier's own software / firmware and a magnet-based system that attracts the floating plate to the actuators. In this position, the inner plate releases the small pupae that the software has identified as males. These pupae are carried by the water current to a collector outside the classifier body, which directs them to the tank corresponding to their sex.

[0044] 7 o- Pupa counting and dosing. All individuals passing through the classifier are photographed and counted using AI-based image analysis algorithms. The classifier can process a user-defined number of pupae thanks to the precise three-dimensional positioning of the internal floating plate. In this way, the classifier can also act as a pupa counter and doser, delivering quantities determined by the user.

[0045] 8°). - Storage of information in a database. All information generated after processing a batch of pupae is stored in a database for subsequent analysis and batch traceability.

[0046] The sexing device, for its part, is composed of the following elements or technical means: 1 oThe main body of the pupa grader contains a sealed inner cavity through which water circulates. This cavity houses a floating plate used for grading, which can move freely in three-dimensional space within the cavity's boundaries. The front of the grader's main body is closed by a transparent plate that acts as a lid and seals the cavity with a gasket. Unlike other graders, the only moving part is confined within the cavity, contributing to safer and hazard-free operation. Furthermore, the water flows exclusively within this sealed cavity, improving cleanliness and ease of use.

[0047] The pupa size grading area is formed between the fixed rear surface of the transparent lid and the movable front surface of the floating plate. The position of the floating plate can be adjusted using a precise positioning system, described later. This adjustment allows the angle formed by both surfaces to be modified in three-dimensional space, along one or two axes, facilitating the gradual separation of the pupae arranged between them according to their size.

[0048] The main body of the pupa calibrator has one or more holes at the top for introducing pupae and larvae by means of a water current into the sealed cavity, and one or more holes at the bottom through which the flow is evacuated with individuals of a sufficiently small size to pass through the space configured between both plates, depending on the angles set.

[0049] 2 o - A pupa and larval dispenser in a water stream consists of a tree-shaped piece, which can be at various levels, that divides the flow of water containing pupae and larvae from a reservoir into several channels, as many as necessary, which flow into the inlet holes of the main body of the pupa grader. This component ensures a homogeneous distribution of pupae and larvae without the need for any moving mechanical parts, reducing the complexity of the system.

[0050] 3 o) - Alternatively to the dispenser described in the previous paragraph, a pupa and larval dispenser in a water stream consists of a Galton board-shaped component that concentrates the flow of water containing pupae and larvae from a reservoir in a normal distribution. This component ensures a higher concentration of pupae and larvae in the central part of the sorting surfaces, following a normal distribution, which can be advantageous when applying certain vision, analysis, and artificial intelligence algorithms for the identification, counting, and recognition of features on the individuals.

[0051] 4 oThe precise three-dimensional positioning subsystem for the movable floating plate consists of a set of actuators located at the rear of the main sorter body. These actuators can be stepper motors or other linear or angular actuation devices that allow for precise control via a digital or analog signal from a control system. An axial spacer passes through the sorter housing into the sealed cavity that houses the floating plate, through holes sealed with gaskets or similar mechanisms. The tip of the spacer is in contact with the rear surface of the floating plate and can push it with high precision.

[0052] The floating plate is attracted to the actuators by magnetic attraction. The floating plate contains magnets that tend to push it against the ends of the actuators, attracted by another set of magnets of opposite polarity embedded in the main housing of the classifier (Figure 4). Alternatively, this latter set of magnets could be located at the front, on the transparent cover, with the same polarity as the magnets on the floating plate, thus causing them to repel each other and, consequently, push against the actuators. Other mechanisms for controlling the position of the floating plate can be implemented, such as actuators attached to the plate by ball joints or spring-based systems that allow the floating plate to be pushed or pulled against the ends of the actuators.Thanks to this positioning system, it is possible to control the three-dimensional position of the floating moving plate with an accuracy of less than one micron and the angle formed by the plates with a resolution of thousandths of a degree.

[0053] 5 o - An image acquisition subsystem consisting of a high-resolution optical camera, controlled by a microprocessor and / or microcontroller-based system. The optical camera is oriented towards the transparent front plate and captures images of the pupae and larvae introduced into the classifier.

[0054] 6 oA software program analyzes the image, identifies each pupa / larva using artificial intelligence algorithms, and creates a histogram of pupa sizes within the image. A subsequent algorithm generates two Gaussian curves, one for the size distribution of pupae for each sex. Once the equation for the female pupa size distribution is known, the software applies the desired probability of females, defined by the user in a previous step, and calculates the position of the actuators to place the floating inner plate in the precise position that allows small male pupae to pass through while retaining large female pupae.

[0055] The intelligent mosquito sexer has the capacity for self-calibration and self-adjustment to different pupal populations with varying average sizes and statistical variations, thanks to the individual analysis of pupal size for each batch introduced into the classifier. All stages of image analysis and subsequent algorithms necessary for classification are performed in real time. In addition to identifying individuals, pupae, and larvae, the application of image analysis and artificial intelligence techniques can enable the detection of other characteristics that are highly useful in mosquito larval rearing programs, such as: sexing through the detection of morphological traits, age estimation based on morphological and color characteristics, detection of recombinant individuals, and more.

[0056] 7 oA database stores all the information obtained from image analysis and artificial intelligence for each pupa that has passed through the classifier, with a timestamp for each data point. This raw data is used to perform predefined analyses that provide valuable information to the insectary manager about the larval rearing process, such as pupal recovery rates for both sexes, average pupal size and distribution, pupation timing, and more. This information can help the insectary manager evaluate different larval feeding regimes, schedules, environmental conditions, and other factors that affect insect rearing efficiency.

[0057] 8 o- A device for eliminating individuals (whether pupae or larvae) consisting of a laser beam or any other electromagnetic or sound wave capable of generating a directional energy beam likely to cause lethal damage to insects through the front plate. The laser head may be mounted on a Cartesian displacer located on the outside of the sorter at the front to precisely direct the beam toward the pupa or larva to be eliminated. Alternatively, the beam may be directed to its target by means of a galvanometer. This option allows the elimination of unwanted individuals based on a characteristic other than size, whether morphological or coloration, thus enabling the sexing of individuals belonging to genetically sexed strains that can be developed in the laboratory. 9 o- A dosing mechanism for batches of pupae and larvae in appropriate quantities for the sexing system (main body). The system consists of an open collector that receives the entire batch of pupae and larvae mixed in a sufficient volume of water, and a smaller, closed regulating vessel. A vacuum creates negative pressure in the regulating vessel, causing a batch of pupae and larvae to be drawn from the collector through a conduit that reaches the bottom of the collector. The collector may include an air bubble injection mechanism to ensure a homogeneous distribution of the pupae / larvae.

[0058] A control unit is responsible for issuing the necessary opening and closing signals to the solenoid valve assembly and for controlling the timing and / or signals of the transducers required for the correct batch dosing. Once the batch is dosed into the regulating vessel, its outlet valve and a vent valve open, allowing the batch of pupae and larvae to be supplied to the main body of the classifier via the distribution stage. The regulating vessel may contain an additional water inlet that allows for controlled injection to clean its surfaces and provide additional flow for improved operation of the main classifying body.

[0059] BRIEF DESCRIPTION OF THE DRAWINGS

[0060] For a better understanding of the general characteristics mentioned above, several drawings are attached to the present invention relating to the apparatus, which is described as specified below:

[0061] Figure 1: General operating scheme of the intelligent mosquito pupa sexer or classifier by stages and by technical means employed, which constitute an illustrative list of them

[0062] A. Deposit of pupae and larvae to be sexed

[0063] (1) container cup

[0064] (2) solenoid valve for water flow control

[0065] (3) pump for bubbling system

[0066] (4) solenoid valve

[0067] B. Batch dispenser

[0068] (5) solenoid valve for pupa admission control

[0069] (6) vent solenoid valve

[0070] (7) water flow control solenoid valve

[0071] (8) vacuum control solenoid valve

[0072] (9) vacuum pump

[0073] (10) regulating vessel

[0074] (11) solenoid valve for pupa discharge control

[0075] C. Boogey Dispenser

[0076] (12) Flow divider tree D. Pupa calibrator, with the following subsystems:

[0077] (13) control unit (electronic control subsystem / control unit)

[0078] (14) floating plate (floating plate positioning subsystem)

[0079] (15) Optical camera (digital image acquisition subsystem)

[0080] (16) Sealed receptacle or cavity

[0081] E. Distributor arm

[0082] (16) mechanical arm

[0083] (17) classified insect containers

[0084] Figure 2: Diagram of the pupa dispenser.

[0085] Figure 3: General view and projected implementation of a 4-way pupa and larval dispenser in a water stream.

[0086] Figure 4: Schematic of a Galton-type pupa dispenser that provides a normal distribution of pupae.

[0087] Figure 5: Vertical section view of the intelligent classifier of immature stages of mosquitoes formed by a receiver or inlet (14.1) of mosquito pupae or larvae by means of a water current, an outlet (14.2) of mosquito pupae or larvae by means of a water current, a fixed front transparent plate (14.3), a movable plate (14.4), whose position is electronically fixed by a set of actuators controlled by an automatic software, a main body of the sizer (14.5), a high resolution optical camera (15), a female pupa (14.6), a male pupa (14.7), magnets (14.8), and actuators or stepper motors (14.9) or any other linear actuation system that allows precise control of the position from a digital or analog signal.

Claims

CLAIMS - An intelligent method for classifying immature mosquito stages by sex based on enhanced size sexual dimorphism using image analysis with algorithms developed by artificial intelligence, characterized by the following stages: 1 o ). - Creation of batches of pupae from the initial deposit containing all the insects to be classified mixed in water by the absorption of predefined volumes of water thanks to a vacuum system generated by a pump. 2 o). - Distributed dispensing of pupae and larvae from a batch to the admission zone of the pupa grader thanks to the division of the initial flow of water with pupae into several channels, as many as necessary and in several stages if necessary to distribute the pupae homogeneously along the entire length of the pupa admission zone of the grader, distributing the pupae uniformly without the need for any motorized element for distribution. 3°). - Acquisition of frontal images of the set of pupae inside the calibrator body, using a high-resolution optical camera located on the outside of the main body of the classifier. 4°) - Analysis of the captured images using artificial intelligence algorithms that identify individual pupae and classify them by sex. The image analysis can identify larval stages and other morphological characteristics of potential interest for classification. The software also performs a statistical analysis of the pupal population and determines the optimal position of the internal movable plate for pupal separation, resulting in a higher pupal recovery rate. males for a user-defined level of female pupal contamination. 5 o- Elimination of unwanted individuals by firing a laser beam or any other electromagnetic or sound wave capable of causing lethal damage to insects through the front plate from outside the main body of the classifier. Unwanted individuals are identified by image analysis software based on characteristics other than size, such as morphology or coloration, as is the case with individuals belonging to genetically sexed strains developed in the laboratory. This stage of the process is optional. 6 o- Precise three-dimensional positioning of the movable inner plate is achieved thanks to actuators electronically controlled by the classifier's own software / firmware and a magnet-based system that attracts the floating plate to the actuators. In this position, the inner plate releases the small pupae that the software has identified as males. These pupae are carried by the water current to a collector outside the classifier body, which directs them to the tank corresponding to their sex. 7°) - Pupa Counting and Dosing. All individuals passing through the classifier are photographed and counted using image analysis algorithms based on artificial intelligence. The classifier can process a user-defined number of pupae thanks to the precise positioning of the internal floating plate in three dimensions. In this way, the classifier can also act as a pupa counter and dispenser, delivering quantities determined by the user. 8°). - Storage of information in a database. All information generated after processing a batch of pupae is stored in a database for subsequent analysis and batch traceability.

2. - An intelligent apparatus for classifying immature mosquito stages by sex based on sexual dimorphism enhanced by the use of image analysis algorithms and artificial intelligence, characterized by a main body of the pupa grader equipped with a sealed inner cavity (14.10) through which water circulates and which houses a movable plate (14.4) used for pupa grading, the position of which can be freely varied in three-dimensional space within the limits of the cavity. The front of the main body of the grader is closed by a transparent plate (14.3) that acts as a lid and seals the cavity by means of a gasket. The grading space for pupa size grading is formed within this cavity between the fixed rear surface of the transparent lid and the movable front surface of the floating plate. The position of the floating plate (14.4) is adjustable by a mechanism described in claim 3. This adjustment allows modification in three-dimensional space of the angle formed by both surfaces, on one or two axes, facilitating the gradual separation of the pupae arranged between them according to their size. Said main body of the pupa calibrator has one or more holes (14.1) in the upper part for the introduction of the pupae and larvae by means of a water current inside the sealed cavity, and one or more holes (14.2) in the lower part through which the water flow is evacuated with the individuals of a sufficiently small size to pass through the space configured between both plates, depending on the established angles.

3. - An intelligent device for classifying immature mosquito stages by sex based on sexual dimorphism enhanced through the use of image analysis algorithms and artificial intelligence, characterized by an electronic three-dimensional positioning subsystem for the internal movable plate of the calibrator. This precise three-dimensional positioning subsystem of the movable plate (14.4) is composed of a set of actuators (14.9) located at the rear of the main body of the calibrator. Actuators can be stepper motors or other linear or angular actuation devices that allow precise control by means of a digital or analog signal from a control system. An axial spacer passes through the classifier housing into the sealed cavity that houses the moving plate, passing through holes sealed with gaskets or similar mechanisms. The tip of the spacer is in contact with the back surface of the moving plate and can push it with high precision. The moving plate is attracted to the actuators by magnetic attraction. This moving plate contains magnets (14.8) that tend to push it against the ends of the actuators, which are attracted by another set of magnets (14.8) of opposite polarity embedded in the housing of the main body of the classifier.

4. An intelligent device for classifying immature mosquito stages by sex based on enhanced sexual dimorphism using image analysis algorithms and artificial intelligence. This device is characterized by an AI image analysis subsystem, integrated into the calibrator, which enables the intelligent determination of the three-dimensional position of the internal moving plate. This subsystem facilitates pupal separation according to user-defined parameters based on male recovery rate and female contamination rate. The device comprises an optical camera, a microprocessor and / or microcontroller for camera control, AI algorithms, and statistical data processing algorithms. The optical camera is oriented towards the transparent front plate and captures images of the pupae and larvae introduced into the calibrator.Software analyzes the image, identifies each pupa / larva using artificial intelligence algorithms, and creates a histogram of pupa sizes within the image. A subsequent algorithm generates two Gaussian curves, one for the size distribution of pupae for each sex. Once the equation for the female pupa size distribution is known, the software applies the desired probability of females, defined by the user in a previous step, and calculates the position of the actuators. The subsystem described in claim 3 is for positioning the floating inner plate in the exact position that allows the passage of small male pupae while retaining large female pupae. A database stores all the information obtained from image analysis and artificial intelligence for each pupa that has passed through the classifier, recording the timestamp for each data entry. The subsystem described in claim 3 is responsible for moving the movable plate to the calculated position. Image analysis and calculation of the moving plate's position are performed in real time for each batch of pupae introduced into the grader, customizing the separation threshold for the specific characteristics of the pupae being analyzed. This results in improved classification parameters, such as the recovery of male pupae and the reduction of female pupae contamination.

5. - An intelligent classification device for immature mosquito stages by sex based on sexual dimorphism enhanced through the use of image analysis algorithms and artificial intelligence, characterized by a dispenser (12) for pupae and larvae in a water stream, consisting of a tree-shaped component that divides the flow of water containing pupae and larvae from a reservoir into several channels, as many as necessary, which flow into the inlet holes of the main body of the pupa sorter. This dispenser ensures a homogeneous distribution of pupae and larvae without the need for any moving mechanical elements, thus reducing the complexity of the system.

6. - An intelligent device for classifying immature mosquito stages by sex based on sexual dimorphism enhanced through the use of image analysis algorithms and artificial intelligence, characterized by a mechanism for dosing batches of pupae and larvae in quantities suitable for the correct operation of the pupa grader. This system comprises an open collector (1) that receives the entire set of pupae and larvae mixed in a sufficient volume of water, and a smaller, closed regulating vessel (10). A vacuum generated by a pump (9) creates negative pressure in the regulating vessel, causing a batch of pupae and larvae to be drawn from the collector through a conduit that reaches the bottom of the vessel. Finally, a control unit (13) sends the necessary opening and closing signals to the set of solenoid valves (5, 6, 7, 8, 11) and controls the timing and / or signals of the transducers required for the correct batch dosing. Once the batch is dosed into the regulating vessel (10), its outlet valve and a vent valve (6) open, allowing the batch of pupae and larvae to be delivered to the main body of the calibrator via the dispensing stage.The regulating vessel (10) may contain an additional water inlet that allows for controlled injection in order to clean its surfaces and add additional flow for improved operation of the calibrator.

7. An intelligent sex-classifying device for immature mosquito stages based on enhanced sexual dimorphism using image analysis algorithms and artificial intelligence, according to claim 2, and / or 3, and / or 4, and / or 5, and / or 6, characterized by a device for eliminating individuals—pupae or larvae—consisting of a laser beam or any other electromagnetic or sound wave capable of generating a directional energy beam that can cause lethal damage to insects through the front plate. The laser head can be mounted on a Cartesian displacer located on the outside of the classifier at the front to precisely direct the beam toward the pupa or larva to be eliminated. Alternatively, the beam can be directed to its target using a galvanometer.This option allows the elimination of unwanted individuals based on a characteristic other than the size of the individual, whether morphological or coloration, allowing the sexing of individuals belonging to genetically sexed strains that can be developed in the laboratory. 8 - An intelligent classification apparatus for immature stages of mosquitoes by sex based on sexual dimorphism enhanced by the use of image analysis algorithms and artificial intelligence according to claim 5 characterized in that, alternatively / complementarily to the dispenser, a dispenser of pupae and larvae in a water stream is used, consisting of a Galton Board-shaped piece, which distributes the pupae and larvae of a batch following a normal distribution of pupae in the classification area inside the body of the calibrator.

9. An intelligent classification apparatus for immature stages of mosquitoes by sex based on sexual dimorphism enhanced by the use of image analysis algorithms and artificial intelligence according to claim 3 characterized in that, alternatively / complementarily to the precise three-dimensional positioning subsystem of the movable plate (14.4), the magnets of the movable plate inside the calibrator are repelled by a second set of magnets of the same polarity embedded in the transparent front plate (14.3) of the main body of the calibrator, thus causing their repulsion and, therefore, their push against the actuators (14.9). 10 - An intelligent classification apparatus for immature stages of mosquitoes by sex based on sexual dimorphism enhanced by the use of image analysis algorithms and artificial intelligence according to claim 3 characterized in that other mechanisms for controlling the position of the floating plate can be implemented alternatively, such as actuators attached to the plate by means of ball joints or springs and coils that allow a push or pull of the floating plate on the ends of the actuators.

Images