Babesia detection ai training method and detection method
Patent Information
- Application Number
- CN202510347267.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-24
- Publication Date
- 2026-09-25
AI Technical Summary
[0007]实际应用中，巴贝斯虫感染率并不太高，为每一份样本都进行高倍数的显微放大，再进行识别处理，无论是人工还是机器识别都会增加工作量
[0021]上述技术方案的技术效果包括：采用感染巴贝斯虫感染的样本，形成有效的样本，训练效率更高。用感染的样本训练巴贝斯虫特征数据集，能提高对巴贝斯虫感染的识别准确性。
Smart Images
Figure CN122821544A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of formed element analysis technology based on magnified microscopic images, and particularly relates to a method for detecting the probability of Babesia infection, especially a method for detecting Babesia. Background Technology
[0002] In the current technology, a microscope with sufficient magnification is usually required to confirm a Babesia infection.
[0003] Babesia, a blood-borne Babesia, primarily infects and transmits through tick bites. After a tick bites, it releases Babesia sporophytes into the host's bloodstream. The sporophytes penetrate red blood cells, divide, and form merozoites and merozoites. The rupture of the red blood cells releases the parasite, which then continues to invade and destroy new red blood cells, repeating the developmental process. It can also be transmitted through blood and vertically through the placenta. Morphologically, it can be divided into two types: Babesia microcarpa and Babesia macrocarpa.
[0004] Babesia microsporum: Primarily Babesia gibrini, 1-3 μm in size, typically characterized by a single ring-like structure, but punctate or rod-like forms can also be observed. Multiple parasites can infect a single red blood cell. Figure 2 The image shows the infection of Babesia minor within red blood cells. Figure 2 This is a magnified micrograph under an oil immersion microscope. The small dots in the red blood cells are Babesia microsporum.
[0005] Babesia macrocarpa: Primarily Babesia canis, Babesia voyeuris, and Babesia revoluta, measuring 3-7 μm in size. A typical characteristic is a pear-seed-like or melon-seed-shaped form, pointed at one end and blunt at the other. Other shapes, such as ring-shaped or amoeboid, can also be observed. Multiple worms can infect a single red blood cell. For example... Figure 3 The image shows the infection of *Babesia macrocephala* within red blood cells. Figure 3 This is a magnified micrograph under an oil immersion microscope. The pear-shaped spots in the red blood cells and the pear-shaped spots outside the cells are Babesia macrocarpa.
[0006] In formed element detection devices, the higher the magnification of the microscope, the higher the cost. Diagnosing Babesia typically requires a microscope with magnification of 60x or higher.
[0007] In practical applications, the infection rate of Babesia is not very high. Performing high-magnification microscopy on each sample before identification, whether manually or by machine, would increase the workload. How to combine it with a low-magnification microscope to improve overall detection efficiency is a technical problem that needs to be solved. Summary of the Invention
[0008] In this application, the inventors proposed a technical solution that first uses a low-power microscope to detect Babesia. Only when the Babesia test results indicate a high probability of suspected infection is a high-power microscope used for further testing, thereby improving the efficiency and overall benefits of the test.
[0009] A method for training an AI to detect Babesia includes: collecting infected blood samples; the infected blood samples being infected with Babesia; preprocessing the blood samples to obtain microscopic samples; setting the microscope magnification and taking pictures of the microscopic samples to obtain images; identifying Babesia in the images; labeling the images as Babesia to obtain labeled images; using the labeled images for AI training to obtain a Babesia feature dataset; and combining the Babesia feature dataset with the corresponding AI algorithm to achieve the ability to identify Babesia.
[0010] The Babesia mentioned above could be Babesia macrocarpa, and the magnification factor could be greater than 60 times.
[0011] The Babesia mentioned above could be a small Babesia, and the magnification mentioned above is greater than 70 times.
[0012] A method for detecting Babesia includes step A: preprocessing blood samples to obtain microscopic samples; step B: performing suspected Babesia detection to identify blood samples containing suspected Babesia; step C: re-obtaining a high-magnification image B from the blood samples containing suspected Babesia; and step D: analyzing the above image B using an AI recognition algorithm in conjunction with a Babesia feature dataset to identify Babesia in the image; wherein the Babesia feature dataset is obtained by training on Babesia images.
[0013] This could be because step B above includes obtaining image A at a magnification of A; using an AI recognition algorithm in conjunction with a Babesia feature dataset to analyze the microscopic sample image A and identify the Babesia in image A; the Babesia feature dataset is obtained by training on Babesia images; the Babesia is a large Babesia, and the magnification is greater than 30 times and less than 50 times; or the Babesia is a small Babesia, and the magnification is greater than 30 times and less than 50 times.
[0014] It may also include obtaining the number of Babesia in the microscopic sample image, obtaining the volume corresponding to the microscopic sample image, and calculating the content of Babesia in a unit volume; if the content of Babesia in the unit volume is greater than a set threshold, the blood sample is considered to contain Babesia.
[0015] It may also include obtaining the number of Babesia in the microscopic sample image, obtaining the number of red blood cells corresponding to the microscopic sample image, and calculating the ratio of the number of Babesia to the number of red blood cells; if the ratio is greater than a set threshold, the blood sample is considered to contain Babesia.
[0016] Yes, the Babesia mentioned above is Babesia macrocarpa, and the magnification B is greater than 60 times.
[0017] Yes, the Babesia mentioned above is Babesia microbabesia, and the magnification B mentioned above is greater than 70 times.
[0018] A method for detecting Babesia includes preprocessing a blood sample to obtain a microscopic sample; setting a microscope magnification and taking a picture of the microscopic sample to obtain a microscopic sample image; using an AI recognition algorithm in conjunction with a Babesia feature dataset to analyze the microscopic sample image and identify Babesia in the image; wherein the Babesia feature dataset is obtained by training on Babesia images; wherein the Babesia is *Babesia macrocarpa* and the magnification is greater than 60 times; or wherein the Babesia is *Babesia septembrittlement* and the magnification is greater than 70 times.
[0019] A computing processing device or detection device includes all or part of a method for running the above-described method; the memory of the computing processing device includes the above-described Babesia feature dataset.
[0020] A data storage device includes storing all or part of the program code used in the above-described method; and storing the Babesia worm feature dataset described above.
[0021] The technical advantages of the above solution include: using samples infected with Babesia to form effective samples, resulting in higher training efficiency; and using infected samples to train the Babesia feature dataset to improve the accuracy of Babesia infection identification.
[0022] The technical effects of the above-mentioned solution include: the ability to classify and identify multiple Babesia species, as well as to identify multiple species of Babesia simultaneously. These include Babesia, Babesia macrocarpa, and Babesia microcarpa.
[0023] The technical effects of the above-mentioned technical solution include: using artificial intelligence to train the ability to identify suspected infections in low-magnification images, and using artificial intelligence to train the ability to identify confirmed infections in high-magnification images. Combining the two can reduce the difficulty of detection and speed up the detection time.
[0024] The technical effects of the above-mentioned technical solution include: the Babesia detection method can detect small-sized Babesia using a low-magnification microscope, which is a cost-effective detection method.
[0025] The technical effects of the above-mentioned technical solution include: first, using a low-power microscope to detect Babesia, and only when the detection results of Babesia indicate a high probability of suspected infection, switching to a high-power magnifying microscope for further detection, thereby improving the efficiency and overall benefits of detection.
[0026] The technical advantages of the above solution include: accurately quantifying the Babesia content per unit volume, making the results more reliable; and providing more credible output results, making it easier to distinguish the severity of infection. The Babesia content per unit volume is used to indicate the degree of Babesia infection with greater accuracy.
[0027] The technical effects of the above-mentioned technical solution include: the Babesia detection method can output Babesia infection assessment values based on the relationship between Babesia content and threshold, and can obtain more accurate diagnostic information with the help of a microscope.
[0028] The technical effects of the above-mentioned technical solution include: the ratio of Babesia worm count to red blood cell count is used to indicate the degree of Babesia worm infection, and with red blood cells as a benchmark, the accuracy is higher. Attached Figure Description
[0029] Figure 1 This is a schematic diagram of the Babesia detection method;
[0030] Figure 2 This is a diagram illustrating the infection of Babesia minimus by red blood cells;
[0031] Figure 3 This is a diagram illustrating the infection of the giant Babes with red blood cells;
[0032] Figure 4 This is a schematic diagram of the AI training method for detecting Babesia.
[0033] Figure 5 This is a schematic diagram of the Babesia detection method. Detailed Implementation
[0034] The contents of this application will be further described in detail below with reference to the accompanying drawings. It should be noted that the following description is of preferred embodiments of the present invention and does not constitute any limitation on the present invention. The description of the preferred embodiments of the present invention is merely an explanation of the general principles of the invention. The designations "first," "second," "A," and "B" used in this invention are for ease of explanation only and do not represent a temporal or spatial order. The combinations of letters and numbers "TA," "TB," and "H" used in this invention are for ease of explanation only, and their specific meanings are determined by the specific terms they represent.
[0035] In this application, the Babesia feature dataset is feature data obtained after training an AI computing model. This feature data, in conjunction with the AI computing model, can perform AI calculations.
[0036] Alternatively, the Babesia feature dataset can be used as an AI computing model obtained after training an AI computing model, which can perform AI computing.
[0037] like Figure 4A method for training an AI to detect Babesia includes: collecting infected blood samples; the infected blood samples being infected with Babesia; preprocessing the blood samples to obtain microscopic samples; setting the microscope magnification and taking pictures of the microscopic samples to obtain images; identifying Babesia in the images; labeling the images as Babesia to obtain labeled images; using the labeled images for AI training to obtain a Babesia feature dataset; and combining the Babesia feature dataset with the corresponding AI algorithm to achieve the ability to identify Babesia.
[0038] In some embodiments, the Babesia described above is Babesia macrocarpa, and the magnification is greater than 60x. In other embodiments, the Babesia described above is Babesia microcarpa, and the magnification is greater than 70x.
[0039] like Figure 5 In one embodiment of a Babesia detection method, the method includes step A: preprocessing a blood sample to obtain a microscopic sample; step B: performing suspected Babesia detection to identify blood samples containing suspected Babesia; step C: re-obtaining a high-magnification image B from the blood sample containing suspected Babesia; and step D: analyzing the image B using an AI recognition algorithm in conjunction with a Babesia feature dataset to identify the Babesia in the image; the Babesia feature dataset is obtained by training on Babesia images.
[0040] Step B above includes obtaining image A at a magnification of A; analyzing the microscopic sample image A using an AI recognition algorithm in conjunction with a Babesia feature dataset to identify Babesia in image A; the Babesia feature dataset is obtained by training on Babesia images; the Babesia is either a large Babesia or a small Babesia, with a magnification greater than 30x and less than 50x; or the Babesia is either a small Babesia or a small Babesia, with a magnification greater than 30x and less than 50x.
[0041] In some embodiments, the method further includes obtaining the number of Babesia in the microscopic sample image, obtaining the volume corresponding to the microscopic sample image, and calculating the content of Babesia per unit volume; if the content of Babesia per unit volume is greater than a set threshold, the blood sample is considered to contain Babesia.
[0042] In some embodiments, the method further includes obtaining the number of Babesia parasites in the microscopic sample image, obtaining the number of red blood cells corresponding to the microscopic sample image, and calculating the ratio of the number of Babesia parasites to the number of red blood cells; if the ratio is greater than a set threshold, the blood sample is considered to contain Babesia parasites.
[0043] In some embodiments, in step C above, the Babesia is a large Babesia, and the magnification B is greater than 60 times; or the Babesia is a small Babesia, and the magnification B is greater than 70 times.
[0044] like Figure 1 In one embodiment of a method for detecting Babesia, a blood sample is preprocessed to obtain a microscopic sample; a microscope magnification is set, and the microscopic sample is photographed to obtain a microscopic sample image; an AI recognition algorithm is used in conjunction with a Babesia feature dataset to analyze the microscopic sample image and identify Babesia in the image; the Babesia feature dataset is obtained by training with Babesia images; the Babesia is *Babesia macrocarpa*, and the magnification is greater than 60 times; or the Babesia is *Babesia septembrittlement*, and the magnification is greater than 70 times.
[0045] A computing processing device or detection device includes all or part of the method described above; the memory of the computing processing device includes the Babesia worm feature dataset described above.
[0046] A data storage device stores all or part of the program code for performing the above-described methods and stores the above-described Babesia feature dataset.
[0047] While the present invention has been described and illustrated with reference to preferred embodiments and several alternatives, the invention is not limited to the specific descriptions herein. Other alternatives or equivalent components may also be used to practice the invention.
Claims
1. A method for training an AI for detecting Babesia, characterized in that, include: Collect infected blood samples; the infected blood samples were infected with Babesia; Blood samples are pre-processed to obtain microscopic samples; Set the microscope magnification, take pictures of the microscopically examined samples, and obtain images; The image of Babesia in the picture is labeled as Babesia, and the labeled image is obtained. AI was trained using labeled images to obtain a dataset of Babesia features. The Babesia worm feature dataset, combined with the corresponding AI algorithm, has the ability to identify Babesia worms.
2. The method according to claim 2, characterized in that, include: The The Babesia is Babesia macrocarpa, and the magnification is greater than 60 times. The Babesia is a small Babesia, and the magnification is greater than 70 times.
3. A method for detecting Babesia, characterized in that, include Step A: Preprocess the blood sample to obtain a microscopic sample; Step B: Perform suspected Babesia testing to identify blood samples suspected of containing Babesia; Step C: For the blood sample suspected of containing Babesia, re-obtain image B at high magnification B; Step D: Using AI recognition algorithms in conjunction with the Babesia feature dataset, analyze image B to identify the Babesia in the image; The Babesia feature dataset was obtained by training on Babesia images.
4. The method according to claim 3, characterized in that, include: Step B includes obtaining image A at a magnification of A; Using AI recognition algorithms in conjunction with a Babesia feature dataset, the microscopic sample image A was analyzed to identify Babesia in image A. The Babesia feature dataset was obtained by training on Babesia images; The Babesia is Babesia macrocarpa, and the magnification is greater than 30x and less than 50x. Or the Babesia is a small Babesia, and the magnification is greater than 30 times and less than 50 times.
5. The method according to claim 4, characterized in that, It also includes obtaining the number of Babesia in the microscopic sample image, obtaining the volume corresponding to the microscopic sample image, and calculating the content of Babesia per unit volume; wherein the content of Babesia per unit volume is greater than a set threshold, and the blood sample contains Babesia.
6. The method according to claim 4, characterized in that, It also includes obtaining the number of Babesia parasites in the microscopic sample image, obtaining the number of red blood cells corresponding to the microscopic sample image, and calculating the ratio of the number of Babesia parasites to the number of red blood cells; if the ratio is greater than a set threshold, the blood sample contains Babesia parasites.
7. The method according to claim 4, characterized in that, include: In step C The Babesia is Babesia macrocarpa, and the magnification B is greater than 60 times. Or the Babesia is a small Babesia, and the magnification B is greater than 70 times.
8. A method for detecting Babesia, characterized in that, include Blood samples are pre-processed to obtain microscopic samples; Set the microscope magnification, take a picture of the microscopic sample, and obtain an image of the microscopic sample; Using AI recognition algorithms in conjunction with a Babesia feature dataset, the microscopic sample images were analyzed to identify Babesia in the images; The Babesia feature dataset was obtained by training on Babesia images; The Babesia is Babesia macrocarpa, and the magnification is greater than 60 times. Or the Babesia is a small Babesia, and the magnification is greater than 70 times.
9. A computing processing device or detection device, characterized in that, Includes any one of the following technical features: TEA1: Used to perform all or part of the method described in any one of claims 1 to 8; TEA2: The memory of the computing processing device includes the Babesia feature dataset as described in any one of claims 1 to 8.
10. A data storage device, characterized in that, Includes any one of the following technical features: TEB1: Stores program code for performing all or part of the method described in any one of claims 1 to 8; TEB2: Stores a dataset of Babesia features as described in any one of claims 1 to 8.