Pathogen pattern scanning system

Abstract

The invention relates to the technical field of pathogen scanning, and particularly discloses a pathogen pattern scanning system, which comprises a pattern scanner module, a data acquisition module, a data processing module and a report generation module, the system can detect functional and metabolic abnormalities through medicament development and multi-band spectrum scanning instead of depending on morphological changes, and the mechanism can discover early lesions, so that a patient can obtain a diagnosis opportunity in a potential hazard stage; according to the pathogen pattern scanning system, a patient only needs to take a developing medicament, comprehensive detection can be completed through spectrum scanning and 3D modeling, invasive examination and radiation exposure are avoided, the system matches diagnosed cases in a database through a deep learning algorithm, similarity matching and risk scoring are carried out, and clear and authoritative medical suggestions are provided for doctors.

Description

Technical Field

[0001] The present invention belongs to the technical field of pathogen scanning, and in particular relates to a pathogen pattern scanning system. Background Art

[0002] In the field of medical diagnosis, the detection of pathogens and internal lesions has always been a crucial link. Traditional detection methods, such as X-rays, CT scans, and MRIs, can provide images of the internal structure of the human body, but they are limited in identifying specific pathogens or subtle lesions. These technologies often rely on the doctor's experience and naked eye observation, and there may be a risk of missed detection or misdiagnosis for some hidden or early lesions.

[0003] In addition, most existing medical imaging technologies use single or limited-band spectral imaging, which limits the richness and depth of image information. In complex cases, single-band images may not fully reveal the detailed features of the lesion area, thus affecting the accuracy of diagnosis. At the same time, traditional medical image processing and analysis methods often lack automation and intelligence, requiring doctors to spend a lot of time and energy on manual interpretation and comparison, which not only increases the workload, but may also lead to inconsistencies in diagnostic results due to human factors.

[0004] In response to these problems, the existing technology urgently needs a method that can detect pathogens and lesions more accurately and efficiently. However, the existing medical imaging technology and analysis methods still have obvious deficiencies in terms of resolution, information volume, degree of automation, and diagnostic accuracy, and cannot meet the urgent needs for high-precision and high-efficiency diagnosis in clinical practice. Therefore, it is particularly important to develop a new pathogen pattern scanning system that can overcome these limitations.

[0005] In response to this, the inventors proposed a pathogen pattern scanning system to solve the above problems. Summary of the invention

[0006] The object of the present invention is to provide a pathogen pattern scanning system to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] Pathogen pattern scanning system, including:

[0009] Pattern Scanner Module: Uses multi-band spectral imaging technology combined with 3D modeling technology to scan the medicine swallowed inside the human body and generate high-resolution organ images;

[0010] Data acquisition module: automatically collects spectral reflectance data after swallowing the drug, uses image fusion technology to combine multi-band images, segment the lesion area, and use the lesion area to generate a 3D image with color identification;

[0011] Data processing module: Apply machine learning algorithms to extract features of 3D images, automatically identify, classify and match database similarity of 3D images of lesion areas, locate pathogen attachment sites and evaluate the degree of lesions;

[0012] Report generation module: Automatically generate test reports, including imaging area analysis, historical comparison data and medical advice.

[0013] Preferably, the expression for obtaining spectral data using multi-band spectral imaging technology is:

[0014]

[0015] Where R(λ): reflectivity of the organ or tissue to wavelength λ;

[0016] Lr(λ): the intensity of reflected light recorded by the detector;

[0017] Li(λ): incident light intensity;

[0018] By calculating the reflectivity at different wavelengths, the spectral characteristics of organ tissues are obtained, and the diseased area is distinguished from healthy tissues.

[0019] Preferably, the generation formula of the organ 3D image is:

[0020] I(x,y,z)=∫ λ w(λ)R(x,y,z,λ)dλ

[0021] Where I(x,y,z): the intensity value of the voxel in the 3D reconstruction;

[0022] R(x,y,z,λ): reflectivity at position (x,y,z) at a specific wavelength;

[0023] w(λ): weight of each band, used to adjust the importance of a specific wavelength;

[0024] By fusing multi-band data, accurate 3D images are generated, highlighting the color differences in the developed areas.

[0025] Preferably, the segmentation of the lesion area is obtained by using a threshold segmentation and edge detection method, and the expression of the threshold segmentation is:

[0026]

[0027] Where ω1, ω2: the probability of pixels on both sides of the threshold;

[0028] μ1, μ2: the mean of the pixels on both sides of the threshold;

[0029] μT: global pixel mean;

[0030] By maximizing the between-class variance Automatically calculate the optimal segmentation threshold;

[0031] The expression of the edge detection method is:

[0032]

[0033] Where Gx, Gy: horizontal and vertical components of image gradient;

[0034] G: Image gradient amplitude, indicating edge strength.

[0035] Preferably, the features of the 3D image include color features and texture features, and a high-dimensional feature vector is generated by extracting the color and texture features for subsequent lesion classification.

[0036] Preferably, the color feature is obtained using the RGB difference formula, which is expressed as:

[0037] ΔR=R 病变 -R 正常 , ΔG=G 病变 -G 正常 , ΔB=B 病变 -B 正常

[0038] Where ΔR, ΔG, ΔB: the color difference between the lesion and the normal area.

[0039] Preferably, the expression of texture color feature is:

[0040]

[0041] Where P(i,j,d,θ): the co-occurrence probability of pixel pairs with grayscale values ​​i and j at distance d and direction θ.

[0042] Preferably, the severity of the lesion is calculated based on the area and color intensity change, and the expression is:

[0043]

[0044] Where Ai: the area of ​​the i-th lesion area;

[0045] I i: color intensity of the i-th lesion area;

[0046] Quantitatively evaluate the distribution and extent of the lesion area and generate the lesion risk level.

[0047] Preferably, the database similarity matching calculation expression is:

[0048]

[0049] in is the feature vector of the current lesion area and the sample in the database;

[0050] is the magnitude of the vector.

[0051] Preferably, the composition of the medicament includes the following conditions:

[0052] Biocompatibility: non-toxic to human body, no side effects, and completely excreted;

[0053] Development ability: It can change the color reflection characteristics of organs and lesions through biochemical reactions, forming spectral characteristics that are significantly different from healthy tissues;

[0054] Targeting: Selectively attach to or visualize pathogens and damaged tissue areas.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] (1) The present invention uses drug development and multi-band spectral scanning to enable the system to detect functional and metabolic abnormalities, rather than relying solely on morphological changes. This mechanism can detect early lesions and provide patients with a diagnosis opportunity at a potential risk stage.

[0057] (2) The pathogen pattern scanning system of the present invention only requires the patient to take a developing agent, and comprehensive detection can be completed through spectral scanning and 3D modeling, avoiding invasive examinations and radiation exposure. The system matches confirmed cases in the database through a deep learning algorithm, performs similarity matching and risk scoring, and provides doctors with clear and authoritative medical advice. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a block diagram of the pathogen pattern scanning system of the present invention.

[0059] Figure 2 This is a multi-band spectral imaging scan of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] Embodiment 1:

[0062] See also Figure 1 and Figure 2 As shown, the pathogen pattern scanning system comprises:

[0063] Pattern Scanner Module: Uses multi-band spectral imaging technology combined with 3D modeling technology to scan the medicine swallowed inside the human body and generate high-resolution organ images;

[0064] Data acquisition module: automatically collects spectral reflectance data after swallowing the drug, combines multi-band images using image fusion technology, segments the lesion area, and uses the lesion area to generate a 3D image with color identification;

[0065] Data processing module: applying machine learning algorithms to automatically identify, classify and match the 3D images of the lesion area, extracting features of the 3D images, locating the attachment sites of pathogens and evaluating the extent of lesions;

[0066] Report generation module: Automatically generate test reports, including imaging area analysis, historical comparison data and medical advice.

[0067] As can be seen above, through drug development and multi-band spectral scanning, the system can detect functional and metabolic abnormalities, rather than relying solely on morphological changes. This mechanism can detect early lesions and give patients a chance to get a diagnosis at the potential risk stage.

[0068] The pathogen pattern scanning system only requires the patient to take a developing agent, and comprehensive testing can be completed through spectral scanning and 3D modeling, avoiding invasive examinations and radiation exposure. The system uses a deep learning algorithm to match confirmed cases in the database, perform similarity matching and risk scoring, and provide doctors with clear and authoritative medical advice.

[0069] Embodiment 2:

[0070] refer to Figure 1 and Figure 2 As shown, the expression for obtaining spectral data by the multi-band spectral imaging technology is:

[0071]

[0072] Where R(λ): reflectivity of the organ or tissue to wavelength λ;

[0073] Lr(λ): the intensity of reflected light recorded by the detector;

[0074] Li(λ): incident light intensity;

[0075] By calculating the reflectivity at different wavelengths, the spectral characteristics of organ tissues are obtained, and the diseased area is distinguished from healthy tissues.

[0076] Specifically, the generation formula of the organ 3D image is:

[0077] I(x,y,z)=∫ λ w(λ)R(x,y,z,λ)dλ

[0078] Where I(x,y,z): the intensity value of the voxel in the 3D reconstruction;

[0079] R(x,y,z,λ): reflectivity at position (x,y,z) at a specific wavelength;

[0080] w(λ): weight of each band, used to adjust the importance of a specific wavelength;

[0081] By fusing multi-band data, accurate 3D images are generated, highlighting the color differences in the developed areas.

[0082] Specifically, the segmented lesion area is obtained by using a threshold segmentation and edge detection method, and the expression of the threshold segmentation is:

[0083]

[0084] Where ω1, ω2: the probability of pixels on both sides of the threshold;

[0085] μ1, μ2: the mean of the pixels on both sides of the threshold;

[0086] μT: global pixel mean;

[0087] By maximizing the between-class variance Automatically calculate the optimal segmentation threshold;

[0088] The expression of the edge detection method is:

[0089]

[0090] Where Gx, Gy: horizontal and vertical components of image gradient;

[0091] G: Image gradient amplitude, indicating edge strength;

[0092] Accurately segment the contour of the lesion area, providing a basis for subsequent 3D image reconstruction and lesion analysis.

[0093] Specifically, the features of the 3D image include color features and texture features. By extracting the color and texture features, a high-dimensional feature vector is generated for subsequent lesion classification.

[0094] Specifically, the color feature is obtained using the RGB difference formula, which is expressed as follows:

[0095] ΔR=R病变 -R 正常 , ΔG=G 病变 -G 正常 , ΔB=B 病变 -B 正常

[0096] Where ΔR, ΔG, ΔB: the color difference between the lesion and the normal area.

[0097] Specifically, the expression of the texture color feature is:

[0098]

[0099] Where P(i,j,d,θ): the co-occurrence probability of pixel pairs with grayscale values ​​i and j at distance d and direction θ.

[0100] Specifically, the evaluation of the degree of lesions is performed by calculating the severity based on the area and color intensity changes, and the expression is:

[0101]

[0102] Where Ai: the area of ​​the i-th lesion area;

[0103] I i: color intensity of the i-th lesion area;

[0104] Quantitatively evaluate the distribution and extent of the lesion area and generate the lesion risk level.

[0105] Specifically, the database similarity matching calculation expression is:

[0106]

[0107] in is the feature vector of the current lesion area and the sample in the database;

[0108] is the magnitude of the vector.

[0109] Specifically, the ingredients of the medicament include the following conditions:

[0110] Biocompatibility: non-toxic to human body, no side effects, and completely excreted;

[0111] Development ability: It can change the color reflection characteristics of organs and lesions through biochemical reactions, forming spectral characteristics that are significantly different from healthy tissues;

[0112] Targeted performance: Selectively attach to or visualize pathogens and damaged tissue areas to improve detection accuracy.

[0113] As can be seen from the above, the system of the present invention can realize the early detection of diseases in the functional change stage (such as Helicobacter pylori infection and precancerous lesions) through color differences and pathogen pattern recognition, thereby greatly improving the cure rate of diseases;

[0114] A full-body scan can cover multiple diseased areas, reduce the economic burden of repeated visits and examinations, alleviate the workload of medical staff, concentrate human resources on the diagnosis and treatment of complex cases, and provide scientific and feasible personalized prevention and treatment plans based on the patient's specific test results and health risk scores.

[0115] Embodiment three:

[0116] This design is specifically used in hospitals and specialist clinics. As an innovative technology that combines medical imaging, artificial intelligence and big data, the pathogen pattern scanning system will play an important role in medical diagnosis, health management, scientific research and public health.

[0117] Provide pathogen classification and distribution information by detecting and monitoring infectious diseases (such as Helicobacter pylori, hepatitis B virus, Streptococcus pneumoniae, etc.);

[0118] Further, the pathogen pattern scanning system implementation case: early liver cancer screening and diagnosis;

[0119] like Figure 2 As shown, the pathogen pattern scanning system can achieve non-invasive screening and diagnosis of early liver cancer.

[0120] Patient: A 40-year-old male with a history of chronic hepatitis B and a recent physical examination revealed abnormal liver function.

[0121] Data Description

[0122] Input Data

[0123] Patient basic information:

[0124] Age: 40

[0125] Gender: Male

[0126] Medical history: Hepatitis B infection for 10 years, no record of cirrhosis

[0127] Drug development data:

[0128] The drug is metabolized and visualized in the liver lesion area, showing different spectral characteristics.

[0129] Data type: Multi-band spectral image acquired after development, resolution 1024×1024.

[0130] Database sample:

[0131] Historical cases: 3D images of 5,000 confirmed early liver cancer lesions.

[0132] Data format: Contains the characteristics of the lesion area (spectral characteristics, color distribution, morphology).

[0133] Output Data

[0134] 3D liver model: showing the difference in imaging between normal tissue and lesion areas.

[0135] Analysis report: location, area, color characteristics and risk score of the lesion area.

[0136] Medical advice: whether further biopsy or treatment is needed.

[0137] 3. Implementation steps

[0138] 1) Drug development and data acquisition

[0139] The patient takes contrast agents, which react with early lesion areas in the liver, showing specific spectral characteristics.

[0140] The system uses a multi-band spectral scanner to collect data and obtain the following information:

[0141] Spectral reflectance data:

[0142]

[0143] The results showed that the reflectivity of the diseased tissue was abnormally high in the infrared light range (λ≈750-900nm).

[0144] 3D modeling imaging data: Generate a 3D image model of the whole liver through voxel modeling.

[0145] 2) Data processing and analysis

[0146] Image Segmentation

[0147] Use Otsu algorithm to automatically segment the lesion area:

[0148] Calculate the optimal threshold σ B 2 , distinguishing the developed area from normal tissue.

[0149] Segmentation results: The lesion area occupies 3% of the total liver volume.

[0150] Feature extraction

[0151] Color characteristics: ΔR=0.15, ΔG=0.10, ΔB=0.08. There is a significant difference in red light reflection between the lesion area and the normal area.

[0152] Texture features: Gray-level co-occurrence matrix analysis showed that the lesion area had obvious changes in texture roughness, with an eigenvalue of 0.75.

[0153] Database matching

[0154] Use the cosine similarity algorithm to calculate the similarity between the current lesion and the sample in the database:

[0155]

[0156] The current lesion matched early liver cancer cases 92% of the time, placing it in the high-risk category.

[0157] Test result generation

[0158] Risk Scoring

[0159] The risk score is calculated based on the lesion area and color characteristics:

[0160]

[0161] Higher than the standard threshold of 0.7, indicating a high possibility of early-stage lesions.

[0162] Generate Report

[0163] The test report contains:

[0164] 3D liver image model: clearly shows the location and color changes of the lesion area.

[0165] Lesion analysis: lesion area size, characteristics and matching status.

[0166] Medical advice: Liver ultrasound and biopsy are recommended to confirm the diagnosis.

[0167] From the above, it can be seen that the precise positioning and pathogen pattern scanning system can achieve high-precision positioning of liver lesion areas with a resolution of millimeter level.

[0168] Through early screening, the system successfully detected early lesions with a matching accuracy rate of 92%, buying patients golden time for treatment.

[0169] Non-invasive diagnosis, the whole process is non-traumatic, only requires taking medicine and scanning, reducing patient discomfort and financial burden.

[0170] Intelligent assistance: The medical advice provided by the system is scientifically based, helping doctors make quick decisions and greatly improving diagnostic efficiency.

[0171] Effect comparison

[0172]

[0173] Through this embodiment, the pathogen pattern scanning system demonstrates its great potential in early disease screening and provides an effective tool for precision medicine.

[0174] Embodiment 4:

[0175] Pathogen pattern scanning system application implementation example

[0176] Target:

[0177] The system detects possible Helicobacter pylori infection in the stomach and generates visual test results and diagnostic recommendations.

[0178] Background and input parameters

[0179] Basic information of the patient:

[0180] Name: Wang

[0181] Age: 45

[0182] Symptoms: intermittent stomach pain, indigestion, no obvious vomiting or fever.

[0183] Testing process:

[0184] The patient takes the contrast agent, and the system uses a multi-band spectral scanner to collect data and obtain the following information:

[0185] Scanning frequency: 50fps.

[0186] Medicine movement speed: 1.5cm / s.

[0187] Helicobacter pylori characteristic data (from the database):

[0188] Standard color values: R1=230, G1=195, B1=140.

[0189] Characteristics of the lesion area: The tissue color is yellowish, accompanied by red spots.

[0190] Common lesion range: 0.5 to 3.0 cm in diameter.

[0191] System detection process

[0192] 1) Data collection and preprocessing

[0193] The medicine moves along the stomach, and the inner wall pattern data is obtained (the simulated data is as follows):

[0194]

[0195] Threshold setting: ∈=30.

[0196] When D(i)>∈, it is determined to be a lesion point.

[0197] 2) 3D modeling and image generation

[0198] Using the formula:

[0199] I(x,y,z)=∫ λ w(λ)R(x,y,z,λ)dλ

[0200] The data sampling points generate the following three-dimensional lesion distribution model:

[0201] Lesion distribution: Located in the middle of the stomach wall, irregular plaque in shape, about 2.1cm in area 2 .

[0202] Infection type: suspected Helicobacter pylori lesion areas appear distinctly yellow with red dots.

[0203] 3) Pathogen identification and classification

[0204] Using the Bayesian classification formula:

[0205]

[0206] Assume that the characteristic probability of Helicobacter pylori P(X|C Helicobacter pylori) = 0.85;

[0207] The sum of other possible pathogens∑ i＝1 P(XC i )P(C i )=0.95.

[0208] Result calculation: P(C Helicobacter pylori|X) = 0.85 × 0.6 / 0.95 = 0.536

[0209] Judgment: The possibility of Helicobacter pylori infection is 53.6%, and further testing is required (such as Helicobacter pylori antigen testing).

[0210] Results and report generation

[0211] Test report content:

[0212] Main findings: There is a suspected Helicobacter pylori infection area in the stomach, with a lesion area of ​​2.1cm 2 .

[0213] Lesion location: left area in the middle section of the stomach wall.

[0214] Health risk score (formula: R i =∑w i F ij ):

[0215] Genetic factors: w1=0.3, F11=1.0.

[0216] Eating habits: w2=0.4, F12=0.8.

[0217] Total risk score: R = 0.3 × 1.0 + 0.4 × 0.8 = 0.62 (medium risk).

[0218] Effect:

[0219] Provides 3D visualization of the entire stomach scan, marking the diseased and healthy areas.

[0220] Generates personalized health recommendations, including:

[0221] Helicobacter pylori antigen testing is recommended to confirm the diagnosis.

[0222] Avoid irritating foods and antibiotic treatment is recommended (if the infection is confirmed).

[0223] Conclusion and application value:

[0224] Detection efficiency: One test takes about 15 minutes, and the patient does not need anesthesia or surgical intervention.

[0225] Accuracy: Through the pathogen identification algorithm supported by big data, the initial diagnosis accuracy rate is over 85%.

[0226] Economical: Avoid multiple visits to the doctor and repeated examinations, and reduce patients' medical costs.

[0227] Scientific research value: Data can be fed back to the database to optimize the early diagnosis criteria for Helicobacter pylori infection.

[0228] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A pathogen pattern scanning system, characterized in that: include: Pattern Scanner Module: Uses multi-band spectral imaging technology combined with 3D modeling technology to scan the medicine swallowed inside the human body and generate high-resolution organ images; Data acquisition module: automatically collects spectral reflectance data after swallowing the drug, combines multi-band images using image fusion technology, segments the lesion area, and uses the lesion area to generate a 3D image with color identification; Data processing module: applying machine learning algorithms to extract features of the 3D image, automatically identifying, classifying and matching the 3D image of the lesion area with database similarity, locating the pathogen attachment site and evaluating the degree of lesion; Report generation module: Automatically generate test reports, including imaging area analysis, historical comparison data and medical advice.

2. The pathogen pattern scanning system according to claim 1, characterized in that: The expression for obtaining spectral data by the multi-band spectral imaging technology is: Where R(λ): reflectivity of the organ or tissue to wavelength λ; Lr(λ): reflected light intensity recorded by the detector; Li(λ): incident light intensity; By calculating the reflectivity at different wavelengths, the spectral characteristics of organ tissues are obtained, and the diseased area is distinguished from healthy tissues.

3. The pathogen pattern scanning system according to claim 1, characterized in that: The generation formula of the organ 3D image is: I(x,y,z)=∫ λ in(λ)R(x,y,z,λ)dλ Where I(x,y,z): the intensity value of the voxel in the 3D reconstruction; R(x,y,z,λ): reflectivity at position (x,y,z) at a specific wavelength; w(λ): weight of each band, used to adjust the importance of a specific wavelength; By fusing multi-band data, accurate 3D images are generated, highlighting the color differences in the developed areas.

4. The pathogen pattern scanning system according to claim 1, characterized in that: The segmented lesion area is obtained by using threshold segmentation and edge detection method, and the expression of the threshold segmentation is: Where ω1, ω2: the probability of pixels on both sides of the threshold; μ1, μ2: the mean of the pixels on both sides of the threshold; μT: global pixel mean; By maximizing the between-class variance Automatically calculate the optimal segmentation threshold; The expression of the edge detection method is: Where Gx, Gy: horizontal and vertical components of image gradient; G: Image gradient amplitude, indicating edge strength.

5. The pathogen pattern scanning system according to claim 1, characterized in that: The features of the 3D image include color features and texture features. By extracting the color and texture features, a high-dimensional feature vector is generated for subsequent lesion classification.

6. The pathogen pattern scanning system according to claim 5, characterized in that: The color feature is obtained by using the RGB difference formula, which is expressed as follows: ΔR=R 病变 -R 正常 ,ΔG=G 病变 -G 正常 ,ΔB=B 病变 -B 正常 Where ΔR, ΔG, ΔB: the color difference between the lesion and the normal area.

7. The pathogen pattern scanning system according to claim 6, characterized in that: The expression of the texture color feature is: Where P(i,j,d,θ): the co-occurrence probability of pixel pairs with grayscale values ​​i and j at distance d and direction θ.

8. The pathogen pattern scanning system according to claim 1, characterized in that: The evaluation of the lesion extent is performed by calculating the severity based on the regional area and the color intensity change, and the expression is: Where Ai: the area of ​​the i-th lesion area; Ii: color intensity of the i-th lesion area; Quantitatively evaluate the distribution and extent of the lesion area and generate the lesion risk level.

9. The pathogen pattern scanning system according to claim 1, characterized in that: The database similarity matching calculation expression is: in is the feature vector of the current lesion area and the sample in the database; is the magnitude of the vector.

10. The pathogen pattern scanning system according to claim 1, characterized in that: The ingredients of the medicament include the following conditions: Biocompatibility: non-toxic to human body, no side effects, and completely excreted; Development ability: It can change the color reflection characteristics of organs and lesions through biochemical reactions, forming spectral characteristics that are significantly different from healthy tissues; Targeting: Selectively attach to or visualize pathogens and damaged tissue areas.

Images