TB and NTM joint detection system and method based on multi-channel immunodetection

Abstract

The invention discloses a TB and NTM joint detection system and method based on multichannel immunodetection, and relates to the technical field of clinical detection. According to the invention, TB and NTM multi-target detection channels are integrated on the same immunochromatography platform, and specific antigen design, a multi-mode immunoreaction mechanism, an advanced signal processing algorithm and an extensible TDM function are combined, so that rapid, accurate and integrated identification of TB infection, NTM infection and co-infection states of the TB infection and the NTM infection is realized; the method is suitable for various application scenes such as primary medical institutions, mobile screening vehicles, border quarantine points and family self-test, and the early diagnosis efficiency and the precise treatment level of mycobacterium infection are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of clinical testing technology, specifically to a combined TB and NTM detection system and method based on multi-channel immunoassay. Background Technology

[0002] Currently, tuberculosis (TB) remains a serious public health threat globally. Non-tumor tract infection (NTM) can cause lung disease highly similar to TB, with patients often presenting with chronic cough, sputum production, hemoptysis, fatigue, and low-grade fever. Its chest imaging features (such as nodules, cavitation, and fibrosis) are almost indistinguishable from pulmonary tuberculosis. This overlap between clinical and imaging findings presents a significant challenge to clinical diagnosis: on the one hand, NTM infection is easily misdiagnosed as pulmonary tuberculosis, leading to unnecessary long-term anti-tuberculosis treatment, adverse drug reactions, and potential drug resistance; on the other hand, misdiagnosing active pulmonary tuberculosis as NTM delays effective treatment, increasing the risk of transmission and disease progression. Therefore, accurately and rapidly differentiating between TB and NTM infection has become an urgent clinical need.

[0003] Current diagnostic technologies have significant limitations. For tuberculosis, although molecular testing tools such as GeneXpert exist, their equipment is expensive and dependent on laboratory conditions, making them difficult to implement in resource-limited areas. Traditional smear microscopy has low sensitivity and is prone to missed diagnoses. While bacterial culture is the "gold standard," it takes 2-6 weeks, delaying treatment decisions. The situation is even more challenging for non-tumor infections: there is currently a lack of rapid point-of-care testing (POCT) methods specifically for NTM. Diagnosis often relies on bacterial identification after culture or on molecular technologies such as metagenomic sequencing. These methods are costly, time-consuming, and require specialized platforms and personnel. In mixed or cross-infection scenarios, the problem becomes even more complex, and clinicians often struggle to identify the pathogen composition quickly enough to develop precise treatment plans.

[0004] In addition, therapeutic drug monitoring (TDM) for tuberculosis is receiving increasing attention for optimizing treatment efficacy and reducing drug resistance. However, most current testing systems fail to integrate pathogen identification with drug concentration monitoring, resulting in a disconnect in the entire management process.

[0005] In summary, there is an urgent clinical need for a multi-channel fusion detection system that simultaneously meets the following criteria: First, it should enable rapid detection in approximately 10 minutes, truly meeting the needs of point-of-care testing (POCT) scenarios; second, it should be able to simultaneously differentiate between Mycobacterium tuberculosis and common non-tumor pathogens (NTMs) such as the avian intracellular mycobacterial complex (MAC), Mycobacterium abscessus (MAB), and Mycobacterium kansasus (KS); third, it should be able to identify mixed TB / NTM infections; and fourth, it should integrate therapeutic drug monitoring functions to provide a basis for personalized medication. Such a system would significantly improve the accuracy of early diagnosis, avoid misdiagnosis and mistreatment, control drug resistance, and optimize patient management, making it particularly suitable for widespread use in areas with a high burden of tuberculosis, primary healthcare institutions, and outpatient clinics. Summary of the Invention

[0006] This invention provides a combined TB and NTM detection system based on multi-channel immunoassay to overcome the technical challenge of existing clinical techniques that make it difficult to identify the composition of lung disease pathogens in a short time, thus making it impossible to formulate precise treatment plans.

[0007] The technical solution adopted in this invention is as follows: A combined TB and NTM detection system based on multi-channel immunoassay includes a multi-channel immunoassay card and a portable reading device; The multichannel immunoassay card is composed of a sample pad, a conjugate pad, a nitrocellulose membrane, an absorbent pad, and a base plate stacked and fixed in sequence. The nitrocellulose membrane has 4-15 independent detection channels, which are physically isolated to prevent signal interference. Each detection channel includes at least one tuberculosis (TB) detection channel, at least one non-tuberculous mycobacterium (NTM) detection channel, a quality control channel, a blank channel, and a cross-reaction monitoring channel. The TB detection channel is coated with a capture molecule targeting TB-specific antigens, and the NTM detection channel is coated with a capture molecule targeting NTM-specific antigens. Each detection channel has a width of 1.0-2.5 mm and a length of 8-15 mm. The portable reading device includes a signal acquisition module and an embedded processor. The signal acquisition module is used to acquire signals from each detection channel on the nitrocellulose membrane. The embedded processor is equipped with a signal processing algorithm module and a reading stability index evaluation unit. The signal processing algorithm module performs the following operations: S1. Perform background subtraction and normalization on the raw signals of each detection channel to obtain the normalized signal intensity of each channel; S2. The normalized signal intensity of each channel is independently judged based on a dynamic threshold mechanism. The dynamic threshold is adjusted in real time based on the distribution model of positive and negative samples continuously updated in the cloud database. S3. Based on the cross-channel deconvolution model, the combined signal intensity of the TB detection channel and the combined signal intensity of the NTM detection channel are processed to obtain the deconvolution signal value R. s The cross-channel deconvolution model is as follows: Among them, I TB I represents the normalized composite signal strength of the TB detection channel. NTM α represents the normalized synthesized signal strength of the NTM detection channel, and β represents constant parameters obtained by calibration using the training set. S4. Based on the deconvolution signal value R s The comparison results with the first preset threshold θ1 and the second preset threshold θ2, combined with the independent interpretation results of each channel, output the final diagnostic conclusion: if the deconvolution signal value R s If the value is greater than the first preset threshold θ1, it is determined to be TB positive; if the deconvolution signal value R s If the value is less than the second preset threshold θ2, it is determined to be NTM positive; if the deconvolution signal value R s The overall signal strength I of the TB detection channel is between the second preset threshold θ2 and the first preset threshold θ1. TB The signal strength I of the NTM detection channel exceeds its independent interpretation threshold. NTM If the signal intensity exceeds its independent interpretation threshold, it is determined to be a double infection of TB and NTM; if the normalized signal intensity of all detection channels does not exceed its corresponding independent interpretation threshold, it is determined to be negative. The interpretation stability index evaluation unit generates the interpretation stability index (RSI) by analyzing the morphological characteristics of the signal curves of each detection channel. The morphological characteristics include the signal peak width, rising edge slope, and signal consistency between channels. When the interpretation stability index (RSI) is lower than the set threshold, the system prompts for resampling or retesting.

[0008] Preferably, the TB-specific antigen is selected from at least one of ESAT-6, CFP-10, Ag85B, and tuberculous lipoarabinomannan (LAM); the NTM-specific antigen is selected from at least one of NTM-LAM, Mycobacterium avium complex (MAC)-specific antigen, and Mycobacterium abscessus (MAB)-specific antigen.

[0009] Preferably, the TB detection channel employs a sandwich immunoassay mode, with a detection antibody pre-loaded with a labeled signal molecule in the binding pad, and the capture molecule coated in the TB detection channel is an antibody that captures TB-specific antigens; the NTM detection channel employs a competitive immunoassay mode, with an NTM-specific antigen-antibody complex pre-loaded with a labeled signal molecule in the binding pad, and the capture molecule coated in the NTM detection channel is an antibody that captures NTM-specific antigens. The signal molecule is a colloidal gold particle or a rare earth-doped fluorescent microsphere, the particle size of which is controlled between 20 and 100 nm, and the concentration of the capture molecule in the corresponding detection channel is 0.5 to 1.5 mg / mL.

[0010] Preferably, the physical isolation structure is a micron-sized trench formed on the surface of the nitrocellulose membrane, the trench having a depth of 10~30 μm and a width of 50~150 μm.

[0011] Preferably, the TB detection channels are centrally arranged on one side, and the NTM detection channels are centrally arranged on the other side, with the TB detection channels and NTM detection channels separated by at least a blank channel and a cross-reaction monitoring channel.

[0012] Preferably, the detection channel further includes at least one drug monitoring channel, wherein the drug monitoring channel is encapsulated with a capture molecule targeting a specific therapeutic drug, and the binding pad is also preloaded with a signal molecule complex labeled with the specific therapeutic drug; the signal processing algorithm module of the portable reading device is further used to quantitatively calculate the concentration of the specific therapeutic drug in the sample based on the competitive signal intensity of the drug monitoring channel, and compare it with a preset treatment window range.

[0013] More preferably, the specific therapeutic drug is selected from at least one of bedaquiline, delamani, paromaline, moxifloxacin, and levofloxacin; the capture molecule coated in the drug monitoring channel is a specific antibody or receptor protein against the corresponding drug, and the coating concentration is 0.8~1.2 mg / mL.

[0014] The detection method using the aforementioned TB and NTM combined detection system includes the following steps: (1) Provide a biological sample to be tested and apply the biological sample to the sample pad of the multichannel immunoassay card; (2) The biological sample to be tested flows sequentially through the conjugation pad, nitrocellulose membrane and absorbent pad under capillary action, and a specific immune reaction occurs in each detection channel of the nitrocellulose membrane; (3) Use a portable reading device to collect signals from each detection channel on the nitrocellulose membrane; (4) The acquired signal is processed by the signal processing algorithm module built into the portable reading device. The processing includes background subtraction, normalization, independent channel interpretation based on dynamic threshold and comprehensive interpretation based on cross-channel deconvolution model. (5) Based on the signal processing results, output the diagnostic conclusions of TB positive, NTM positive, TB and NTM double infection or negative, and the interpretation stability index. When the interpretation stability index is lower than the set threshold, prompt for resampling or retesting.

[0015] Furthermore, the biological sample to be tested is whole blood, serum, plasma, or urine; when the biological sample to be tested is whole blood, a hemolysis buffer is added before or simultaneously with the application of the sample.

[0016] Furthermore, when the detection system includes a drug monitoring channel, step (5) also quantitatively calculates and outputs the concentration assessment result of a specific therapeutic drug in the sample based on the signal from the drug monitoring channel.

[0017] In summary, compared with the prior art, the present invention has the following advantages and beneficial effects: This invention integrates TB and NTM multi-target detection channels on the same immunochromatographic platform, and combines specific antigen design, multimodal immune response mechanisms, advanced signal processing algorithms, and scalable TDM functions to achieve rapid, accurate, and integrated identification of TB infection, NTM infection, and co-infection status. It is applicable to various scenarios such as primary healthcare institutions, mobile screening vehicles, border quarantine points, and home self-testing, significantly improving the efficiency of early diagnosis and precision treatment of mycobacterial infections. Detailed Implementation

[0018] The present invention will be described in detail below with reference to specific embodiments and examples, thereby making the advantages and various effects of the present invention more clearly apparent. Those skilled in the art should understand that these specific embodiments and examples are for illustrative purposes only and are not intended to limit the present invention.

[0019] Throughout this specification, unless otherwise specified, the terminology used herein should be understood to have the meaning commonly used in the art. Therefore, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. In the event of any conflict, this specification shall prevail.

[0020] Unless otherwise specified, all raw materials, reagents, instruments and equipment used in this invention can be purchased from the market or prepared by existing methods.

[0021] The technical solution of the present invention will be described in detail below with reference to specific embodiments, so that those skilled in the art can fully understand and implement the present invention.

[0022] Example 1 This embodiment provides a combined detection system integrating TB and NTM multi-target detection channels on the same immunochromatographic platform. The core component of the system is a multi-channel immunoassay card, which consists of a sample pad, a conjugate pad, a nitrocellulose membrane, an absorbent pad, and a base plate stacked and fixed in sequence. The sample pad is located at the front end and is used to receive the biological sample to be tested; the conjugate pad is immediately following it and contains a detection antibody or antigen-antibody complex labeled with a signal molecule; the nitrocellulose membrane serves as the core reaction area and has multiple independent detection channels; the absorbent pad is located at the end and is used to absorb excess liquid and maintain capillary flow; the base plate is a supporting structure made of polystyrene or polypropylene, which has good mechanical strength and chemical inertness, ensuring that the positions of each functional layer remain stable and do not deform or shift throughout the detection process.

[0023] In this embodiment, the nitrocellulose membrane is provided with 10 detection channels, including 4 tuberculosis (TB) detection channels (ESAT-6, CFP-10, Ag85B, TB-LAM), 3 non-tuberculous mycobacteria (NTM) detection channels (NTM-LAM, MAC antigen, MAB antigen), 1 quality control channel (for verifying the effectiveness of the detection process), 1 blank channel (for background noise calibration), and 1 cross-reactivity monitoring channel (for assessing the intensity of potential cross signals). The channels are isolated from each other by a physical isolation structure to prevent signal interference. This physical isolation structure is a micron-level trench formed on the surface of the nitrocellulose membrane by laser etching or imprinting, with a depth of 20 μm and a width of 100 μm, effectively blocking the lateral penetration of liquid between adjacent channels.

[0024] The Mycobacterium tuberculosis (TB) detection channel employs a sandwich immunoassay mode. Specifically, a detection antibody pre-loaded with colloidal gold particles or rare-earth-doped fluorescent microspheres is placed in the conjugate pad. This detection antibody recognizes non-overlapping epitopes on antigens (ESAT-6, CFP-10, Ag85B, TB-LAM). A capture antibody is coated at the corresponding channel position on the nitrocellulose membrane. This capture antibody recognizes another epitope on the same antigen. When the sample contains the target antigen, the antigen first binds to the labeled detection antibody in the conjugate pad to form a complex. This complex then flows with the chromatography solvent to the capture antibody region and is captured, forming a "capture antibody-antigen-labeled detection antibody" sandwich structure, thereby generating a detectable signal at the detection line.

[0025] Nontuberculous mycobacteria (NTM) detection channel: Employs a competitive immunoassay mode. An antigen-antibody complex pre-labeled with colloidal gold or fluorescent microspheres is loaded into the conjugate pad. This complex consists of antigens (NTM-LAM, MAC antigen, MAB antigen) and specific monoclonal antibodies pre-bound. Capture antibodies are coated at the corresponding channel positions on the nitrocellulose membrane. When the sample does not contain antigen, the pre-loaded labeled complex flows freely to the capture antibody region and binds, generating a strong signal. When the sample contains antigen, the free antigen competes with the labeled complex for binding to the capture antibody, resulting in a weakened detection line signal. The signal intensity is negatively correlated with the antigen concentration in the sample.

[0026] The capture molecules used in each channel were immobilized in different regions of the corresponding channel using a partitioned coating technique. The coating process was completed using a high-precision spray nozzle, with the spray volume of each nozzle controlled at 0.5–1.2 μL and the coating concentration ranging from 0.5–1.5 mg / mL. This concentration range has been experimentally verified to ensure stable signal response capabilities for each channel while avoiding non-specific adsorption or signal crosstalk between adjacent channels due to excessively high concentrations. After coating, the nitrocellulose membrane was dried at 37°C and 40–60% relative humidity for 2 hours, and then sealed in an aluminum foil bag containing desiccant for storage, with a shelf life of no less than 12 months.

[0027] The labeling agents (signal molecules) used in the conjugate pads are surface-modified colloidal gold particles or rare-earth-doped fluorescent microspheres. The particle size of the colloidal gold particles is generally controlled between 20 and 100 nm; the fluorescent microspheres are europium- or terbium-doped polystyrene microspheres, with the particle size also controlled between 20 and 100 nm, preferably 60 nm in this embodiment. The labeling process uses an EDC / NHS coupling chemical method to covalently link the antibody or antigen to the surface of the microspheres, with a labeling efficiency of not less than 85%. After labeling, the microspheres are purified by ultracentrifugation, resuspended in a buffer solution containing trehalose, BSA, and Tween-20, sprayed onto glass fiber conjugate pads, dried at 37°C for 2 hours, and then sealed for storage.

[0028] The channels on the nitrocellulose membrane are arranged asymmetrically. Specifically, the TB detection channels are concentrated on one side of the membrane, and the NTM detection channels are concentrated on the other side, separated by blank channels and cross-reaction monitoring channels. This maximizes the spatial distance between the TB and NTM detection channels, reducing lateral permeation effects caused by liquid diffusion. The width of each channel is set to 1.0–2.5 mm, preferably 1.8 mm in this embodiment; the length is 8–15 mm, preferably 12 mm in this embodiment. This size design ensures sufficient capture area to improve sensitivity while also considering chromatography speed, allowing the entire detection process to be completed within ten minutes.

[0029] The detection system provided in this embodiment also includes a portable reading device, which integrates a high-resolution CMOS imaging sensor, a controllable LED light source array, and an embedded processor. The light source system includes a white LED and excitation light of a specific wavelength (365 nm ultraviolet light is used in this embodiment to match the fluorescent microspheres), which can automatically switch according to the type of detection card. The imaging sensor has a resolution of 2048×1536 pixels and a dynamic range of no less than 12 bits, enabling precise capture of the signal intensity distribution of each channel. After the device is started, it automatically acquires images of the nitrocellulose membrane and transmits the raw data to the embedded processor to execute signal processing algorithms.

[0030] The signal processing algorithm module is deployed in the embedded processor of the reading device, and its execution flow is as follows: First, the original image undergoes background subtraction processing, eliminating interference from ambient light scattering and uneven reflection from the membrane substrate through Gaussian filtering and morphological operations; second, the signal intensity of each channel is normalized using the Z-score normalization method to ensure that all channel signals are under a uniform dimension; subsequently, a dynamic threshold mechanism is introduced, which is adjusted in real time based on a continuously updated positive / negative sample distribution model in a cloud database. Before each test, the device automatically downloads the latest calibration parameters via a wireless communication module (supporting Wi-Fi or 4G / 5G), including the mean μ and standard deviation σ of each channel. The dynamic interpretation threshold is set as μ + kσ, where k is a coefficient set according to clinical sensitivity requirements, typically ranging from 2.5 to 3.5.

[0031] Furthermore, the algorithm module employs a cross-channel deconvolution α-β weighting model to suppress potential cross-reactions between the TB and NTM channels. This model is defined as follows: R s = (I TB - I NTM × β) / α Among them, I TB This represents the combined signal strength of the TB detection channels after normalization, calculated as a weighted average of the signals from the ESAT-6, CFP-10, Ag85B, and TB-LAM channels; I NTM The normalized NTM detection channel's overall signal intensity is represented by the weighted average of the NTM-LAM, MAC antigen, and MAB antigen signals. α and β are constant parameters calibrated using a large training set. α reflects the true amplification factor of the TB signal, and β reflects the cross-interference coefficient of the NTM signal on the TB channel. In a training set containing 32,768 clinical samples, α = 1.08 and β = 0.23 were optimized by minimizing the false positive rate.

[0032] If the calculated R sIf the value is greater than the first preset threshold θ1 (in this embodiment, θ1 is preset to 0.85 based on multiple training results), it is determined to be TB positive; if R s If the value is less than the second preset threshold θ2 (in this embodiment, θ2 is preset to -0.65 based on multiple training results), it is determined to be an NTM positive; if R s Between θ2 and θ1, but the TB detection channel signal I TB Exceeding its independent interpretation threshold and the NTM channel synthesized signal I NTM If the test results exceed their respective independent interpretation thresholds, it is determined to be a double infection of TB and NTM; if none of the test channels exceed their corresponding interpretation thresholds, it is determined to be negative.

[0033] The result interpretation module outputs the final diagnostic conclusion based on the above algorithm and presents four distinct results in a visual interface: TB positive, NTM positive, TB+NTM double infection, and both negative.

[0034] To further ensure accurate detection, this embodiment also integrates a Reliability Stability Index (RSI) evaluation unit. RSI generates a stability score by analyzing the morphological characteristics of the signal curves of each channel. Specific parameters include signal peak width (FWHM), rise slope, signal plateau smoothness, and inter-channel consistency. The RSI calculation formula is: Among them, FWHM avg FWHM is the average of the full width at half maximum (FWHM) of each positive channel. ref The reference full width at half maximum (FWHM) for an ideal chromatographic peak (2.5 mm in this example); Slope avg Slope is the average slope of the rising edge. max σ is the theoretical maximum slope. channel With μ channel ω1, ω2, and ω3 are the standard deviation and mean of the signal strength of each channel, respectively; ω1, ω2, and ω3 are weighting coefficients, with a sum of 1. In this embodiment, ω1 is 0.4, ω2 is 0.3, and ω3 is 0.3. When the RSI is lower than the set threshold (0.7 in this embodiment), the system prompts the user "Insufficient signal quality, it is recommended to resample or retest".

[0035] The detection system provided in this embodiment was used to test 200 clinical samples, including serum and urine. Among the samples, 60 were confirmed TB patients, 50 were NTM infected, 30 were TB+NTM co-infected, and 60 were healthy controls. The same 200 samples were tested separately as control groups using a traditional single-channel TB test strip (detecting only CFP-10) and a separate NTM ELISA kit. The results are shown in Table 1. Table 1. Summary of test results for 200 clinical samples from different detection systems.

[0036] In Table 1, the cross-reactivity rate is defined as the proportion of NTM-infected individuals who are misdiagnosed as TB-positive, or the proportion of TB patients who are misdiagnosed as NTM-positive. This invention significantly reduces the cross-reactivity rate through antigen-specific design and an α-β deconvolution algorithm. Regarding co-infection identification, traditional methods, lacking a simultaneous detection mechanism, are completely unable to identify co-infection states. This invention, however, achieves high-accuracy co-infection determination through independent channel signal superposition and logical interpretation rules. Example

[0037] This embodiment, based on Embodiment 1, prepares a monitoring system including a therapeutic drug monitoring (TDM) extension module. This module is implemented by adding two dedicated detection channels to the nitrocellulose membrane of the original nitrocellulose detection card, specifically including a BDQ detection channel (for bedaquiline) and a DLM detection channel (for delamani). Each channel is coated with a specific antibody or receptor protein of the corresponding drug at a concentration of 0.8–1.2 mg / mL. The free drug in the sample competes with the pre-loaded labeled drug (i.e., drug-fluorescent microsphere conjugate) in the binding pad for binding to the captured molecules on the stationary phase, forming a competitive immune response. The signal intensity is negatively correlated with the drug concentration in the sample, and the blood drug concentration can be quantitatively calculated using a standard curve. The TDM module and the main detection module share the same reading platform and algorithm engine. The reading device has multiple sets of standard curve parameters built-in, covering the dose-response relationship of different drugs. After the test is completed, the system automatically calls the standard curve of the corresponding drug, converts the competing signal into a concentration value, and compares it with the therapeutic window range (e.g., the valley concentration of bedaquiline should be maintained at 0.5~2.0 μg / mL), generating prompts such as "normal concentration", "too low concentration" or "too high concentration" to assist clinical medication decisions.

[0038] Thirty TB patients receiving bedaquiline treatment were selected, and blood samples were simultaneously collected for TDM detection using the gold standard of this invention combined with high-performance liquid chromatography-mass spectrometry (HPLC-MS). The results showed that the correlation coefficient R between the bedaquiline concentration measured by this invention and the HPLC-MS results was [value missing]. 2 = 0.94, with an average deviation of 8.3%, which meets the requirements for clinical monitoring.

[0039] Finally, it should be noted that the terms "comprising," "including," or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make further changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0040] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various changes and improvements without departing from the concept of the technical solution of this application, and these all fall within the scope of protection of this application.

Claims

1. A combined TB and NTM detection system based on multi-channel immunoassay, characterized in that, Including multi-channel immune detection cards and portable reading devices; The multichannel immunoassay card is composed of a sample pad, a conjugate pad, a nitrocellulose membrane, an absorbent pad, and a base plate stacked and fixed in sequence. The nitrocellulose membrane has 4-15 independent detection channels, which are physically isolated to prevent signal interference. Each detection channel includes at least one tuberculosis (TB) detection channel, at least one non-tuberculous mycobacterium (NTM) detection channel, a quality control channel, a blank channel, and a cross-reaction monitoring channel. The TB detection channel is coated with a capture molecule targeting TB-specific antigens, and the NTM detection channel is coated with a capture molecule targeting NTM-specific antigens. Each detection channel has a width of 1.0-2.5 mm and a length of 8-15 mm. The portable reading device includes a signal acquisition module and an embedded processor. The signal acquisition module is used to acquire signals from each detection channel on the nitrocellulose membrane. The embedded processor is equipped with a signal processing algorithm module and a reading stability index evaluation unit. The signal processing algorithm module performs the following operations: S1. Perform background subtraction and normalization on the raw signals of each detection channel to obtain the normalized signal intensity of each channel; S2. The normalized signal intensity of each channel is independently interpreted based on a dynamic threshold mechanism, wherein the dynamic threshold is adjusted in real time based on the distribution model of positive and negative samples continuously updated in the cloud database; S3. Based on the cross-channel deconvolution model, the combined signal intensity of the TB detection channel and the combined signal intensity of the NTM detection channel are processed to obtain the deconvolution signal value R. s The cross-channel deconvolution model is as follows: Among them, I TB I represents the normalized composite signal strength of the TB detection channel. NTM α represents the normalized synthesized signal strength of the NTM detection channel, and β represents constant parameters obtained by calibration using the training set. S4. Based on the deconvolution signal value R s The comparison results with the first preset threshold θ1 and the second preset threshold θ2, combined with the independent interpretation results of each channel, output the final diagnostic conclusion: if the deconvolution signal value R s If the value is greater than the first preset threshold θ1, it is determined to be TB positive; if the deconvolution signal value R s If the value is less than the second preset threshold θ2, it is determined to be NTM positive; if the deconvolution signal value R s The overall signal strength I of the TB detection channel is between the second preset threshold θ2 and the first preset threshold θ1. TB The signal strength I of the NTM detection channel exceeds its independent interpretation threshold. NTM If the signal intensity exceeds its independent interpretation threshold, it is determined to be a double infection of TB and NTM; if the normalized signal intensity of all detection channels does not exceed its corresponding independent interpretation threshold, it is determined to be negative. The interpretation stability index evaluation unit generates the interpretation stability index (RSI) by analyzing the morphological characteristics of the signal curves of each detection channel. The morphological characteristics include the signal peak width, rising edge slope, and signal consistency between channels. When the interpretation stability index (RSI) is lower than the set threshold, the system prompts for resampling or retesting.

2. The TB and NTM combined detection system as described in claim 1, characterized in that, The TB-specific antigen is selected from at least one of ESAT-6, CFP-10, Ag85B, and tuberculous lipoarabinomannan (LAM); the NTM-specific antigen is selected from at least one of NTM-LAM, Mycobacterium avium complex (MAC)-specific antigen, and Mycobacterium abscessus (MAB)-specific antigen.

3. The TB and NTM combined detection system as described in claim 1, characterized in that, The TB detection channel employs a sandwich immunoassay mode, with a detection antibody pre-loaded with a labeled signal molecule in the binding pad. The capture molecule coated in the TB detection channel is an antibody that captures TB-specific antigens. The NTM detection channel employs a competitive immunoassay mode, with an NTM-specific antigen-antibody complex pre-loaded with a labeled signal molecule in the binding pad. The capture molecule coated in the NTM detection channel is an antibody that captures NTM-specific antigens. The signal molecule is a colloidal gold particle or a rare earth-doped fluorescent microsphere, with the particle size controlled between 20 and 100 nm. The concentration of the capture molecule in the corresponding detection channel is 0.5 to 1.5 mg / mL.

4. The TB and NTM combined detection system as described in claim 1, characterized in that, The physical isolation structure is a micron-sized trench formed on the surface of the nitrocellulose membrane, the trench having a depth of 10~30 μm and a width of 50~150 μm.

5. The TB and NTM combined detection system as described in claim 1, characterized in that, The TB detection channels are centrally arranged on one side, and the NTM detection channels are centrally arranged on the other side. The TB detection channels and the NTM detection channels are separated by at least a blank channel and a cross-reaction monitoring channel.

6. The TB and NTM combined detection system as described in claim 1, characterized in that, The detection channel further includes at least one drug monitoring channel, which is encapsulated with a capture molecule targeting a specific therapeutic drug. The binding pad is also preloaded with a signal molecule complex labeled with the specific therapeutic drug. The signal processing algorithm module of the portable reading device is also used to quantitatively calculate the concentration of the specific therapeutic drug in the sample based on the competing signal intensity of the drug monitoring channel and compare it with a preset treatment window range.

7. The TB and NTM combined detection system as described in claim 6, characterized in that, The specific therapeutic drug is selected from at least one of bedaquiline, delamani, paromaline, moxifloxacin, and levofloxacin; the capture molecule coated in the drug monitoring channel is a specific antibody or receptor protein against the corresponding drug, and the coating concentration is 0.8~1.2 mg / mL.

8. A detection method using the TB and NTM combined detection system as described in any one of claims 1 to 7, characterized in that, The method includes the following steps: (1) Provide a biological sample to be tested and apply the biological sample to the sample pad of the multichannel immunoassay card; (2) The biological sample to be tested flows sequentially through the conjugation pad, nitrocellulose membrane and absorbent pad under capillary action, and a specific immune reaction occurs in each detection channel of the nitrocellulose membrane; (3) Use a portable reading device to collect signals from each detection channel on the nitrocellulose membrane; (4) The acquired signal is processed by the signal processing algorithm module built into the portable reading device. The processing includes background subtraction, normalization, independent channel interpretation based on dynamic threshold and comprehensive interpretation based on cross-channel deconvolution model. (5) Based on the signal processing results, output the diagnostic conclusions of TB positive, NTM positive, TB and NTM double infection or negative, and the interpretation stability index. When the interpretation stability index is lower than the set threshold, prompt for resampling or retesting.

9. The detection method as described in claim 8, characterized in that, The biological sample to be tested is whole blood, serum, plasma, or urine; when the biological sample to be tested is whole blood, a hemolysis buffer is added before or simultaneously with the application of the sample.

10. The detection method as described in claim 8, characterized in that, When the detection system includes a drug monitoring channel, step (5) also quantitatively calculates and outputs the concentration assessment result of a specific therapeutic drug in the sample based on the signal from the drug monitoring channel.

Images