Artificial intelligence-based information provision method for lung cancer diagnosis using lung cancer diagnostic biomarkers and clinical information of test subjects

Abstract

The present invention relates to an artificial intelligence-based information provision method for lung cancer diagnosis using lung cancer diagnostic biomarkers and clinical information of a test subject. More specifically, the present invention relates to a metabolite biomarker composition for lung cancer diagnosis, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), and an artificial intelligence-based information provision method for lung cancer diagnosis using the metabolite biomarkers and clinical information of a test subject. Using the lung cancer diagnostic biomarkers and clinical information selected in the present invention, an artificial intelligence-based algorithm model for lung cancer diagnosis was established. As a result, it was confirmed that the screening ability for early lung cancer was 90-92% in sensitivity, 93-95% in specificity, and 92-93% in accuracy, which is significantly higher than existing lung cancer screening methods. Therefore, the present invention can effectively provide information regarding lung cancer diagnosis.

Description

Detailed Description of the Invention

[0001] [Technical field] The present invention relates to an artificial intelligence-based information provision method for lung cancer diagnosis using lung cancer diagnostic biomarkers and clinical information of a test subject. More specifically, the present invention relates to a metabolite biomarker composition for lung cancer diagnosis, comprising kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), and an artificial intelligence-based information provision method for lung cancer diagnosis using the metabolite biomarkers and clinical information of a test subject.

[0002] [Background technology] Cancer is a disease in which cells grow indefinitely and disrupt the function of normal cells. Representative cancers include liver cancer, lung cancer, stomach cancer, breast cancer, colon cancer, and ovarian cancer, but it can occur in virtually any tissue.

[0003] Early cancer diagnosis was based on external changes in biological tissues caused by the growth of cancer cells, but in recent years, attempts have been made to diagnose cancer by detecting trace amounts of biomolecules present in biological tissues or cells, such as blood, glycochains, and DNA.However, the most commonly used cancer diagnosis method is diagnosis using tissue samples obtained through biopsies or imaging.

[0004] Among these, biopsies have the disadvantages of causing great pain to patients, being expensive, and taking a long time to diagnose. Furthermore, if a patient does develop cancer, there is a risk that the cancer may metastasize during the biopsy process. Furthermore, if a tissue sample cannot be obtained through a biopsy, a diagnosis of the disease cannot be made until the suspected tissue is surgically removed.

[0005] Lung cancer, in particular, is one of the cancers with the highest mortality rates worldwide, and currently, lung cancer diagnosis relies heavily on imaging methods (e.g., X-rays, CT scans, and MRIs). However, more than half of lung cancer patients are already inoperable at the time of diagnosis. Even in those who are deemed operable, complete resection is often not possible. Therefore, early diagnosis and treatment of lung cancer are crucial to increasing the lung cancer cure rate. However, the limited number of diagnostic markers for lung cancer makes such diagnosis difficult. Therefore, it is necessary to identify cancer-specific markers present in biological samples and develop methods that utilize these markers to diagnose cancer with high accuracy and precision.

[0006] Recently, methods using artificial intelligence (AI) have been researched for more accurate cancer diagnosis. AI-based analysis mainly uses machine learning or deep learning, and there is a trend of various researches using AI in the bio field (Korean Patent Publication No. 10-2014-0002149, Korean Patent Registration No. 10-2268963).

[0007] [Summary of the Invention] [Problem to be solved by the invention] Therefore, the present inventors have made extensive efforts to screen for markers that can more accurately diagnose lung cancer. As a result, they selected biomarkers including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), and confirmed that the expression levels of these biomarkers showed different patterns between patient and control groups. Furthermore, the quantitative values ​​of these seven biomarkers and the clinical information of the patient and control groups were analyzed using an artificial intelligence-based algorithm, and it was confirmed that the diagnostic ability for lung cancer was improved, thereby completing the present invention.

[0008] Therefore, an object of the present invention is to provide a biomarker composition for diagnosing lung cancer. Another object of the present invention is to provide a lung cancer diagnostic composition containing a substance for measuring the level of the lung cancer diagnostic biomarker, and a lung cancer diagnostic kit using the same.

[0009] Another object of the present invention is to provide a method for providing artificial intelligence-based information for diagnosing lung cancer by utilizing the above-mentioned lung cancer diagnostic biomarkers and clinical information of test subjects.

[0010] [Means for solving the problem] To achieve the above-mentioned object, the present invention provides a biomarker composition for diagnosing lung cancer, comprising kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC).

[0011] In a preferred embodiment of the present invention, the biomarkers can be extracted from blood. In another preferred embodiment of the present invention, the blood may be whole blood, plasma or serum.

[0012] To achieve another object, the present invention provides a composition for diagnosing lung cancer, which comprises a preparation for measuring the blood concentration of the above-mentioned biomarker composition for diagnosing lung cancer. In a preferred embodiment of the present invention, the preparation for measuring the blood concentration of the lung cancer diagnostic biomarker composition may be a preparation for mass spectrometry.

[0013] The present invention also provides a kit for diagnosing lung cancer, which comprises a preparation for measuring the blood concentration of the above-mentioned biomarker composition for diagnosing lung cancer. In order to achieve another object, the present invention provides a method for measuring the levels of lung cancer diagnostic biomarkers, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), from the blood of a subject; and (b) A method for providing information for diagnosing lung cancer using artificial intelligence, which includes applying the levels of the lung cancer diagnostic biomarkers and clinical information of the test subject to a machine learning algorithm model.

[0014] In a preferred embodiment of the present invention, the blood in step (a) above may be whole blood, plasma or serum. In another preferred embodiment of the present invention, the concentration of the biomarker in step (a) can be obtained by mass spectrometry of a whole blood, plasma, or serum sample based on the mass peak area. Specifically, the concentration can be obtained via a liquid chromatography-mass spectrometer (LC-MS). The mass spectrometer may be any of a triple TOF, triple quadrupole, or MALDI TOF capable of quantitative measurement.

[0015] In another preferred embodiment of the present invention, in the step (b) of applying to the algorithm model, the biomarker level in the blood of the test subject and clinical information of the test subject can be input into the algorithm model, and the presence or absence of lung cancer can be output as an output value.

[0016] In another preferred embodiment of the present invention, the clinical information in step (b) above may be any one or more selected from the group consisting of age, BMI, and smoking history. In another preferred embodiment of the present invention, the algorithm model in step (b) is (1) measuring the levels of lung cancer diagnostic biomarkers, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), from the blood of lung cancer patients and control groups; and (2) A step of generating a lung cancer onset prediction model by learning the levels of the lung cancer diagnostic biomarkers and clinical information of a lung cancer patient group and a control group using a machine learning algorithm.

[0017] In another preferred embodiment of the present invention, the artificial intelligence is machine learning or deep learning, and more specifically, the algorithm in step (b) is any one selected from a linear or nonlinear classification algorithm including a K-nearest neighbor algorithm; a logistic regression algorithm; a discriminant analysis algorithm; a partial least squares-discriminant analysis algorithm; a support vector machine algorithm; a decision tree algorithm; a decision tree ensemble algorithm; and a neural network algorithm.

[0018] In another preferred embodiment of the present invention, when the algorithm is a support vector machine algorithm, the kernel function can be expressed as the following Equation 1:

[0019]

number

[0020] The present invention also provides an AI-based lung cancer diagnosis and prediction device, which includes a measurement unit that measures the level of a lung cancer diagnostic biomarker in the blood of a test subject; and a cancer diagnosis unit that inputs the biomarker level and clinical information of the test subject into a trained AI algorithm to determine whether or not lung cancer has developed.

[0021] [Effects of the invention] In the present invention, seven biomarkers capable of more accurately diagnosing lung cancer were selected, and an AI-based algorithm for lung cancer diagnosis was established using the above biomarkers for lung cancer diagnosis and clinical information from lung cancer patients and control groups. The ability to screen for early-stage lung cancer using the algorithm developed in the present invention was confirmed to be significantly higher in accuracy than existing lung cancer screening methods, with a sensitivity of 90-92%, a specificity of 93-95%, and an accuracy of 92-93%, making the present invention capable of effectively providing information related to lung cancer diagnosis. DETAILED DESCRIPTION OF THE INVENTION

[0022] The present invention will be described in detail below. <Biomarker composition for diagnosing lung cancer> The present invention relates to a biomarker composition for diagnosing lung cancer, comprising kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC).

[0023] In the present invention, the biomarker is extracted from blood, which may be whole blood, plasma or serum. The term "diagnosis" as used herein means to confirm the presence or characteristics of a pathological condition. For purposes of the present invention, diagnosis refers to confirming the presence or absence of lung cancer.

[0024] The term "diagnostic biomarker" as used in the present invention refers to an organic biomolecule such as a polypeptide, nucleic acid (e.g., mRNA), lipid, glycolipid, glycoprotein, or sugar (monosaccharide, disaccharide, oligosaccharide, etc.) that shows a significant increase or decrease in a lung cancer patient group compared to a normal control group, and is preferably the above-mentioned lung cancer diagnostic biomarker composition.

[0025] In one specific example of the present invention, blood samples were collected from a control group (Con) and a lung cancer (LC) patient group, and the concentrations of KN, HC, OC, DC, DDC, MC, and PC in the blood were measured. It was confirmed that the quantitative values ​​of the blood concentrations of the biomarkers of the present invention in the lung cancer patient group and the control group were significantly different, and it was confirmed that the concentrations of KN metabolites in the blood of the lung cancer patient group were increased, and the concentrations of HC, OC, DC, DDC, MC, and PC metabolites were decreased, compared to the control group (Tables 1 and 2).

[0026] <Composition for diagnosing lung cancer> From another aspect, the present invention relates to a composition for diagnosing lung cancer, which comprises a preparation for measuring the blood concentration of the biomarker composition for diagnosing lung cancer of the present invention.

[0027] The lung cancer diagnostic composition according to the present invention is the same as the above <Biomarker composition for diagnosing lung cancer>. The biomarkers of the present invention are metabolites, and the formulation for measuring the blood concentration of the lung cancer diagnostic biomarker composition may be a formulation for mass spectrometry.

[0028] The above-mentioned preparation for mass spectrometry is a preparation that can analyze the mass of a marker in whole blood, plasma, or serum, and specifically means a preparation that can be used for liquid chromatography-mass spectrometry (LC-MS).

[0029] <Lung cancer diagnostic kit> In another aspect, the present invention relates to a kit for diagnosing lung cancer, which comprises a preparation for measuring the blood concentration of the lung cancer diagnostic biomarker composition of the present invention.

[0030] The lung cancer diagnostic composition according to the present invention is the same as the above <Biomarker composition for diagnosing lung cancer>. The kit can be prepared by a conventional method known in the art and may contain, for example, a lyophilized antibody, a buffer, a stabilizer, an inactive protein, etc.

[0031] <Method for providing information on lung cancer diagnosis using an artificial intelligence-based algorithm> In another aspect, the present invention provides a method for treating lung cancer comprising the steps of: (a) measuring the levels of lung cancer diagnostic biomarkers, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), from the blood of a subject; and (b) A method for providing information for diagnosing lung cancer using artificial intelligence, which includes applying the levels of the lung cancer diagnostic biomarkers and clinical information of the test subject to a machine learning algorithm model.

[0032] In the present invention, the blood in step (a) above may be whole blood, plasma or serum. In the present invention, the concentration of the biomarker in step (a) can be obtained by mass spectrometry of a whole blood, plasma, or serum sample based on the mass peak area. Specifically, it is obtained using a liquid chromatography-mass spectrometer (LC-MS). The mass spectrometer may be any of a Triple TOF, a Triple Quadrupole, or a MALDI TOF capable of quantitative measurement.

[0033] In the present invention, the step of applying to the algorithm model in step (b) can input the biomarker level in the blood of the test subject and the clinical information of the test subject into the algorithm model, and output the presence or absence of lung cancer as an output value.

[0034] In the present invention, the clinical information in step (b) may be any one or more selected from the group consisting of age, BMI, and smoking history. In the present invention, the algorithm model in the above step (b) is (1) measuring the levels of lung cancer diagnostic biomarkers, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), from the blood of lung cancer patients and control groups; and (2) A step of generating a lung cancer onset prediction model by learning the levels of the lung cancer diagnostic biomarkers and clinical information of a lung cancer patient group and a control group using a machine learning algorithm.

[0035] In the present invention, the artificial intelligence may be machine learning or deep learning, and more specifically, the algorithm in step (b) is any one selected from a linear or nonlinear classification algorithm including a k-nearest neighbor algorithm; a logistic regression algorithm; a discriminant analysis algorithm; a partial least squares discriminant analysis algorithm; a support vector machine algorithm; a decision tree algorithm; a decision tree ensemble algorithm; and a neural network algorithm.

[0036] In the present invention, when the above algorithm is a support vector machine algorithm, the kernel function can be expressed by the following Equation 1:

[0037]

number

[0038] In one specific example of the present invention, a prediction model was developed using quantitative values ​​of seven biomarkers of the present invention measured in the blood of lung cancer patients and control groups and clinical information of the lung cancer patients and control groups, and the algorithm used was a support vector machine with a radial basis function as the kernel.The diagnostic ability of the developed prediction model for early-stage lung cancer was confirmed, and the model showed a sensitivity of 90-92%, a specificity of 93-95%, and an accuracy of 92-93%.

[0039] That is, it was confirmed that the AI-based lung cancer diagnosis method using the seven biomarkers of the present invention and clinical information of the test subjects has a much higher accuracy than other existing diagnostic methods.

[0040] In another aspect, the present invention provides a measuring unit for measuring the level of a lung cancer diagnostic biomarker, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), in the blood of a subject; and The present invention provides an artificial intelligence-based lung cancer diagnosis and prediction device, which includes a cancer diagnosis unit that inputs the lung cancer diagnostic biomarker level and clinical information of the test subject into a trained artificial intelligence algorithm to determine whether or not lung cancer has developed.

[0041] The present invention will be described in more detail below through examples. It will be obvious to those skilled in the art that these examples are merely for the purpose of illustrating the present invention, and that the scope of the present invention is not to be construed as being limited by these examples.

[0042] Example 1: Measurement of biomarker concentrations in the blood of lung cancer patients and control groups 1-1: Sample preparation To confirm whether the biomarkers of the present invention can diagnose lung cancer, a total of 152 individuals were selected, specifically 92 control individuals and 60 lung cancer (LC) patient individuals, through Ajou University Hospital, and their blood samples were collected. Clinical information regarding age, BMI, and smoking history was also collected from the lung cancer patient and control groups.

[0043] 1-2: Measurement of biomarker concentrations in blood Plasma was separated from the blood samples and measured for kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), and decanoyl-L-carnitine (DC). Standard concentrations required for the biomarker standard curves of dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC) were prepared.

[0044] 20 μl of plasma and standard samples were mixed with 400 μl of lipid extraction buffer (Abnova, Taiwan), vortexed, and then centrifuged (10,000 g, 5 min, 4°C). After centrifugation, the supernatant was transferred to a new tube and dried for 12–16 hours using a concentrator. 50 μl of 100% methanol containing 0.1% formic acid (FA) was added to the dried metabolite extract, which was then thoroughly dissolved using a vortexer. The mixture was then analyzed using LC-MS / MS (liquid chromatography-mass spectrometry / mass spectrometry).

[0045] The LC system used was a Shimadzu LC 40 system, and the MS system was an AB Sciex Triple Quad 5500+ system. The MS was equipped with a turbospray ion source. The analytical samples were separated on a BEH C18 column (1.7 μm, 2.1 × 50 mm; Waters) on the Shimadzu LC 40 system. The solvents used were a two-step linear gradient (solvent A, 0.1% FA in water; solvent B, 0.1% FA in 100% acetonitrile; 5–55% solvent B for 2.5 min, 55% solvent B for 5.5 min, 55–95% solvent B for 7.5 min, 95% solvent B for 11 min, 95–5% solvent B for 11.1 min, and 5% solvent B for 14.5 min).

[0046] Mass spectrometry (MS / MS) was performed using multiple reaction monitoring (MRM) mode. The area of ​​the mass peaks with the same mass values ​​in the mass spectrum at the same time as the metabolites corresponding to each biomarker emerged from the liquid chromatography was calculated. A standard curve was created using the mass peaks of each biomarker standard substance, and the mass peaks of each sample were substituted into the standard curve to measure the concentration of each biomarker.

[0047] [Table 1] JPEG2025534327000004.jpg229169JPEG2025534327000005.jpg227169JPEG20255343270 00006.jpg228169JPEG2025534327000007.jpg228169JPEG2025534327000008.jpg149169

[0048] [Table 2] JPEG2025534327000010.jpg227169JPEG2025534327000011.jpg227169JPEG2025534327000012.jpg162169

[0049] As a result, as shown in Tables 1 and 2, it was confirmed that there was a significant difference in the blood concentrations of seven biomarkers between the quantitative values ​​of the lung cancer patient group and the control group, and the concentrations of KN metabolites in the blood of the lung cancer patient group increased, while the concentrations of HC, OC, DC, DDC, MC, and PC metabolites decreased, compared to the control group.

[0050] Example 2: Development of an artificial intelligence-based algorithm model for lung cancer diagnosis In the present invention, a prediction model capable of diagnosing the presence or absence of lung cancer was developed by applying a support vector machine algorithm using a radial basis function as the kernel to the quantitative values ​​of seven biomarkers and clinical information (age and BMI) of lung cancer patients and control groups.

[0051] A lung cancer prediction model was trained by tuning the algorithm parameters using the kernel function expressed in the following equation 1:

[0052]

number

[0053] The parameter gamma (γ) in Equation 1 determines the extent of influence of a single training sample, and another parameter of the support vector machine, C, which is used independently of the kernel function, determines the tolerance for misclassification of a training sample. Since the values ​​of both parameters can lead to underfitting or overfitting of the learning model, the optimal parameters were selected through iterative cross-validation.

[0054] Table 3 shows the results of an example in which the presence or absence of lung cancer was determined using a support vector machine model trained on the quantitative values ​​of seven biomarkers measured for 629 lung cancer patient samples and 511 control group samples, along with the clinical information (age and smoking history) collected together.

[0055] [Table 3]

[0056] The diagnostic ability of early stage lung cancer was confirmed using the developed prediction model, and it was found to have a sensitivity of 90-92%, a specificity of 93-95%, and an accuracy of 92-93%, as shown in Table 3. In other words, it was confirmed that the AI-based lung cancer diagnostic method of the present invention using the seven biomarkers and clinical information has a much higher accuracy than other existing diagnostic methods.

[0057] The screening ability for early stage lung cancer using the algorithm developed in the present invention was confirmed to be extremely high in accuracy compared to existing lung cancer screening methods, with a sensitivity of 90-92%, a specificity of 93-95%, and an accuracy of 92-93%, so the present invention can be effectively applied to providing information for lung cancer diagnosis.

Claims

1. A biomarker composition for diagnosing lung cancer, comprising kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC).

2. The lung cancer diagnostic biomarker composition according to claim 1, wherein the lung cancer diagnostic biomarker is extracted from blood.

3. The biomarker composition for diagnosing lung cancer according to claim 2, wherein the blood is whole blood, plasma or serum.

4. A lung cancer diagnostic composition comprising a preparation for measuring the blood concentration of the lung cancer diagnostic biomarker composition according to any one of claims 1 to 3.

5. 5. The lung cancer diagnostic composition according to claim 4, wherein the preparation for measuring the blood concentration of the lung cancer diagnostic biomarker composition is a preparation for mass spectrometry.

6. A lung cancer diagnostic kit comprising a formulation for measuring the blood concentration of the lung cancer diagnostic biomarker composition according to any one of claims 1 to 3.

7. (a) measuring the levels of lung cancer diagnostic biomarkers, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), from the test subject's blood; and (b) A method for providing information for diagnosing lung cancer using artificial intelligence, comprising a step of applying the level of the lung cancer diagnostic biomarker and clinical information of the test subject to a machine learning algorithm model.

8. The method for providing information for lung cancer diagnosis using artificial intelligence according to claim 7, wherein the blood in step (a) is whole blood, plasma, or serum.

9. The method for providing information for lung cancer diagnosis using artificial intelligence according to claim 7, wherein in step (a), the biomarker level measurement is obtained via liquid chromatography-mass spectrometry (LC-MS).

10. The method for providing information for lung cancer diagnosis using artificial intelligence as described in claim 7, characterized in that the step (b) of applying to the algorithm model includes inputting the biomarker levels and clinical information in the blood of the test subject into the algorithm model and outputting the presence or absence of lung cancer as an output value.

11. The method for providing information for lung cancer diagnosis using artificial intelligence described in claim 7, characterized in that the clinical information in step (b) is any one or more selected from the group consisting of age, BMI, and smoking history.

12. The algorithm model in the above step (b) is as follows: (1) measuring the levels of lung cancer diagnostic biomarkers, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), from the blood of lung cancer patients and control groups; (2) A step of generating a lung cancer onset prediction model by learning the levels of the lung cancer diagnostic biomarkers and clinical information of a lung cancer patient group and a control group using a machine learning algorithm.

13. The method for providing information for lung cancer diagnosis using artificial intelligence according to claim 7, wherein the algorithm in step (b) is any one selected from a linear or nonlinear classification algorithm including a K-nearest neighbor algorithm; a logistic regression algorithm; a discriminant analysis algorithm; a partial least squares-discriminant analysis algorithm; a support vector machine algorithm; a decision tree algorithm; a decision tree ensemble algorithm; and a neural network algorithm.

14. 8. The method for providing information for lung cancer diagnosis using artificial intelligence according to claim 7, wherein when the algorithm in step (b) is a support vector machine algorithm, the kernel function is expressed by the following Equation 1: [Equation 1] In the above, χ is the measured blood level of the lung cancer diagnostic biomarker composition and the clinical information of the test subject, and γ is a parameter for the flexibility (curvature) of the decision boundary.

15. a measuring portion for measuring the level of a lung cancer diagnostic biomarker, including kynurenine (KN), hexanoyl-L-carnitine (HC), octanoyl-L-carnitine (OC), decanoyl-L-carnitine (DC), dodecanoyl-L-carnitine (DDC), myristoyl-L-carnitine (MC), and palmitoyl-L-carnitine (PC), in the blood of a subject; and and a cancer diagnosis unit that inputs the lung cancer diagnostic biomarker level and clinical information of the test subject into a trained artificial intelligence algorithm to determine whether or not lung cancer has developed.

Images