A liver cancer detection system, kit and application thereof

Abstract

The embodiments of the present specification provide a liver cancer detection system, a kit and applications thereof, including a sample processing module, an evaluation module and an interaction module; the sample processing module is configured to determine the Ct value of the gene to be detected in the target sample corresponding to the subject; the evaluation module is configured to determine a comprehensive evaluation value based on the Ct value, and determine a comprehensive judgment result based on the comprehensive evaluation value; the interaction module is configured to display the comprehensive judgment result and perform risk warning based on the comprehensive judgment result.

Description

Technical Field

[0001] The present specification relates to the biomedical field, and in particular to a liver cancer detection system, a kit and applications thereof. Background Art

[0002] Primary liver cancer is a common digestive system malignancy worldwide. According to new data released by GLOBOCAN in 2018, the number of new cases of liver cancer worldwide reached 841,000 per year. There are no obvious symptoms in the early stages of liver cancer. The detection rate of early liver cancer by traditional early screening methods is low, and liver cancer cannot be directly observed and diagnosed using laparoscopic technology, which increases the difficulty of early screening and diagnosis of liver cancer.

[0003] Therefore, there is an urgent need to propose a liver cancer detection system, a kit and its application to provide a reasonable reference for the early diagnosis of liver cancer. Summary of the invention

[0004] One or more embodiments of the present specification provide a liver cancer detection system, including a sample processing module, an evaluation module and an interaction module; the sample processing module is configured to determine the Ct value of the gene to be detected in the target sample corresponding to the subject; the evaluation module is configured to determine a comprehensive evaluation value based on the Ct value, and determine a comprehensive judgment result based on the comprehensive evaluation value; the interaction module is configured to display the comprehensive judgment result and issue a risk warning based on the comprehensive judgment result.

[0005] One of the embodiments of the present specification provides a liver cancer detection kit, which is applied to the aforementioned liver cancer detection system, including a first detection reaction liquid, a second detection reaction liquid, an enzyme system, a negative control and a positive control; the first detection reaction liquid includes a HOXA1 forward primer, a HOXA1 reverse primer, a HOXA1 probe, a NKX6-2 forward primer, a NKX6-2 reverse primer, a NKX6-2 probe, an ACTB forward primer, an ACTB reverse primer, an ACTB probe, a dNTP mixed solution and a Buffer buffer; the second detection reaction liquid includes a GNB4 forward primer, a GNB4 reverse primer, a GNB4 probe, a TSPYL5 forward primer, a TSPYL5 reverse primer, a TSPYL5 probe, an ACTB forward primer, an ACTB reverse primer, an ACTB probe, a dNTP mixed solution and a Buffer buffer. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0007] Figure 1 is an exemplary module diagram of a liver cancer detection system according to some embodiments of this specification;

[0008] Figure 2 is an exemplary flow chart of the application of the liver cancer detection system according to some embodiments of this specification;

[0009] Figure 3 It is a schematic diagram of determining the gene to be detected according to some embodiments of this specification;

[0010] Figure 4 is a schematic diagram of a risk assessment model according to some embodiments of this specification;

[0011] Figure 5 is a schematic diagram of determining the periodic review interval duration according to some embodiments of this specification;

[0012] Figure 6 is a ROC curve shown in some embodiments of the present specification. DETAILED DESCRIPTION

[0013] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0014] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0015] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0016] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 It is an exemplary module diagram of the liver cancer detection system shown in the embodiment of this specification.

[0018] In some embodiments, the liver cancer detection system 100 includes a sample processing module 110 , an evaluation module 120 , and an interaction module 130 .

[0019] The sample processing module 110 is configured to determine the Ct value of the gene to be detected in the target sample corresponding to the subject.

[0020] The evaluation module 120 is configured to determine a comprehensive evaluation value based on the Ct value, and determine a comprehensive judgment result based on the comprehensive evaluation value.

[0021] In some embodiments, the evaluation module 120 is further configured to determine a comprehensive evaluation value based on the Ct value of the gene to be detected through a risk evaluation model.

[0022] The interactive module is configured to display the comprehensive judgment result and issue a risk warning based on the comprehensive judgment result.

[0023] For more information about the sample processing module and the evaluation module, see Figure 2 and Figure 4 For more information about the interaction module, see Figure 5 and its related description.

[0024] It should be noted that the above description of the liver cancer detection system and its modules is only for convenience of description and does not limit the present specification to the scope of the embodiments. It is understandable that, after understanding the principle of the system, those skilled in the art may arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from the principle. In some embodiments, Figure 1 The sample processing module 110, the evaluation module 120, and the interaction module 130 disclosed in the specification can be different modules in a system, or a module can realize the functions of two or more modules. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the scope of protection of this specification.

[0025] Figure 2 is an exemplary flow chart of a liver cancer detection application according to some embodiments of this specification. In some embodiments, process 200 can be performed by various modules of liver cancer detection system 100. The operation diagram of process 200 presented below is illustrative. In some embodiments, one or more additional operations not described and / or one or more operations not discussed can be used to complete the process. In addition, Figure 2The order in which the operations of flow 200 are illustrated and described below is not intended to be limiting.

[0026] Step 210, determining the Ct value of the gene to be detected in the target sample corresponding to the subject.

[0027] The target sample corresponding to the subject may be blood (eg, peripheral blood), tissue sample, urine, ascites, etc.

[0028] The gene to be detected refers to a gene related to liver cancer detection.

[0029] In some embodiments, the sample processing module can determine at least one gene to be evaluated based on the methylation sequencing results, and determine the gene to be detected based on the at least one gene to be evaluated. For more detailed information about methylation sequencing and determining the gene to be detected, see Figure 3 and its related description.

[0030] In some embodiments, the genes to be detected include NKX6-2, GNB4, HOXA1, and TSPYL5.

[0031] The Ct value refers to the number of cycles experienced when the fluorescent signal in each reaction tube reaches the set threshold in real-time fluorescence quantitative PCR.

[0032] In some implementations, the sample processing module can detect the pre-treated target sample by real-time fluorescence quantitative PCR to obtain the Ct value of the gene to be detected. As an example only, the sample processing module can process the target sample by a kit to determine the Ct value of the gene to be detected in the target sample. The target sample can be the blood of the subject, and the pre-treatment process can include: plasma separation of the target sample, extraction of peripheral blood; using a free DNA extraction reagent to extract nucleic acids by a magnetic bead method; using a DNA bisulfite conversion reagent to convert its DNA through sulfite conversion, adding a specific capture sequence during the conversion process to enrich the DNA, and obtaining a pre-treated sample.

[0033] For more information about the kit, please refer to this instruction manual. Figure 3 Related description in .

[0034] In some embodiments, the sample processing module can perform methylation detection of the gene to be detected on the pre-treated sample, and determine the Ct value of the gene to be detected by fluorescent quantitative PCR. The Ct value of the gene to be detected may include the Ct value of NKX6-2, GNB4, HOXA1 and TSPYL5.

[0035] Step 220, determining a comprehensive evaluation value based on the Ct value, and determining a comprehensive judgment result based on the comprehensive evaluation value.

[0036] The comprehensive assessment value refers to a numerical value that reflects the subject's risk of developing liver cancer.

[0037] In some embodiments, the comprehensive evaluation value can be expressed as a numerical value, for example, a numerical value between 0-100%, or other numerical ranges, etc. The higher the comprehensive evaluation value, the higher the risk of the subject suffering from liver cancer.

[0038] In some embodiments, the evaluation module may determine the comprehensive evaluation value based on the Ct value of the gene to be detected by querying a reference risk table, wherein the reference risk table includes at least one reference Ct value of the gene to be detected and its corresponding reference risk value, which may be constructed based on historical data.

[0039] In some embodiments, the evaluation module may determine a comprehensive evaluation value based on the Ct value of the gene to be detected through a risk evaluation model.

[0040] In some embodiments, the risk assessment model may include at least one of a logistic regression model, a random forest model, a SVM model, or other models capable of performing logical analysis.

[0041] As an example only, the evaluation module can determine the joint formula between the Ct value of at least one gene to be detected in the historical data and the comprehensive evaluation value through at least one of the logistic regression model, the random forest model, and the SVM model, and determine the comprehensive evaluation value of the current subject based on the determined joint formula and the current Ct value corresponding to at least one gene to be detected of the current subject.

[0042] In some embodiments, the genes to be detected may include HOXA1, NKX6-2, GNB4, and TSPYL5, and the joint formula between the Ct value and the comprehensive evaluation value of at least one gene to be detected may include at least one of the interpretation based on the Ct value and the analysis based on dataCT (gene-IC). Among them, the interpretation based on the Ct value has a specificity of up to 96.90% and a sensitivity of up to 94.78%; the specificity based on the dataCT (gene-IC) analysis can reach 99.22% and a sensitivity of up to 89.55%. It can be seen that the interpretation based on the Ct value has a higher sensitivity, and the analysis based on dataCT (gene-IC) has a better specificity, which can be selected according to actual needs.

[0043] In some embodiments, the interpretation based on the Ct value can be performed based on the following formula (I): Formula (I)

[0044] In the formula, is the liver cancer comprehensive index, for Ct value, for The Ct value, for The Ct value, for Ct value.

[0045] In some embodiments, the dataCT (gene-IC) analysis may be performed based on the following formula (II): Formula (II)

[0046] Where, LIVER-CRA2 is the liver cancer comprehensive index, is the dataCt value of NKX6, is the dataCt value of GNB4, is the dataCt value of HOXA1, is the dataCt value of TSPYL5.

[0047] In some embodiments, the risk assessment model may include a machine learning model, which may include at least one sub-model and a fusion layer corresponding to the gene to be detected, the input may include a Ct value corresponding to at least one gene to be detected, and the output may include a comprehensive assessment value. For more information about the risk assessment model, see Figure 4 and its related description.

[0048] The comprehensive judgment result refers to the result of judging the risk of cancer.

[0049] In some embodiments, the comprehensive judgment result may include one of low risk and high risk.

[0050] In some embodiments, the evaluation module can determine a comprehensive judgment result based on the comprehensive evaluation value and a preset threshold value, and in response to the comprehensive evaluation value being less than the preset threshold value, the comprehensive judgment result of the subject's risk of liver cancer is judged to be low risk. Exemplarily, when judging based on the Ct value, the preset threshold value can be -0.26, if the comprehensive evaluation value is less than -0.26, the subject is judged to have a low risk of liver cancer; if the comprehensive evaluation value is greater than or equal to -0.26, the subject is judged to have a high risk of liver cancer; when judging based on the dataCt value, the preset threshold value can be -1.031, if the comprehensive evaluation value is less than -1.031, the subject is judged to have a low risk of liver cancer; if the comprehensive evaluation value is greater than or equal to -1.031, the subject is judged to have a high risk of liver cancer.

[0051] In some embodiments, the preset threshold may be determined based on historical experience.

[0052] In some embodiments, the preset threshold can also be determined based on the basic information of the subject and the Ct value of the gene to be detected.

[0053] Among them, the basic information of the subject may include age, gender, living habits, etc. The aforementioned living habits may include living habits that have a greater impact on the liver, such as whether staying up late, the degree of staying up late, whether drinking alcohol, the degree of drinking alcohol, etc.

[0054] In some embodiments, the evaluation module can construct a first feature vector based on the basic information of the subject and the Ct value of the gene to be detected. Exemplarily, the first feature vector can be expressed as (age, gender, living habit 1, degree of living habit 1, living habit 2, degree of living habit 2, Ct value of gene 1, Ct value of gene 2, Ct value of gene 3, Ct value of gene 4), and the degree corresponding to the living habit can be pre-set to the level corresponding to different situations, such as staying up late, staying up until 10 o'clock every day is level one, staying up until midnight every day is level two, staying up until two o'clock every day is level three, etc. For example, the first feature vector (40, 1, 1, 3, 0, 0, 20, 15, 35, 27) indicates that the subject is 40 years old; male; living habit 1 (such as alcoholism) exists, the corresponding degree is level 3; living habit 2 does not exist, the corresponding degree is level 0; the Ct value of gene 1 is 20; the Ct value of gene 2 is 15; the Ct value of gene 3 is 35; the Ct value of gene 4 is 27.

[0055] In some embodiments, the evaluation module can determine the preset threshold corresponding to the subject by querying the reference threshold table based on the first feature vector. The reference threshold table may include at least one reference feature vector and its corresponding reference threshold. The reference feature vector and its corresponding reference threshold can be determined based on historical data. For example, for a reference feature vector, subjects with a comprehensive evaluation index higher than 0.7 in historical data similar to the feature vector were diagnosed with liver cancer in subsequent examinations, then 0.7 can be determined as the reference threshold corresponding to the reference feature vector.

[0056] Step 230, display the comprehensive judgment result, and issue a risk warning based on the comprehensive judgment result.

[0057] In some embodiments, the interactive module can display the comprehensive judgment result and issue a risk warning based on the comprehensive judgment result.

[0058] Risk warning refers to the corresponding risk warning to the subjects based on the comprehensive judgment results. For more information about risk warning, see Figure 5 and its related description.

[0059] In some embodiments, the assessment module can prompt the patient to undergo corresponding follow-up examinations based on the comprehensive judgment result. For example, if the comprehensive judgment result is high risk, the risk warning includes prompting the subject to undergo a special examination; for another example, if the comprehensive judgment result is low risk, the risk warning includes prompting the subject to undergo a regular review.

[0060] The embodiments of this specification use a liver cancer detection system to collect multiple genes to be detected in the subject's sample, and obtain the Ct value of the gene to be detected, determine a comprehensive evaluation value based on the Ct value, and then determine a comprehensive judgment result, and provide risk warnings to the subject based on the comprehensive judgment result, which can effectively improve the accuracy of early diagnosis of liver cancer and provide timely reminders when the subject is at high risk of cancer.

[0061] It should be noted that the above description of the process 200 is only for example and illustration, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the process 200 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0062] Figure 3 It is a schematic diagram of determining the gene to be detected according to some embodiments of this specification.

[0063] In some embodiments, the sample processing module may further determine at least one gene to be evaluated 310 based on the methylation sequencing result 410 , and determine a gene to be detected 320 based on the at least one gene to be evaluated 310 .

[0064] In some embodiments, the sample processing module can determine the gene loci with significant differences based on the methylation sequencing results of hepatocellular carcinoma tissue and adjacent tissue, and use them as genes to be evaluated. The methylation sequencing results can include information such as methylated sites, methylation levels, and methylation patterns, usually presented in the form of a data set.

[0065] As an example only, the sample processing module can perform layered filtering on the initial sample (including the aforementioned hepatocyte self-tissue and adjacent tissue) by reduced representation bisulfite sequencing (RRBS sequencing technology) to obtain the processed sample, and extract the methylation ratio of the processed sample by sequence alignment; differentially methylated regions (DMR) detection is used to find the differentially methylated regions in the processed sample, and through further filtering, the abnormally methylated genes related to hepatocellular carcinoma are determined for further research. In some embodiments, the aforementioned filtering conditions may include the following conditions: removing gene sites with a CpG density of 0.05-0.3, removing gene sites with empty annotations for gene promoters and gene bodies, removing X chromosome and Y chromosome genes, and removing chrUn genes.

[0066] In some embodiments, the liver cancer cell-related abnormal methylation genes may include at least one of NKX6-2, B4GALNT, TSPYL5, GNB4, HOXA1, LINC00682, RASSF1, TSPAN18, GRASP-AS1, and EMX1.

[0067] In some embodiments, the sample processing module can further screen the above liver cancer cell-related abnormal methylation genes to determine the genes to be evaluated.

[0068] As an example only, BSP sequencing primers can be designed 200bp upstream and downstream in the above-mentioned region for the above-mentioned liver cancer cell-related abnormal methylation genes, and 10 liver cancer tissues and 10 adjacent cancer tissues from early liver cancer patients are selected for Sanger methylation sequencing. The gene methylation differences between liver cancer tissues and adjacent cancer tissues are compared, and the target genes are further filtered and screened, and the gene loci with significant differences between early liver cancer tissues and adjacent cancer tissues are identified as genes to be evaluated.

[0069] In some embodiments, a differential analysis is performed based on the methylation sequencing results, and seven gene loci regions with significant differences in adjacent cancer tissues and liver cancer tissues, including NKX6-2, B4GALNT, RASSF1, EMX1, GNB4, HOXA1, and TSPYL5, are screened out as genes to be evaluated.

[0070] In some embodiments, the sample processing module can further verify the aforementioned at least one gene to be evaluated. As an example only, the gene to be evaluated can be combined with MethPrimer Logo methylation primer design software to design MSP primers, and based on SYBR dye method fluorescence quantitative PCR, the plasma samples of liver cancer and other digestive tract cancers (gastric cancer, esophageal cancer, intestinal cancer, etc.) are tested, and the gene to be detected is further determined by performing ROX curve analysis on the test results.

[0071] In some embodiments, genes with an area under the curve (AUC) greater than 0.9 can be selected as genes to be detected.

[0072] The AUC calculation results of the significantly different gene loci in the above paracancerous tissues and liver cancer tissues are shown in Table 1:

[0073] Table 1 AUC calculation results

[0074] Gene segment AUC value NKX62 0.997 B4GALNT 0.818 RASSF1A 0.742 EMX1 0.942 GNB4 0.992 HOXA1 0.983 TSPYL5 1.000

[0075] As can be seen from the above table, in some embodiments, the genes to be detected may include NKX6-2, EMX1, GNB4, HOXA1 and TSPYL5.

[0076] In some embodiments, the sample processing module can design methylation primers and probes based on the nucleic acid sequence of at least one gene to be detected, screen PCR probes and primers for methylation of related genes according to the characteristics of the primer probes themselves and the conditions that the methylation primers need to meet, and synthesize the designed primers and probes. The specific sequences are shown in Table 2:

[0077] Table 2 Primer and probe sequences to be synthesized

[0078]

[0079] In some embodiments, for each gene to be detected, 2-3 groups can be set up to form a double system with the housekeeping gene ACTB, and the prepared liver cancer positive cell line bisDNA gradient (gradient is 5-fold dilution) is used for testing and evaluation, and the combination is screened by amplification curve, fluorescence increment and sensitivity, and finally it is determined that each gene region is set up with a group and the housekeeping gene ACTB to form a double system, and the specific sequence is shown in Table 3;

[0080] Table 3 Specific primers and probes

[0081]

[0082] After the above combination and housekeeping gene composition dual system, the primer combination was further evaluated through multiple groups of samples to screen the optimal combination for multiple system construction. The aforementioned multiple groups of samples included healthy plasma samples from the control group and liver cancer plasma samples. The test results are shown in Table 4;

[0083] Table 4 Test results

[0084] , ,

[0085]

[0086] In some embodiments, the above results can be preliminarily studied by ROX curve according to statistical analysis, and the Ct value corresponding to the maximum value of the equal index can be used as the cutoff value for data analysis to analyze the data. The ROC curve is as follows: Figure 6 As shown, the diagnostic ability of EMX1 gene is lower than that of the other four genes (NKX6-2, GNB4, HOXA1 and TSPYL5), and the detection abilities of the four genes NKX6-2, GNB4, HOXA1 and TSPYL5 are complementary to each other. Therefore, in some embodiments, the kit can be constructed and optimized based on the four genes NKX6-2, GNB4, HOXA1 and TSPYL5.

[0087] In some embodiments, a triple system can be cross-constructed according to the four gene probe modification groups of NKX6-2, GNB4, HOXA1 and TSPYL5 to obtain four test combinations, as shown in Table 5:

[0088] Table 5 Triple system test combination

[0089]

[0090] In some embodiments, optimization and amplification can be performed based on the above test combinations to obtain amplification curves, and two groups of combinations with higher sensitivity can be selected for the construction of the kit.

[0091] In some embodiments, the amplification effect and sensitivity of combination 1 and combination 3 are lower than those of combination 2 and combination 4, as shown in Table 6:

[0092] Table 6 Kit components

[0093]

[0094] In some embodiments, the gene to be detected can be processed based on a kit, and the kit may include a first detection reaction solution and a second detection reaction solution. The first detection reaction solution includes a HOXA1 forward primer, a HOXA1 reverse primer, a HOXA1 probe, a NKX6-2 forward primer, a NKX6-2 reverse primer, a NKX6-2 probe, an ACTB forward primer, an ACTB reverse primer, an ACTB probe, a dNTP mixed solution, and a Buffer buffer; the second detection reaction solution includes a GNB4 forward primer, a GNB4 reverse primer, a GNB4 probe, a TSPYL5 forward primer, a TSPYL5 reverse primer, a TSPYL5 probe, an ACTB forward primer, an ACTB reverse primer, an ACTB probe, a dNTP mixed solution, and a Buffer buffer. The dNTP mixed solution may be a 5×buffer solution, and the Buffer buffer may be a 25 mM dNTPs solution.

[0095] In some embodiments, the aforementioned kit further comprises the following components: enzyme system, negative control and positive control. As an example only, the aforementioned enzyme system may include HiTaq DNA Polymerase, HiTaq DNA Polymerase diluent; the negative control may include non-liver cancer cell line DNA; the positive control may include liver cancer cell line DNA.

[0096] In some embodiments, the detection effect of the kit can be analyzed and determined by testing the plasma samples of the validation set using the kit, wherein the validation set includes a plurality of positive samples including liver cancer cells and a plurality of negative samples including paracancerous cells.

[0097] In some embodiments, for each test of the samples in the validation set, real-time fluorescence quantitative PCR is used to obtain the Ct values ​​of the methylation genes of NKX6-2, GNB4, HOXA1 and TSPYL5, and the Ct values ​​of the tested genes are substituted into the joint formula.

[0098] As an example only, the interpretation can be based on the calculation results of the above formula (I). As an example only, LIVER-CRA1 is the risk score for liver cancer, the standard is LIVER-CRA1=-0.260, LIVER-CRA1>-0.260 is negative, and LIVER-CRA<-0.260 is positive. For detailed description of the combined formula and formula (I), please refer to this manual Figure 2 The interpretation results are shown in Table 7:

[0099] Table 7 Interpretation results

[0100] ,

[0101] ,

[0102] ,

[0103] ,

[0104]

[0105] According to the test results of the samples in the validation set, when calculated using the calculation formula LIVER-CRA1, the sensitivity was 94% and the specificity was 94.19%.

[0106] By using the kit provided in some embodiments of the present specification, the positive detection rate and specificity of early hepatocellular carcinoma can be significantly improved, and the detection sensitivity of biomarkers can be increased to picograms / nanograms of DNA molecules; by jointly detecting multi-gene methylated DNA in plasma samples, a higher sensitivity in the detection of liver cancer, especially advanced tumors, can be achieved compared to serum marker detection methods, which can significantly improve the detection rate of tumors.

[0107] In some embodiments of the present specification, based on the methylation sequencing results, the genes to be evaluated related to liver cancer are preliminarily screened, and the genes to be evaluated are screened to further determine the genes to be detected, which can effectively narrow the detection scope and thus improve the detection efficiency.

[0108] Figure 4 is an exemplary schematic diagram of a risk assessment model according to some embodiments of the present specification.

[0109] The risk assessment model refers to a model used to assess the comprehensive liver cancer index. In some embodiments, the risk assessment model can be a machine learning model, such as a neural network (NN) model, or other machine learning models that can achieve corresponding functions.

[0110] In some embodiments, the input of the risk assessment model includes a Ct value 440 corresponding to at least one gene to be detected, for example, the Ct value of gene 1 to be detected, the Ct value of gene 2 to be detected, ..., the Ct value of gene N to be detected, etc.; the output includes a comprehensive evaluation index 490 of the subject. For detailed descriptions of Ct values ​​and comprehensive evaluation indexes, please refer to the specification Figure 2 Related description in .

[0111] In some embodiments, the genes to be detected may include NKX6-2, GNB4, HOXA1, and TSPYL5.

[0112] In some embodiments, the risk assessment model includes at least one sub-model 450 corresponding to the gene to be detected, and a fusion layer 470 .

[0113] A submodel refers to a model for estimating a comprehensive evaluation value based on a gene to be detected. In some embodiments, each submodel corresponds to a gene to be detected. For example, when the genes to be detected include NKX6-2, GNB4, HOXA1, and TSPYL5, the corresponding number of submodels is 4, corresponding to the aforementioned four genes to be detected.

[0114] like Figure 4 As shown, in some embodiments, the input of submodel 450 may include the Ct value of a gene to be detected, and the output includes an intermediate result determined based on the aforementioned gene to be detected. For example, the Ct value of gene 1 to be detected is input into submodel 1, and the intermediate result 1 corresponding to gene 1 to be detected can be output; the Ct value of gene 2 to be detected is input into submodel 2, and the intermediate result 2 corresponding to gene 2 to be detected can be output; ...; the Ct value of gene N to be detected is input into submodel N, and the intermediate result N corresponding to gene N to be detected can be output, etc.

[0115] The intermediate result refers to the liver cancer risk value predicted based on the Ct value of a single gene to be detected, which can reflect the possibility of liver cancer. For example, the larger the value, the greater the possibility of the subject suffering from liver cancer.

[0116] In some embodiments, the input of the multiple sub-models also includes a historical Ct value and a historical interval duration corresponding to at least one gene to be detected.

[0117] The historical Ct value refers to the Ct value of the liver cancer test result of the subject at a historical time point. The historical interval duration refers to the duration between the test time corresponding to the historical Ct value and the current test time of the subject. In some embodiments, the historical Ct value and the historical interval duration can be obtained based on the historical test records of the subject.

[0118] In some cases, the subject needs to be re-examined or examined regularly, during which the Ct value may change dynamically, which can represent the change in the subject's physical condition.

[0119] In some embodiments of the present specification, by using historical Ct values ​​and historical interval durations as inputs to the sub-model, it is helpful to combine the historical conditions of the subject, thereby better judging the comprehensive evaluation value and obtaining a more accurate judgment result.

[0120] In some embodiments, at least one sub-model may be obtained through separate training.

[0121] In some embodiments, when training at least one sub-model separately, the first sample can be input into the initial sub-model, a loss function can be constructed based on the output of the initial sub-model and the first label, and the parameters of the initial sub-model can be iteratively updated based on the loss function until the preset conditions are met, and the training ends to obtain the trained sub-model. The preset conditions can include but are not limited to the convergence of the loss function and the reaching of the threshold value of the training cycle.

[0122] In some embodiments, the first sample may include multiple groups of sub-model training samples, and each group of sub-model training samples can be used for training the sub-models corresponding to the respective genes to be detected. Each group of sub-model training samples includes at least one sample Ct value of a sample gene, and a sample risk value corresponding to the sample Ct value, and the sample risk value refers to the liver cancer risk value predicted based on the aforementioned sample gene.

[0123] In some embodiments, at least one sample Ct value of a sample gene can be obtained based on historical test data, and the sample risk value can be determined based on the subsequent conditions of the historical subjects corresponding to the aforementioned sample gene. For example, the evaluation module can obtain test samples at multiple historical time points and screen them, retaining the test samples with regular inspection or review records as target test samples.

[0124] In some embodiments, the corresponding label can be determined based on the subsequent situation of the subject corresponding to the target test sample, wherein the subsequent situation may include but is not limited to whether the subject is diagnosed, the diagnosis interval of the subject, the liver cancer grade of liver cancer, and determine the risk value of the sample.

[0125] Exemplarily, when the subject has a follow-up examination but is not diagnosed, the sample risk value is determined to be 0. When the subject has a follow-up examination and is diagnosed, the sample risk value is determined according to the liver cancer grade and the confirmation interval. The higher the liver cancer grade in the follow-up examination and the shorter the confirmation interval, the higher the risk and the greater the sample risk value.

[0126] In some embodiments, the sample risk value can be calculated using formula (1):

[0127]

[0128] in, is the sample risk value; and is the preset coefficient; is the grade of liver cancer; B Confirm the interval length. Liver cancer grade It refers to the clinical classification of liver cancer based on its severity. Liver cancer patients are divided into stage I liver cancer patients, stage II liver cancer patients, stage III liver cancer patients, and stage IV liver cancer patients. The corresponding liver cancer grades are 1, 2, 3, and 4. The larger the liver cancer grade number, the more serious the liver cancer. Confirmation interval time B refers to the interval time between the detection time and the confirmed time corresponding to the target detection sample.

[0129] like Figure 4 As shown, in some embodiments, the input of the fusion layer 470 may include the intermediate results 460 output by the sub-model 450 , for example, intermediate result 1, intermediate result 2, ..., intermediate result N; and the output may include a comprehensive evaluation value 490 .

[0130] In some embodiments, the fusion layer can be obtained by joint training with the trained sub-model.

[0131] In some embodiments, the second sample of the joint training may be the same as the first sample, including a sample Ct value corresponding to at least one sample gene, and the second label of the joint training may be a sample risk value corresponding to the sample Ct value. Exemplarily, when there are 4 genes to be detected, the second sample of the joint training includes training samples of each sub-model. The method for obtaining the second sample is the same as that of the first sample, and please refer to the relevant description above. The second label can be determined based on whether the subject corresponding to the sample data suffers from liver cancer. If suffering from liver cancer, it is marked as 1, and if not suffering from liver cancer, it is marked as 0.

[0132] In some embodiments, the second sample can be input into the trained sub-model to obtain the sample intermediate result output by the sub-model; the sample intermediate result is input into the initial fusion layer, a loss function is constructed based on the second label and the output of the sample fusion layer, and the parameters of the initial fusion layer are iteratively updated based on the loss function until the preset conditions are met, the training ends, and the trained fusion layer is obtained. The preset conditions may include but are not limited to the convergence of the loss function and the reaching of the threshold value of the training cycle.

[0133] In some embodiments of the present specification, the fusion layer is jointly trained with the trained sub-model. In this process, the training of the sub-model can be used as pre-training of the risk assessment model, thereby accelerating the convergence speed of the risk assessment model.

[0134] In some embodiments, the input to the fusion layer 470 may also include a preliminary screening result 484 .

[0135] The initial screening results refer to the data obtained by screening based on the methylation sequencing results of the whole genome, which includes liver cancer-related gene sites and the degree of liver cancer relevance.

[0136] In some embodiments, the processor may determine the preliminary screening result 484 based on the methylation sequencing result 410 through the preliminary screening model 480. The methylation sequencing result may be obtained by processing the sample processing module. For more information on determining the methylation sequencing result, see Figure 2 Related description in .

[0137] In some embodiments, the initial screening model 480 includes a gene determination layer 481 and a result determination layer 483 , wherein the gene determination layer 481 is used to determine relevant genes and the degree of correlation, and the result determination layer 483 is used to determine the initial screening result 484 .

[0138] In some embodiments, the input of the gene determination layer 481 is the methylation sequencing result 410 , and the output is the genes related to liver cancer and the degree of correlation 482 .

[0139] Liver cancer-related genes refer to genes that are associated with liver cancer after methylation sequencing and are determined by the gene determination layer. The degree of correlation can reflect the degree of association between a gene and liver cancer, and the degree of correlation can be expressed by the Pearson correlation coefficient. For example, -1 indicates a complete negative correlation; 1 indicates a complete positive correlation.

[0140] In some embodiments, the input of the result determination layer may be the output of the gene determination layer, and the output of the result determination layer may be the initial screening result of liver cancer.

[0141] In some embodiments, the preliminary screening model can be obtained by joint training of the gene determination layer and the result determination layer.

[0142] In some embodiments, the third sample for joint training of the gene determination layer and the result determination layer may include multiple sample methylation sequencing data, and the corresponding third label may include the liver cancer grade corresponding to the sample.

[0143] In some embodiments, the third sample may include methylation sequencing data samples of healthy people, patients with liver cancer stage I, patients with liver cancer stage II, patients with liver cancer stage III, and patients with liver cancer stage IV. The third sample can be obtained based on liver cancer detection cases at historical time points, and the third label can be obtained based on the re-examination results of liver cancer detection cases. The acquisition of the third sample is similar to the acquisition steps of the first sample, and the relevant content can be referred to above.

[0144] The specific process of joint training is similar to the training process of the risk assessment model, which can be found in the relevant content above.

[0145] In some embodiments of the present specification, by dividing the initial screening model into a gene determination layer and a result determination layer, the gene determination layer can be used to screen from a large number of methylation gene sites to exclude related sites that are not related to liver cancer; the gene determination layer can be used to obtain the required initial screening results. Compared with using a single model for prediction, more accurate results can be obtained.

[0146] In some embodiments, the fusion layer input further includes related genes and correlation levels 482. The related genes and correlation levels include at least one related gene related to liver cancer and its corresponding correlation level.

[0147] In some embodiments, when the input of the fusion layer also includes the initial screening result 484 and the related genes and the correlation degree 482, the second sample also includes the sample initial screening result and the sample related genes and the correlation degree. When the sub-model and the initial fusion layer are jointly trained, the output of the sub-model, the sample initial screening result and the correlation degree of the sample related genes need to be used as the input of the initial fusion layer. The specific training process can be found above.

[0148] In some embodiments of this specification, due to the complexity of genes and the diversity of regulatory processes, there are usually both genes with a high degree of correlation with liver cancer and genes with a low degree of correlation. If only individual genes with a high degree of correlation are considered, the potential impact of other genes with a low degree of correlation on liver cancer will often be ignored. By using the initial screening model to analyze and screen the methylation sequencing results of the whole genome tissue, potential related genes can be effectively mined and used as input to the risk assessment model to improve the comprehensiveness and accuracy of the model prediction.

[0149] In some embodiments, the risk assessment model further includes a feature extraction layer 420, which is configured to determine a methylation feature distribution 430 of the gene to be detected based on the methylation sequencing result 410. For more information on the methylation sequencing result, please refer to the above description.

[0150] In some embodiments, the input of the feature extraction layer may include methylation sequencing results 410; the output may include methylation feature distribution 430 of at least one gene to be detected, for example, methylation feature distribution of gene 1 to be detected, methylation feature distribution of gene 2 to be detected, ..., methylation feature distribution of gene N to be detected, etc.

[0151] The methylation feature distribution can reflect the methylation level of the base site on each gene and the proportion of the methylated region of the entire gene. In some embodiments, the methylation level of the base site on each gene can be represented by a sequence, and the order of the sequence is the same as the order of the bases on the gene region.

[0152] In some embodiments, the feature extraction layer can be obtained through separate training.

[0153] In some embodiments, the third sample for training the feature extraction layer may include at least one sample methylation sequencing result, and the third label includes the sample methylation feature distribution corresponding to the sample gene. The training process includes inputting the third sample into the initial feature extraction layer, constructing a loss function based on the output of the initial feature extraction layer and the third label, iteratively updating the parameters of the initial feature extraction layer based on the loss function, until the preset conditions are met, the training ends, and the trained feature extraction layer is obtained. The preset conditions may include but are not limited to the convergence of the loss function, the training cycle reaching a threshold, etc.

[0154] The training labels of the feature extraction layer can be obtained by analyzing and processing the sample methylation sequencing results. For example, the sample methylation sequencing results can be analyzed for differentially methylated regions (DMRs), and the obtained methylation feature distribution results can be determined as the required training labels.

[0155] In some embodiments, the output of the feature extraction layer 420 can be used as the input of the sub-model 450. For example, the methylation feature distribution of the gene to be detected 1 can be used together with the Ct value of the gene to be detected 1 as the input of the sub-model 1, the methylation feature distribution of the gene to be detected 2 can be used together with the Ct value of the gene to be detected 2 as the input of the sub-model 2, ..., the methylation feature distribution of the gene to be detected N can be used together with the Ct value of the gene to be detected N as the input of the sub-model N.

[0156] In some embodiments, when the input of the sub-model also includes the methylation feature distribution 430 of the gene to be detected, the first sample trained by the sub-model also includes the methylation feature distribution of the sample gene.

[0157] In some embodiments of the present specification, methylation features are extracted through a feature extraction layer and input into corresponding subsequent sub-models, so that the potential reflection of methylation features on liver cancer risk can be utilized to improve the accuracy of liver cancer comprehensive index prediction.

[0158] In some embodiments of this specification, by using different sub-models to analyze and process data for specific genes to be detected, the specificity of the model can be improved, and the liver cancer risk factors contained in a single gene can be fully explored. The outputs of multiple sub-models are fused through the fusion layer to obtain the final required liver cancer comprehensive index, which can fully consider the different weights of different genes for liver cancer and the impact on the prediction of the liver cancer comprehensive index. It helps to improve the accuracy and reliability of the prediction results.

[0159] Figure 5 It is a schematic diagram of determining the periodic review interval duration according to some embodiments of this specification.

[0160] In some embodiments, in response to the comprehensive judgment result 510 being high risk, the risk warning includes prompting the subject to undergo a special examination 520 ; in response to the comprehensive judgment result 510 being low risk, the risk warning includes prompting the subject to undergo a regular review 530 .

[0161] Risk warning refers to the prediction and warning of the liver cancer risk of the subject by the interactive module based on the comprehensive judgment result. In some embodiments, the risk warning may include the length of the regular review interval.

[0162] In some embodiments, the interaction module can determine that the comprehensive judgment result 510 of the subject is low risk or high risk based on a variety of methods. In some embodiments, the processor can judge the liver cancer risk of the subject based on the comprehensive evaluation value and the preset threshold. In response to the comprehensive evaluation value not exceeding the preset threshold, the liver cancer risk of the subject is judged to be low risk. Exemplarily, the liver cancer comprehensive index of the subject is 0.43, which is less than the risk threshold of 0.72, and the subject is determined to be low risk. Among them, the preset threshold can be preset based on prior experience.

[0163] In some embodiments, in response to the comprehensive judgment result of liver cancer being low risk, the interactive module may prompt the subject to undergo regular review, including: in response to the comprehensive judgment result of liver cancer being low risk, based on a preset threshold and a comprehensive liver cancer index, determining a regular review interval 540.

[0164] The regular review interval refers to the interval between the time when the subject obtains the comprehensive liver cancer diagnosis result and the recommended next review time.

[0165] In some embodiments, the interaction module can determine the periodic review interval based on a variety of methods. In some embodiments, the interaction module can calculate the difference based on a preset threshold and the liver cancer comprehensive index, and determine the periodic review interval based on the positive correlation between the periodic review interval and the absolute value of the difference. For example, the larger the absolute value of the difference, the longer the periodic review interval.

[0166] In some embodiments, the interactive module can also establish a first preset table based on historical preset thresholds, historical liver cancer comprehensive indexes, and historical regular review intervals. The first preset table includes the correspondence between historical preset thresholds, historical liver cancer comprehensive indexes, and different historical regular review intervals. The interactive module can determine the regular review interval according to the current preset threshold and the historical liver cancer comprehensive index through the first preset table.

[0167] In some embodiments of the present specification, the regular review interval is determined based on the difference between a preset threshold and a comprehensive liver cancer index. When the subject's comprehensive liver cancer index is far from the threshold, the subject does not need to undergo frequent liver cancer examinations. When the subject's comprehensive liver cancer index is close to the threshold, although the subject is judged to be at low risk, there is a risk of turning into a high risk. A more frequent regular review interval can be set to ensure that the subject's liver will not become cancerous due to a long period of non-review.

[0168] In some embodiments, the interactive module can also determine the regular review interval based on the comprehensive evaluation value, at least one historical comprehensive evaluation value, the Ct value of the gene to be detected this time, at least one group of historical Ct values ​​of the genes to be detected, and the interval between the detection time corresponding to at least one historical comprehensive evaluation value and the current detection time, and at least one historical comprehensive evaluation value corresponds to at least one group of historical Ct values ​​of multiple genes to be detected.

[0169] In the case where the subject has undergone historical testing, the historical data of the subject's liver cancer monitoring can be used as an influencing factor for the length of the reexamination interval. In some embodiments, different subjects have different numbers of historical tests, and the amount of historical data to be obtained can be determined based on the number of historical tests the subject has undergone.

[0170] For example, if the subject has only been examined once, the test results of this time are obtained as a set of data; if the subject has been examined twice, the test results of these two times are obtained as two sets of data; if the subject has been examined no less than three times, the test data of the last three times can be obtained for subsequent analysis.

[0171] The data of liver cancer monitoring can be represented by a second feature vector. In some embodiments, the second feature vector is constructed based on the comprehensive evaluation value of each subject's current test, the historical comprehensive evaluation value, the Ct value of the gene to be tested this time, the historical Ct value of the gene to be tested, and the interval length.

[0172] In some embodiments, when determining sample data based on historical data, the historical data can be screened, and only the historical data in which the subject has a reexamination record and the condition is improved, controlled, or not significantly deteriorated during the reexamination time is retained as the sample data. For example, after the first liver cancer test of subject A, the doctor requires a reexamination two months later based on the Ct value of the gene to be tested. The reexamination results in a significant deterioration of the condition, which means that the two-month reexamination interval is unreasonable, so this set of data will not be used.

[0173] Taking the acquisition of one historical test data as an example, when the subject has one historical test data, the current liver cancer comprehensive index is A, the previous liver cancer comprehensive index is B, the current Ct value of gene 1 is C, ..., the current Ct value of gene N is D, the previous Ct value of gene 1 is E, ..., the previous Ct value of gene N is F, and the historical interval length is G, then the corresponding second eigenvector is [(A, B), (C, ..., D), (E, ..., F), G].

[0174] The interaction module can cluster the second feature vector in the sample data by a clustering algorithm. In some embodiments, the interaction module can classify the second feature vector according to the historical number of liver cancer monitoring of the subject, and perform separate clustering according to the classification. After clustering, one or more cluster centers are obtained. And the reference reexamination interval duration corresponding to the cluster center is determined based on the one or more cluster centers.

[0175] In some embodiments, all second eigenvectors in each cluster can be sorted according to the reexamination results of the corresponding subjects, and the reexamination interval corresponding to the second eigenvector with the most ideal reexamination effect can be determined as the required reference reexamination interval. The most ideal reexamination effect means that the condition of the subject has improved.

[0176] When it is necessary to determine the regular review interval based on the liver cancer test of the subject, the corresponding second eigenvector is constructed based on the comprehensive evaluation value of the subject, at least one historical comprehensive evaluation value, the Ct value of the gene to be tested this time, at least one group of historical Ct values ​​of the genes to be tested, and the interval between the detection time and the current detection time corresponding to at least one historical comprehensive evaluation value, and the vector distance with each cluster center is calculated to determine the cluster to which it belongs, and the corresponding reference review interval is used as the next regular review interval of the subject.

[0177] There may be many types of clustering algorithms, for example, clustering algorithms may include K-Means clustering, density-based clustering method (DBSCAN), etc. There is no limitation on the types of clustering algorithms.

[0178] In some embodiments of the present specification, the review interval is determined based on the historical test data of the subject, which can take into account the impact of different changes in liver cancer risks due to different physiological factors of the subjects, and can improve the applicability of the regular review interval for different subjects.

[0179] In some embodiments of the present specification, the subject is judged to be at low risk or high risk based on the comprehensive judgment result of liver cancer, and a risk warning is issued. When the risk is high, the subject is immediately reminded to undergo special examinations to avoid delays in the disease; when the risk is low, the subject is reminded to undergo regular checkups to prevent the potential risk of liver cancer.

[0180] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0181] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0182] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0183] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0184] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0185] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification is hereby incorporated by reference in its entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0186] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A liver cancer detection system, characterized in that: It includes sample processing module, evaluation module and interaction module; The sample processing module is configured to determine the Ct value of the gene to be detected in the target sample corresponding to the subject; The evaluation module is configured to determine a comprehensive evaluation value based on the Ct value, and determine a comprehensive judgment result based on the comprehensive evaluation value; The interactive module is configured to display the comprehensive judgment result and issue a risk warning based on the comprehensive judgment result; The sample processing module is further configured to: Based on the methylation sequencing results, determining at least one gene to be evaluated; Based on the at least one gene to be evaluated, the gene to be detected is determined; wherein, The genes to be detected include NKX6-2, GNB4, HOXA1 and TSPYL5.

2. The liver cancer detection system according to claim 1, characterized in that: The evaluation module is further configured to: Based on the Ct value of the gene to be detected, the comprehensive evaluation value is determined through a risk assessment model.

3. The liver cancer detection system according to claim 1, characterized in that: The evaluation module is further configured to: Based on the comprehensive evaluation value and a preset threshold, the comprehensive judgment result is determined.

4. The liver cancer detection system according to claim 1, characterized in that: The interaction module is further configured to: In response to the comprehensive judgment result being a high risk, the risk warning includes prompting the subject to undergo a special examination; In response to the comprehensive judgment result being low risk, the risk warning includes prompting the subject to undergo regular review.

5. A liver cancer detection kit, characterized in that: The detection kit is applied to the liver cancer detection system as described in any one of claims 1 to 4, comprising a first detection reaction solution, a second detection reaction solution, an enzyme system, a negative control and a positive control; The first detection reaction solution includes a HOXA1 forward primer, a HOXA1 reverse primer, a HOXA1 probe, a NKX6-2 forward primer, a NKX6-2 reverse primer, a NKX6-2 probe, an ACTB forward primer, an ACTB reverse primer, an ACTB probe, a dNTP mixed solution and a Buffer buffer; The second detection reaction solution includes a GNB4 forward primer, a GNB4 reverse primer, a GNB4 probe, a TSPYL5 forward primer, a TSPYL5 reverse primer, a TSPYL5 probe, an ACTB forward primer, an ACTB reverse primer, an ACTB probe, the dNTP mixed solution and the Buffer buffer.

6. The liver cancer detection kit according to claim 5, characterized in that The enzyme system includes at least one of HiTaq DNA Polymerase and HiTaq DNA Polymerase diluent.

7. The liver cancer detection kit according to claim 5, characterized in that The negative control included non-hepatoma cell line DNA.

8. The liver cancer detection kit according to claim 5, characterized in that The positive control included liver cancer cell line DNA.

9. Application of a liver cancer detection kit, characterized in that: The liver cancer detection kit is configured in a sample processing module of the liver cancer detection system according to any one of claims 1 to 4, and is used to assist in determining the Ct value of the gene to be detected in the target sample corresponding to the subject.

Images