Method and system for converting hormone detection data

Abstract

The present application relates to a method for converting hormone data detected by different detection methods, which comprises: obtaining a first detection value for the hormone from a first platform; obtaining a second detection value for the hormone from a second platform; fitting the first detection value and the second detection value using Passing‑Bablok regression; judging whether there is a systematic difference between the first detection value and the second detection value based on the Passing‑Bablok regression fitting result; if there is no systematic difference between the first detection value and the second detection value, converting the first detection value or the second detection value based on the fitting formula of Passing‑Bablok regression; if there is a systematic difference between the first detection value and the second detection value, fitting the first detection value and the second detection value based on spline regression, and converting the first detection value or the second detection value based on the fitting formula of spline regression.

Description

Technical Field

[0001] The present invention relates to a method and system for transforming hormone detection data obtained by different detection methods, and in particular to a method and system for transforming sex hormone detection data. Background Art

[0002] Hormones are substances that are secreted directly into the blood by endocrine organs or tissues of humans and animals and have special effects on the body. Digestive organs and placenta and other tissues also secrete hormones, such as secretagogue, gastrin, chorionic gonadotropin, etc. The coordinated action of various hormones is necessary to maintain the metabolism and function of the body. In terms of chemical properties, some hormones are phenolic derivatives, such as adrenaline and thyroxine; some are polypeptides or proteins, such as pituitary hormone-releasing factor, pituitary hormone, insulin, glucagon, calcitonin, parathyroid hormone, etc.; some are steroid compounds, such as androgens, estrogens, adrenocortical hormones, etc. Many hormone preparations and synthetic products have important uses in medicine and animal husbandry.

[0003] One of the important categories of hormones is sex hormones. Sex hormones (chemical essence is lipid) refer to hormones synthesized by the gonads of animals, as well as tissues such as the placenta and the reticular zone of the adrenal cortex. They have the functions of promoting the maturation of sexual organs, the development of secondary sexual characteristics and maintaining sexual function. The ovaries of female animals mainly secrete two sex hormones - estrogen and progesterone, and the testicles of male animals mainly secrete androgens, mainly testosterone. The hormones secreted by the hypothalamus-pituitary-gonadal axis or the hypothalamus-pituitary-adrenal axis are generally divided into three categories according to their chemical structure. The first category is steroids, such as adrenal cortical hormones and steroidal sex hormones. The second category is amino acids, including thyroxine, adrenal medullary hormones, pineal hormones, etc. The structure of the third category of hormones is peptides and proteins, such as hypothalamic hormones, pituitary hormones, gastrointestinal hormones, calcitonin, etc. The third category of hormones are basically protein macromolecules. From the perspective of immunological methods, the third category of hormones has multiple antigenic markers, and can generally be detected by the double antibody sandwich method.

[0004] Anti-Mullerian hormone (AMH) contributes to the regression of the Mullerian ducts during sexual differentiation of male embryos and was first discovered by Jost in 1947. In 1999, Dulinger et al. used AMH knockout mice to find that AMH was responsible for inhibiting the recruitment of primordial follicles and that AMH belonged to the third class of hormones mentioned above. Since then, researchers have begun research and clinical applications of AMH in reproductive science and developed a kit for detecting AMH. During ovarian stimulation, AMH is an important marker for measuring ovarian reserve. AMH is also considered an ideal marker for determining the individualized ovarian stimulation regimen for the dose of human recombinant follicle-stimulating hormone (rFSH) adrenocorticotropic hormone in IVF treatment. As AMH has gradually been recognized as an important parameter in reproductive science. Therefore, accurate, efficient and robust AMH detection methods have become particularly critical.

[0005] At present, there are many detection methods for AMH in the relevant field, among which the new automatic anti-Mullerian hormone detection method is gradually replacing the original ELISA detection method. Currently on the market, the commercial automatic anti-Mullerian hormone detection methods include the Elecsys AMH method provided by Roche (hereinafter referred to as "Roche method" or "Roche method" or "Elecsys method"), the Access AMH method provided by Beckman Coulter (hereinafter referred to as "Beckman method" or "Beckman method" or "Access method"), the chemiluminescent AMH detection method provided by Kangrun (hereinafter referred to as "Kangrun chemiluminescent method" or "Kangrun chemiluminescent method") and the enzyme-linked immunosorbent assay AMH detection method provided by Kangrun (hereinafter referred to as Kangrun Anshlab method or Kangrun enzyme-linked immunosorbent assay Anshlab method).

[0006] Roche launched the Elecsys assay in 2014, which is based on the monoclonal antibodies F2B 12 / H and F2B 7 / A. Beckman subsequently launched the Access assay using the same monoclonal antibodies. Since Roche and Beckman used similar antibodies, the main difference between the two is the development of their assays. In addition, Kangrun, a Chinese company, has developed an automated AMH assay that uses a different monoclonal antibody, which is coded as 1B63D7.

[0007] Estrogen is a major female hormone. It promotes the maturation of female accessory organs and the appearance of secondary sexual characteristics, and maintains normal sexual desire and reproductive function, mainly estradiol (E2). It is inactivated in the liver, converted into estriol and estrone, and excreted in the urine after conjugation with glucuronic acid. During pregnancy, the placenta can secrete a large amount of estriol. Estradiol (E2) was first discovered by Edward Doisy 77 years ago and has been shown to play a key role in a variety of physiological processes. Therefore, the quantification of E2 has been widely used in clinical assessment of female or male conditions or diseases, such as fertility testing, ovarian stimulation monitoring during assisted reproductive technology, pregnancy, polycystic ovary syndrome (PCOS), hypogonadism, amenorrhea, E2-related tumors, etc. Methods for E2 quantification include immunoassay, gas chromatography-mass spectrometry (GC-MS), and liquid chromatography-tandem mass spectrometry (LC–MS / MS). Mass spectrometry methods have better specificity and sensitivity than immunoassays, but their application as routine clinical methods is limited due to cost and technical complexity. Automated immunoassay platforms that do not require complex sample extraction processes have become the first choice for most clinical laboratories due to their high throughput, reproducibility, and short turnaround time.

[0008] Most direct automated immunoassay platforms are optimized to measure E2 concentrations between 20 and 2000 pg / mL. Such a concentration range can meet almost all clinical purposes in obstetrics and gynecology and can meet the needs of assisted reproductive technology (ART) in hospitals. However, it is not easy to standardize E2 measurements over such a wide concentration range. It has been reported that the assay sensitivity of low-level E2 tests is affected. In hospital clinical laboratories, clinicians have found that there is a large heterogeneity between different automated device immunoassay platforms, especially at low E2 levels, such as <200 pg / ml. This may be caused by the lack of international calibrators for E2, different calibration point assignments, and systematic differences in detection. Variations in E2 levels, especially in the low-level range, will definitely cause great confusion for clinicians in interpreting the results. Summary of the invention

[0009] Given that several major sex hormones described in the above background technology, such as AMH and E2, all play an important role in clinical testing and diagnosis, there are different methods and platforms for the detection of these sex hormones, which will result in the inability to effectively and uniformly evaluate and analyze the test results from different sources.

[0010] Taking AMH as an example, in view of the different AMH detection methods currently existing in the market, and the inconsistent situation of the AMH values ​​detected by these methods. The patient's AMH result obtained by using a certain AMH detection method in a certain hospital cannot be adopted by the hospital using other AMH detection methods, and such a problem greatly increases the medical burden of the patient and the complexity of clinical practice, and hinders the clinical application of the universal relevant online assessment tool for AMH at present. Taking AMH as one of the most important predictive indicators, the inventor of the present application has established several online prediction tools before, including ovarian reserve assessment tool (http: / / 121.43.113.123:9999 / ), PCOS screening assessment tool (http: / / 121.43.113.123:8888 / ), egg cell retrieval prediction tool (http: / / 121.43.113.123:8002 / ) and the online tool (http: / / 121.43.113.123:8004) for guiding the initiation and adjustment of exogenous FSH dosage. Given that AMH is an increasingly important evaluation parameter in reproductive medicine, if we want to easily apply and uniformly calculate the AMH data of patients tested by different hospitals and testing institutions, this field urgently needs a calculation method and system that can convert or unify the AMH values ​​obtained by different testing methods.

[0011] Take estradiol as an example. Estradiol is a C18 steroid hormone and the most biologically active hormone among estrogens. It is produced by male testicles, female ovaries and pregnant placentas, and its levels vary with the menstrual cycle. The main physiological function of estradiol is to promote the generation and development of follicles, the development of ovaries and female sexual organs, the appearance of female secondary sexual characteristics, and the synthesis of various transport proteins in the liver. Circulating blood estradiol concentration is one of the indicators for examining the function of the hypothalamus-pituitary-reproductive target gland axis, and has certain diagnostic significance for diagnosing endocrine diseases and gynecological diseases such as precocious puberty, dysplasia, infertility, etc. In the clinical laboratories of hospitals, clinicians have found that there is a large heterogeneity between different automatic device immunoassay platforms, especially when the E2 level is low, such as <200pg / ml. This may be due to the lack of an international calibrator for E2 and the different allocation of calibration points. Changes in E2 levels, especially in the low-level range, will definitely cause great confusion for clinicians in interpreting the results.

[0012] Specifically, the technical solution of this application is as follows:

[0013] 1. A method for converting hormone data detected by different detection methods, comprising:

[0014] Obtaining a first detection value for the hormone from a first platform;

[0015] Obtaining a second detection value for the hormone from a second platform;

[0016] Passing-Bablok regression is used to fit the first detection value and the second detection value;

[0017] Determining whether there is a systematic difference between the first detection value and the second detection value based on the passing-Bablok regression fitting result;

[0018] If there is no systematic difference between the first detection value and the second detection value, the first detection value or the second detection value is converted based on the fitting formula of Passing-Bablok regression.

[0019] If there is a systematic difference between the first detection value and the second detection value, the first detection value and the second detection value are fitted based on the spline regression, and the first detection value or the second detection value is converted based on the fitting formula of the spline regression.

[0020] 2. The method according to item 1, wherein judging whether there is a systematic difference between the first detection value and the second detection value based on the Passing-Bablok regression fitting result is judging whether there is a systematic difference between the first detection value and the second detection value by calculating the 95% confidence interval of the intercept of the linear regression result.

[0021] 3. The method according to claim 2, wherein:

[0022] If the 95% confidence interval of the intercept includes 0, it is judged that there is no systematic difference between the first test value and the second test value;

[0023] If the 95% confidence interval of the intercept does not include 0, it is determined that there is a systematic difference between the first test value and the second test value.

[0024] 4. The method according to any one of items 1 to 3, wherein

[0025] When fitting the first detection value and the second detection value based on the spline regression, the nodes of the spline regression are confirmed by using the inherent nodes and the statistical index.

[0026] 5. The method according to item 4, wherein the statistical indicator is an AIC (Akaike Information Criterion) or a BIC (Bayesian Information Criterion) indicator.

[0027] 6. The method according to any one of items 1 to 5, wherein the hormone is a sex hormone, preferably the sex hormone is selected from the group consisting of hypothalamus-pituitary-gonadal axis hormones, including hypothalamic hormones, pituitary prolactin (PRL), follicle-stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH), inhibin B (inhibinB), estradiol (E2), testosterone (T), dehydroepiandrosterone sulfate (DHEA-S), estriol (E3), progesterone (P), preferably anti-Mullerian hormone (AMH) or estradiol (E2).

[0028] 7. The method according to item 6, wherein the sex hormone is anti-Mullerian hormone (AMH), and when the first platform is the Roche AMH detection platform and the second platform is the Kangrun chemiluminescent AMH detection platform, Passing-Bablok regression is used to convert the detection results of the Roche AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0029] The preferred formula for converting Kangrun chemiluminescent AMH (X) to Roche AMH (Y) is:

[0030] Roche AMH = a1 + b1 * Kangrun chemiluminescent AMH;

[0031] Wherein a1 is any value selected from -0.0035 to 0.0303; b1 is any value selected from 0.85 to 0.8844;

[0032] The preferred formula for converting Roche AMH (X) to Kangrun chemiluminescent AMH (Y) is:

[0033] Kangrun chemiluminescence AMH = a2 + b2 * Roche AMH;

[0034] Wherein a2 is any value selected from -0.0356 to 0.00393; b2 is any value selected from 1.1307 to 1.1765.

[0035] 8. The method according to item 6, wherein the sex hormone is anti-Mullerian hormone (AMH), and when the first platform is the Beckman AMH detection platform and the second platform is the Roche AMH detection platform, spline regression is used to transform the detection results of the Beckman AMH detection platform and the Roche AMH detection platform;

[0036] The preferred formula for converting Roche AMH (X) to Beckman AMH (Y) is:

[0037] Beckman AMH = a3 + b3 * Roche AMH + c3 * g1 (Roche AMH-1) + d3 * g2 (Roche AMH-9),

[0038] In the formula, if Roche AMH <= 1 ng / mL, then g1 = 0, g2 = 0, and the formula is:

[0039] Beckman AMH = a3 + b3 * Roche AMH

[0040] If Roche AMH>1 ng / mL and <=9 ng / mL, then g1=1, g2=0, and the formula is:

[0041] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1)

[0042] If Roche AMH>9 ng / mL, then g1=1, g2=1, and the formula is:

[0043] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1) + d3 * (Roche AMH-9)

[0044] Wherein a3 is selected from any value between -0.28809 and 0.30461; b3 is selected from any value between 0.665236 and 1.430824; c3 is selected from any value between 0.073606 and 0.900514; d3 is selected from any value between -0.70344 and -0.42116;

[0045] The preferred formula for converting Beckman AMH (X) to Roche AMH (Y) is:

[0046] Roche AMH = a4 + b4 * Beckmann AMH + c4 * g1 (Beckmann AMH-0.5) + d4 * g2 (Beckmann AMH-1)

[0047] In the formula, if the Beckman AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0048] Roche AMH = a4 + b4 * Beckmann AMH

[0049] If the Beckman AMH value is >0.5 ng / mL and <=1 ng / mL, then g1=1, g2=0, and the formula is:

[0050] Roche AMH = a4 + b4 * Beckmann AMH + c4 * (Beckmann AMH - 0.5)

[0051] If the Beckman AMH value is > 1 ng / mL, then g1 = 1, g2 = 1, and the formula is:

[0052] Roche AMH = a4 + b4*Beckmann AMH + c4*(Beckmann AMH-0.5) + d4*(Beckmann AMH-1);

[0053] a4 is selected from any value between -0.42466 and 0.41822; b4 is selected from any value between -0.11202 and 2.17363; c4 is selected from any value between -1.96122 and 1.21734; d4 is selected from any value between -0.49419 and 0.66553.

[0054] 9. The method according to item 6, wherein the sex hormone is anti-Mullerian hormone (AMH), and when the first platform is the Beckman AMH detection platform and the second platform is the Kangrun chemiluminescent AMH detection platform, spline regression is used to convert the detection results of the Beckman AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0055] The preferred formula for converting Beckman AMH (X) to Kangrun chemiluminescent AMH (Y) is:

[0056] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*g1(Beckman AMH-1)+d5*g2(Beckman AMH-4)+e5*g3(Beckman AMH-13)

[0057] In the formula, if the Beckman AMH value is less than or equal to 1 ng / ml, then g1=0, g2=0, g3=0, and the formula is:

[0058] Kangrun chemiluminescence AMH=a5+b5*Beckmann AMH

[0059] If the Beckman AMH value is >1 ng / mL and <=4 ng / mL, then g1=1, g2=0, g3=0, and the formula is:

[0060] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)

[0061] If the Beckman AMH value is >4 ng / mL and <=13 ng / mL, then g1=1, g2=1, g3=0, and the formula is:

[0062] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)

[0063] If the Beckman AMH value is >13 ng / mL, then g1=1, g2=1, g3=1, and the formula is:

[0064] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)+e5*(Beckman AMH-13),

[0065] Wherein, a5 is selected from any value between -0.24101 and 0.26258; b5 is selected from any value between 0.72643 and 1.42037; c5 is selected from any value between -0.78107 and 0.04326; d5 is selected from any value between 0.29488 and 0.57880; e5 is selected from any value between -0.71455 and -0.47130;

[0066] The preferred formula for converting Kangrun chemiluminescence AMH (X) to Beckman AMH (Y) is:

[0067] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * g1 (Kangrun chemiluminescence AMH-0.5) + d6 * g2 (Kangrun chemiluminescence AMH-13);

[0068] In the formula, if the Kangrun chemiluminescence AMH value is less than or equal to 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0069] Beckman AMH=a6+b6*Kangrun chemiluminescence AMH

[0070] If the Kangrun chemiluminescent AMH value is >0.5 and <=13 ng / ml, then g1=1, g2=0, and the formula is:

[0071] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5)

[0072] If the Kangrun chemiluminescent AMH value is >13 ng / ml, then g1=1, g2=1, and the formula is:

[0073] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5) + d6 * (Kangrun chemiluminescence AMH-13);

[0074] Among them, a6 is selected from any value between -0.45082 and 0.54656; b6 is selected from any value between -0.35329 and 1.82169; c6 is selected from any value between -0.63742 and 1.56010; d6 is selected from any value between -0.52283 and -0.22757.

[0075] 10. The method according to item 6, wherein the sex hormone is anti-Mullerian hormone (AMH), wherein the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is Kangrun chemiluminescent AMH detection platform, and the second platform is Kangrun enzyme-immune (Anshlab) AMH detection platform, and spline regression is used to convert the detection results of Kangrun chemiluminescent AMH detection platform and Kangrun enzyme-immune (Anshlab) AMH detection platform;

[0076] The preferred formula for converting Kangrun chemiluminescence AMH (X) to Kangrun chemiluminescence AMH (Y) is:

[0077] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 * (Kangrun chemiluminescent AMH-2) + d7 * g2 * (Kangrun chemiluminescent AMH-4) + e7 * g3 * (Kangrun chemiluminescent AMH-7)

[0078] Wherein, a7 is selected from any value in the range of -0.31000 to 0.20149, b7 is selected from any value in the range of 0.73863 to 1.10064, c7 is selected from any value in the range of -0.08941 to 0.51000, d7 is selected from any value in the range of 0.08121 to 0.54582, and e7 is selected from any value in the range of -0.70493 to -0.42738;

[0079] further,

[0080] If Kangrun chemiluminescent AMH <= 2 ng / ml, then g1 = 0, and the formula is:

[0081] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH;

[0082] If Kangrun chemiluminescent AMH>2ng / ml and <=4ng / ml, then g1=1, g2=0, and the formula is:

[0083] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2);

[0084] If Kangrun chemiluminescent AMH>4 ng / ml and <=7 ng / ml, then g1=1, g2=1, and the formula is:

[0085] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2) + d7 * g2 (Kangrun chemiluminescent AMH-4)

[0086] If Kangrun chemiluminescent AMH>7 ng / ml, then g1=1, g2=1, g3=1, and the formula is:

[0087] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7*Kangrun chemiluminescent AMH + c7*g1 (Kangrun chemiluminescent AMH-2) + d7*g2 (Kangrun chemiluminescent AMH-4) + e7*g3 (Kangrun chemiluminescent AMH-7).

[0088] 11. The method according to item 7, wherein the sex hormone is estradiol E2, wherein when the first platform is the Kangrun estradiol detection platform and the second platform is the Siemens estradiol detection platform, spline regression is used to convert the detection results of Kangrun estradiol and Siemens estradiol;

[0089] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * g1 (Kangrun E2 test value - 660) + d8 * g2 (Kangrun E2 test value - 2030)

[0090] In the formula,

[0091] If the Kangrun E2 test value is <= 660pmol / L, then g1=0, g2=0, and the formula is:

[0092] Siemens E2 test value = a8 + b8 * Kangrun E2 test value

[0093] If the Kangrun E2 test value is >660pmol / L and <=2030pmol / L, then g1=1, g2=0, and the formula is:

[0094] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660)

[0095] If the Kangrun E2 test value is >2030pmol / L, then g1=1, g2=1, and the formula is:

[0096] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660) + d8 * (Kangrun E2 test value - 2030)

[0097] Among them, a7 is selected from any value between -93.069 and -6.91207; b7 is selected from any value between 1.310892 and 1.569808; c7 is selected from any value between -0.8957 and -0.52322; d7 is selected from any value between -0.08505 and 0.085318.

[0098] 12. A system for converting hormone data detected by different detection methods, comprising:

[0099] A submodule for acquiring a first detection value for the hormone from a first platform;

[0100] A submodule for acquiring a second detection value for the hormone from a second platform;

[0101] A submodule for fitting the first detection value and the second detection value using Passing-Bablok regression;

[0102] A submodule for determining whether there is a systematic difference between the first detection value and the second detection value based on the passing-Bablok regression fitting result;

[0103] In the submodule for determining whether there is a systematic difference between the first detection value and the second detection value based on the passing-Bablok regression fitting result, the following steps are performed:

[0104] If there is no systematic difference between the first detection value and the second detection value, the first detection value or the second detection value is converted based on the fitting formula of Passing-Bablok regression.

[0105] If there is a systematic difference between the first detection value and the second detection value, the first detection value and the second detection value are fitted based on the spline regression, and the first detection value or the second detection value is converted based on the fitting formula of the spline regression.

[0106] 13. A system according to item 12, wherein the determination of whether there is a systematic difference between the first detection value and the second detection value based on the Passing-Bablok regression fitting result is performed by calculating a 95% confidence interval of the intercept of the linear regression result to determine whether there is a systematic difference between the first detection value and the second detection value.

[0107] 14. A system according to claim 13, wherein:

[0108] If the 95% confidence interval of the intercept includes 0, it is judged that there is no systematic difference between the first test value and the second test value;

[0109] If the 95% confidence interval of the intercept does not include 0, it is determined that there is a systematic difference between the first test value and the second test value.

[0110] 15. A system according to any one of items 12 to 14, wherein:

[0111] When fitting the first detection value and the second detection value based on the spline regression, the nodes of the spline regression are confirmed by using the inherent nodes and the statistical index.

[0112] 16. A system according to item 15, wherein the statistical indicator is an AIC (Akaike Information Criterion) or a BIC (Bayesian Information Criterion) indicator.

[0113] 17. A system according to any one of items 12 to 16, wherein the hormone is a sex hormone, which is selected from the hypothalamus-pituitary-gonadal axis hormone, including hypothalamic hormones, pituitary prolactin (PRL), follicle-stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH), inhibin B (inhibinB), estradiol (E2), testosterone (T), dehydroepiandrosterone sulfate (DHEA-S), estriol (E3), progesterone (P), preferably anti-Mullerian hormone (AMH) or estradiol (E2).

[0114] 19. The system according to item 18, wherein the sex hormone is anti-Mullerian hormone (AMH), wherein when the first platform is the Roche AMH detection platform and the second platform is the Kangrun chemiluminescent AMH detection platform, Passing-Bablok regression is used to convert the detection results of the Roche AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0115] The preferred formula for converting Kangrun chemiluminescent AMH (X) to Roche AMH (Y) is:

[0116] Roche AMH = a1 + b1 * Kangrun chemiluminescent AMH;

[0117] Wherein a1 is any value selected from -0.0035 to 0.0303; b1 is any value selected from 0.85 to 0.8844;

[0118] The preferred formula for converting Roche AMH (X) to Kangrun chemiluminescent AMH (Y) is:

[0119] Kangrun chemiluminescence AMH = a2 + b2 * Roche AMH;

[0120] Wherein a2 is any value selected from -0.0356 to 0.00393; b2 is any value selected from 1.1307 to 1.1765.

[0121] 20. The system according to item 18, wherein the sex hormone is anti-Mullerian hormone (AMH), and when the first platform is the Beckman AMH detection platform and the second platform is the Roche AMH detection platform, spline regression is used to transform the detection results of the Beckman AMH detection platform and the Roche AMH detection platform;

[0122] The preferred formula for converting Roche AMH (X) to Beckman AMH (Y) is:

[0123] Beckman AMH = a3 + b3 * Roche AMH + c3 * g1 (Roche AMH-1) + d3 * g2 (Roche AMH-9),

[0124] In the formula, if Roche AMH <= 1 ng / mL, then g1 = 0, g2 = 0, and the formula is:

[0125] Beckman AMH = a3 + b3 * Roche AMH

[0126] If Roche AMH>1 ng / mL and <=9 ng / mL, then g1=1, g2=0, and the formula is:

[0127] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1)

[0128] If Roche AMH>9 ng / mL, then g1=1, g2=1, and the formula is:

[0129] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1) + d3 * (Roche AMH-9)

[0130] Wherein a3 is selected from any value between -0.28809 and 0.30461; b3 is selected from any value between 0.665236 and 1.430824; c3 is selected from any value between 0.073606 and 0.900514; d3 is selected from any value between -0.70344 and -0.42116;

[0131] The preferred formula for converting Beckman AMH (X) to Roche AMH (Y) is:

[0132] Roche AMH = a4 + b4 * Beckmann AMH + c4 * g1 (Beckmann AMH-0.5) + d4 * g2 (Beckmann AMH-1)

[0133] In the formula, if the Beckman AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0134] Roche AMH = a4 + b4 * Beckmann AMH

[0135] If the Beckman AMH value is >0.5 ng / mL and <=1 ng / mL, then g1=1, g2=0, and the formula is:

[0136] Roche AMH = a4 + b4 * Beckmann AMH + c4 * (Beckmann AMH - 0.5)

[0137] If the Beckman AMH value is > 1 ng / mL, then g1 = 1, g2 = 1, and the formula is:

[0138] Roche AMH = a4 + b4*Beckmann AMH + c4*(Beckmann AMH-0.5) + d4*(Beckmann AMH-1);

[0139] a4 is selected from any value between -0.42466 and 0.41822; b4 is selected from any value between -0.11202 and 2.17363; c4 is selected from any value between -1.96122 and 1.21734; d4 is selected from any value between -0.49419 and 0.66553.

[0140] 21. The system according to item 18, wherein the sex hormone is anti-Mullerian hormone (AMH), wherein when the first platform is the Beckman AMH detection platform and the second platform is the Kangrun chemiluminescent AMH detection platform, spline regression is used to transform the detection results of the Beckman AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0141] The preferred formula for converting Beckman AMH (X) to Kangrun chemiluminescent AMH (Y) is:

[0142] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*g1(Beckman AMH-1)+d5*g2(Beckman AMH-4)+e5*g3(Beckman AMH-13)

[0143] In the formula, if the Beckman AMH value is less than or equal to 1 ng / ml, then g1=0, g2=0, g3=0, and the formula is:

[0144] Kangrun chemiluminescence AMH=a5+b5*Beckmann AMH

[0145] If the Beckman AMH value is >1 ng / mL and <=4 ng / mL, then g1=1, g2=0, g3=0, and the formula is:

[0146] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)

[0147] If the Beckman AMH value is >4 ng / mL and <=13 ng / mL, then g1=1, g2=1, g3=0, and the formula is:

[0148] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)

[0149] If the Beckman AMH value is >13 ng / mL, then g1=1, g2=1, g3=1, and the formula is:

[0150] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)+e5*(Beckman AMH-13),

[0151] Wherein, a5 is selected from any value between -0.24101 and 0.26258; b5 is selected from any value between 0.72643 and 1.42037; c5 is selected from any value between -0.78107 and 0.04326; d5 is selected from any value between 0.29488 and 0.57880; e5 is selected from any value between -0.71455 and -0.47130;

[0152] The preferred formula for converting Kangrun chemiluminescence AMH (X) to Beckman AMH (Y) is:

[0153] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * g1 (Kangrun chemiluminescence AMH-0.5) + d6 * g2 (Kangrun chemiluminescence AMH-13);

[0154] In the formula, if the Kangrun chemiluminescence AMH value is less than or equal to 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0155] Beckman AMH=a6+b6*Kangrun chemiluminescence AMH

[0156] If the Kangrun chemiluminescent AMH value is >0.5 and <=13 ng / ml, then g1=1, g2=0, and the formula is:

[0157] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5)

[0158] If the Kangrun chemiluminescent AMH value is >13 ng / ml, then g1=1, g2=1, and the formula is:

[0159] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5) + d6 * (Kangrun chemiluminescence AMH-13);

[0160] Among them, a6 is selected from any value between -0.45082 and 0.54656; b6 is selected from any value between -0.35329 and 1.82169; c6 is selected from any value between -0.63742 and 1.56010; d6 is selected from any value between -0.52283 and -0.22757.

[0161] 22. A system according to item 18, wherein the sex hormone is anti-Mullerian hormone (AMH), wherein the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is Kangrun chemiluminescent AMH detection platform, and the second platform is Kangrun enzyme-immune (Anshlab) AMH detection platform, and spline regression is used to convert the detection results of Kangrun chemiluminescent AMH detection platform and Kangrun enzyme-immune (Anshlab) AMH detection platform;

[0162] The preferred formula for converting Kangrun chemiluminescence AMH (X) to Kangrun chemiluminescence AMH (Y) is:

[0163] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 * (Kangrun chemiluminescent AMH-2) + d7 * g2 * (Kangrun chemiluminescent AMH-4) + e7 * g3 * (Kangrun chemiluminescent AMH-7)

[0164] Wherein, a7 is selected from any value in the range of -0.31000 to 0.20149, b7 is selected from any value in the range of 0.73863 to 1.10064, c7 is selected from any value in the range of -0.08941 to 0.51000, d7 is selected from any value in the range of 0.08121 to 0.54582, and e7 is selected from any value in the range of -0.70493 to -0.42738;

[0165] further,

[0166] If Kangrun chemiluminescent AMH <= 2 ng / ml, then g1 = 0, and the formula is:

[0167] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH;

[0168] If Kangrun chemiluminescent AMH>2ng / ml and <=4ng / ml, then g1=1, g2=0, and the formula is:

[0169] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2);

[0170] If Kangrun chemiluminescent AMH>4 ng / ml and <=7 ng / ml, then g1=1, g2=1, and the formula is:

[0171] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2) + d7 * g2 (Kangrun chemiluminescent AMH-4)

[0172] If Kangrun chemiluminescent AMH>7 ng / ml, then g1=1, g2=1, g3=1, and the formula is:

[0173] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7*Kangrun chemiluminescent AMH + c7*g1 (Kangrun chemiluminescent AMH-2) + d7*g2 (Kangrun chemiluminescent AMH-4) + e7*g3 (Kangrun chemiluminescent AMH-7).

[0174] 23. The system according to item 18, wherein the sex hormone is estradiol E2, wherein when the first platform is the Kangrun estradiol detection platform and the second platform is the Siemens estradiol detection platform, spline regression is used to convert the detection results of Kangrun estradiol and Siemens estradiol;

[0175] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * g1 (Kangrun E2 test value - 660) + d8 * g2 (Kangrun E2 test value - 2030)

[0176] In the formula,

[0177] If the Kangrun E2 test value is <= 660pmol / L, then g1=0, g2=0, and the formula is:

[0178] Siemens E2 test value = a8 + b8 * Kangrun E2 test value

[0179] If the Kangrun E2 test value is >660pmol / L and <=2030pmol / L, then g1=1, g2=0, and the formula is:

[0180] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660)

[0181] If the Kangrun E2 test value is >2030pmol / L, then g1=1, g2=1, and the formula is:

[0182] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660) + d8 * (Kangrun E2 test value - 2030)

[0183] Among them, a7 is selected from any value between -93.069 and -6.91207; b7 is selected from any value between 1.310892 and 1.569808; c7 is selected from any value between -0.8957 and -0.52322; d7 is selected from any value between -0.08505 and 0.085318.

[0184] Effects of the Invention

[0185] Through the method and system of the present application, the serum hormone test values ​​obtained by different hormone detection methods can be converted and calculated, and patients can easily standardize the hormone test data obtained in different hospitals or testing institutions, that is, the results of the same hormone test items of different testing systems in different hospitals are mutually recognized through certain formula transformation. Through this data conversion, on the one hand, it is possible to reduce the cost of sex hormone testing for patients when they visit different hospitals or testing institutions; on the other hand, it is conducive to the promotion and application of artificial intelligence software based on a certain testing platform data to different testing platforms for the same indicator items, which is conducive to the development of subsequent digital medicine. For example, our team has established a variety of digital medical-related applications using the reproductive medicine big data of Peking University Third Hospital, including tools such as evaluating ovarian reserve function, evaluating the number of oocytes, screening PCOS, and calculating the starting and adjusting doses of exogenous FSH. These tools all need to input the level of AMH, but our tools are only based on the prediction of the Kangrun chemiluminescent AMH hormone level of the Kangrun Anshlab platform. If it is applied to different AMH detection brands, it is necessary to apply the present invention for conversion, which can reduce the cost of patient hormone testing and ensure the accuracy of digital medical tools. BRIEF DESCRIPTION OF THE DRAWINGS

[0186] By reading the detailed description of the preferred specific embodiments below, various other advantages and benefits of the present application will become clear to those of ordinary skill in the art. The drawings in the specification are only used for the purpose of illustrating the preferred embodiments and are not considered to be limitations of the present application. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work. Moreover, the same reference numerals are used to represent the same components throughout the drawings.

[0187] Figure 1 A. Figure 1 B and Figure 1 C shows the fitting results of Roche AMH and Kangrun AMH, the consistency of the measurement results of Roche AMH and Kangrun AMH, and the mountain graph of Kangrun AMH and Roche AMH, respectively.

[0188] Figure 2 For the E2 results, Deming regression and Passing-Bablok regression were used to fit the result graph.

[0189] Figure 3 Bland-Altman plot of the consistency between Kangrun and Siemens for the E2 result.

[0190] Figure 4 A mountain-shaped graph showing Siemens E2 test value - Kangrun E2 test value.

[0191] Figure 5 Comparison of predictions between E2 platform-switched Passing-Bablok regression and spline regression.

[0192] Figure 6 Kangrun chemiluminescent AMH test results and the number of nodes of Kangrun enzymatic AMH test linear spline regression (Anshlab).

[0193] Figure 7 Kangrun chemiluminescent AMH test results and Kangrun enzymatic immunoassay (Anshlab) AMH test data grouping.

[0194] Figure 8 All data are calculated using the platform conversion formula for Anshlab AMH data.

[0195] Fig. 9 The low value (<7ng / ml) data were calculated using the platform conversion formula using the Anshlab AMH data. DETAILED DESCRIPTION

[0196] The specific embodiments of the present invention will be described in more detail below. However, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0197] It should be noted that certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that technicians may use different nouns to refer to the same component. This specification and claims do not use the difference in nouns as a way to distinguish components, but use the difference in the functions of the components as the criterion for distinction. As mentioned throughout the specification and claims, "including" or "comprising" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment of the present invention, but the description is based on the general principles of the specification and is not intended to limit the scope of the present invention. The scope of protection of the present invention shall be determined by the attached claims.

[0198] Passing-Bablok regression is mainly used to fit the consistency of the linear relationship between the dependent variable and the independent variable. This regression assumes that both the dependent variable and the independent variable are random. The idea of ​​Passing-Bablok regression is to calculate the slope of any two points and finally take the median of all slopes as the slope estimate. Passing-Bablok regression mainly tests the difference between the intercept and 0 and the difference between the slope and 1 to make a statistical analysis of whether the two methods are consistent.

[0199] Spline regression is a piecewise regression. Unlike linear regression, which fits a linear relationship on all data, spline regression can divide the data into several segments and fit a linear relationship in each segment. Therefore, spline regression is a piecewise polynomial function that is smoothly connected at the nodes. Spline regression requires determining the number of nodes and node values, as well as the form of the fitting curve in each segment, which can be a linear term (linear), a quadratic term, a cubic term, etc., which needs to be determined in combination with actual data.

[0200] The nodes in the spline regression process are usually determined based on the inherent nodes that have been confirmed in the field, supplemented by statistical indicators. When there are relatively clear nodes in the field based on the characteristics of the data itself (i.e., inherent nodes based on the characteristics of the data), they are determined based on the inherent nodes first; when there are no clear inherent nodes in the field, consider judging based on statistical indicators. Commonly used indicators include AIC (Akaike Information Criterion), BIC (Bayesian Information Criterion), etc., which are used to measure the performance of the model. The lower the AIC or BIC, the better the model fitting effect.

[0201] The present application provides a method for converting hormone data detected by different detection methods, which comprises:

[0202] Step 1: Obtaining a first detection value for the hormone from a first platform;

[0203] Step 2: Obtaining a second detection value for the hormone from a second platform;

[0204] Step 3: Use Passing-Bablok regression to fit the first detection value and the second detection value;

[0205] Step 4: judging whether there is a systematic difference between the first detection value and the second detection value based on the passing-Bablok regression fitting result;

[0206] If there is no systematic difference between the first detection value and the second detection value, the first detection value or the second detection value is converted based on the fitting formula of Passing-Bablok regression.

[0207] If there is a systematic difference between the first detection value and the second detection value, the first detection value and the second detection value are fitted based on the spline regression, and the first detection value or the second detection value is converted based on the fitting formula of the spline regression.

[0208] Specifically, the order of step 1 and step 2 is not critical.

[0209] Through the method of the present application, the serum hormone test values ​​obtained by different hormone detection methods can be converted and calculated, and patients can easily standardize the hormone test data obtained in different hospitals or testing institutions, that is, the results of the same hormone test items in different testing systems in different hospitals can be mutually recognized through certain formula transformations. Through this data conversion, on the one hand, it can reduce the cost of targeted hormone testing for patients when they visit different hospitals or testing institutions; on the other hand, it is conducive to the promotion and application of artificial intelligence software based on a certain testing platform data to different testing platforms for the same indicator item, which is conducive to the subsequent development of digital medicine.

[0210] In the present application, judging whether there is a systematic difference between the first detection value and the second detection value based on the Passing-Bablok regression fitting result is to judge whether there is a systematic difference between the first detection value and the second detection value by calculating the 95% confidence interval of the intercept of the linear regression result.

[0211] A confidence interval is an estimated interval of a population parameter constructed from a sample statistic. In statistics, a confidence interval of a probability sample is an interval estimate of a population parameter of the sample. The confidence interval shows the degree to which the true value of the parameter has a certain probability of falling around the measurement result. The confidence interval gives the degree of credibility of the measured value of the measured parameter, that is, the "probability" required above. This probability is called the confidence level, and the confidence space at a confidence level of 0.95 can also be expressed as a 95% confidence interval.

[0212] Specifically, if the 95% confidence interval of the intercept includes 0, it is determined that there is no systematic difference between the first test value and the second test value;

[0213] Specifically, if the 95% confidence interval of the intercept does not include 0, it is determined that there is a systematic difference between the first detection value and the second detection value.

[0214] In the present application, when fitting the first detection value and the second detection value based on spline regression, the nodes of the spline regression are confirmed by using inherent nodes and statistical indicators.

[0215] Specifically, the statistical indicator is an AIC (Akaike Information Criterion) or a BIC (Bayesian Information Criterion) indicator.

[0216] Akaike Information Criterion (AIC) is one of the commonly used criteria for comparing the goodness of fit of multiple regression models with different numbers of explanatory variables. This criterion requires that an explanatory variable be added to the original model only if the added explanatory variable can reduce the AIC value. If the added explanatory variable has no explanatory power, it will not help much in reducing the residual sum of squares, but will increase the number of parameters to be estimated, which may lead to an increase in the AIC value.

[0217] The Bayesian information criterion, also known as the Bayesian decision criterion, is a basic method in statistical pattern recognition. The Bayesian decision criterion takes into account both the probability of occurrence of various reference populations and the size of the loss caused by misjudgment, and has strong discrimination ability. The Bayesian method is more suitable for the following occasions:

[0218] (1) The number (capacity) of samples (sub-samples) is not large enough, so the statistical theory of large sub-samples is not suitable.

[0219] (2) The experiment is hereditary, which means that in statistics, there must be prior information before the experiment.

[0220] In the present application, the hormone is a sex hormone, preferably the sex hormone is selected from the group consisting of hypothalamus-pituitary-gonadal axis hormones, including hypothalamic hormones, pituitary prolactin (PRL), follicle-stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH), inhibin B, estradiol (E2), testosterone (T), dehydroepiandrosterone sulfate (DHEA-S), estriol (E3), and progesterone (P).

[0221] Specifically, the hormone is anti-Mullerian hormone (AMH) or estradiol (E2).

[0222] In the present application, the first platform can be Roche AMH detection platform, Kangrun chemiluminescent AMH detection platform, Beckman AMH detection platform, Kangrun enzyme immune (Anshlab) AMH detection platform, Kangrun estradiol detection platform and Siemens estradiol detection platform.

[0223] In the present application, the second platform can be Roche AMH detection platform, Kangrun chemiluminescent AMH detection platform, Beckman AMH detection platform, Kangrun enzyme immune (Anshlab) AMH detection platform, Kangrun estradiol detection platform and Siemens estradiol detection platform.

[0224] Specifically, when the first platform is the Roche AMH detection platform, the second platform can be the Kangrun chemiluminescent AMH detection platform, the Beckman AMH detection platform, the Kangrun enzyme immune (Anshlab) AMH detection platform, the Kangrun estradiol detection platform and the Siemens estradiol detection platform.

[0225] Specifically, when the first platform is the Kangrun chemiluminescent AMH detection platform, the second platform can be the Roche AMH detection platform, the Beckman AMH detection platform, the Kangrun enzyme immune (Anshlab) AMH detection platform, the Kangrun estradiol detection platform and the Siemens estradiol detection platform

[0226] Specifically, when the first platform is the Beckman AMH detection platform, the second platform can be the Roche AMH detection platform, the Kangrun chemiluminescent AMH detection platform, the Kangrun enzyme-immune (Anshlab) AMH detection platform, the Kangrun estradiol detection platform and the Siemens estradiol detection platform.

[0227] Specifically, when the first platform is the Anshlab AMH detection platform, the second platform can be the Roche AMH detection platform, the Anshlab chemiluminescent AMH detection platform, the Beckman AMH detection platform, the Anshlab estradiol detection platform and the Siemens estradiol detection platform.

[0228] Specifically, when the first platform is the Kangrun estradiol detection platform, the second platform can be the Roche AMH detection platform, the Kangrun chemiluminescent AMH detection platform, the Beckman AMH detection platform, the Kangrun enzyme-immune (Anshlab) AMH detection platform and the Siemens estradiol detection platform.

[0229] Specifically, when the first platform is the Siemens estradiol detection platform, the second platform can be the Roche AMH detection platform, the Kangrun chemiluminescent AMH detection platform, the Beckman AMH detection platform, the Kangrun enzyme-immune (Anshlab) AMH detection platform, the Kangrun estradiol detection platform and the Siemens estradiol detection platform.

[0230] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Roche AMH detection platform and the second platform is the Kangrun chemiluminescent AMH detection platform, Passing-Bablok regression is used to convert the detection results of the Roche AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0231] Specifically, the formula for converting Kangrun chemiluminescent AMH (X) to Roche AMH (Y) is:

[0232] Roche AMH = a1 + b1 * Kangrun chemiluminescent AMH;

[0233] Wherein a1 is any value selected from -0.0035 to 0.0303; b1 is any value selected from 0.85 to 0.8844;

[0234] Further, a1 can be -0.0035, -0.0015, 0, 0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.010, 0.012, 0.014, 0.016, 0.018, 0.020, 0.022, 0.024, 0.026, 0.028, 0.030 or 0.0303.

[0235] b1 can be 0.85, 0.86, 0.87, 0.88, 0.881, 0.882, 0.883, 0.884, 0.8841, 0.8842, 0.8843, and 0.8844.

[0236] Specifically, the formula for converting Roche AMH (X) to Kangrun chemiluminescent AMH (Y) is:

[0237] Kangrun chemiluminescence AMH = a2 + b2 * Roche AMH;

[0238] Wherein a2 is any value selected from -0.0356 to 0.00393; b2 is any value selected from 1.1307 to 1.1765.

[0239] Further, a2 may be -0.0356, -0.0366, -0.0376, -0.0386 or -0.0393.

[0240] b2 can be 1.1307, 1.1407, 1.1507, 1.1607, 1.1707 or 1.1765.

[0241] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Beckman AMH detection platform, and the second platform is the Roche AMH detection platform, and spline regression is used to convert the detection results of the Beckman AMH detection platform and the Roche AMH detection platform;

[0242] Specifically, the formula for converting Roche AMH (X) to Beckman AMH (Y) is:

[0243] Beckman AMH = a3 + b3 * Roche AMH + c3 * g1 (Roche AMH-1) + d3 * g2 (Roche AMH-9),

[0244] In the formula, if Roche AMH <= 1 ng / mL, then g1 = 0, g2 = 0, and the formula is:

[0245] Beckman AMH = a3 + b3 * Roche AMH

[0246] If Roche AMH>1 ng / mL and <=9 ng / mL, then g1=1, g2=0, and the formula is:

[0247] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1)

[0248] If Roche AMH>9 ng / mL, then g1=1, g2=1, and the formula is:

[0249] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1) + d3 * (Roche AMH-9)

[0250] Wherein a3 is selected from any value between -0.28809 and 0.30461; b3 is selected from any value between 0.665236 and 1.430824; c3 is selected from any value between 0.073606 and 0.900514; d3 is selected from any value between -0.70344 and -0.42116;

[0251] Further, a3 can be -0.28809, -0.18809, 0, 0.18809, 0.28809 or 0.30461.

[0252] b3 can be 0.665236, 0.765236, 0.865236, 0.965236, 1 or 1.430824.

[0253] c3 can be 0.073606, 0.083606, 0.093606, 0.103606, 0.203606, 0.303606, 0.403606, 0.503606, 0.603606, 0.703606, 0.803606 or 0.900514.

[0254] d3 can be -0.70344, -0.60344, -0.50344, or -0.42116.

[0255] Specifically, the formula for converting Beckman AMH (X) to Roche AMH (Y) is:

[0256] Roche AMH = a4 + b4 * Beckmann AMH + c4 * g1 (Beckmann AMH-0.5) + d4 * g2 (Beckmann AMH-1)

[0257] In the formula, if the Beckman AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0258] Roche AMH = a4 + b4 * Beckmann AMH

[0259] If the Beckman AMH value is >0.5 ng / mL and <=1 ng / mL, then g1=1, g2=0, and the formula is:

[0260] Roche AMH = a4 + b4 * Beckmann AMH + c4 * (Beckmann AMH - 0.5)

[0261] If the Beckman AMH value is > 1 ng / mL, then g1 = 1, g2 = 1, and the formula is:

[0262] Roche AMH = a4 + b4*Beckmann AMH + c4*(Beckmann AMH-0.5) + d4*(Beckmann AMH-1);

[0263] a4 is selected from any value between -0.42466 and 0.41822; b4 is selected from any value between -0.11202 and 2.17363; c4 is selected from any value between -1.96122 and 1.21734; d4 is selected from any value between -0.49419 and 0.66553.

[0264] Further, a4 may be -0.42466, -0.32466, -0.22466, -0.12466, -0.02466, 0, 0.12466, 0.22466, 0.32466 or 0.41822.

[0265] b4 can be -0.11202, -0.01202, 0, 0.11202, 0.21202, 0.31202, 0.41202, 0.51202, 0.61202, 0.71202, 0.81202, 0.91202, 1.11202, 1.21202, 1.31202, 1.41202, 1.51202, 1.61202, 1.71202, 1.81202, 1.91202, 2.01202, 2.11202 or 2.17363.

[0266] c4 can be -1.96122, -1.86122, -1.76122, -1.66122, -1.56122, -1.46122, -1.36122, -1.26122, -1.16122, -1.06122, -0.96122, -0.86122, -0.76122, -0.66122, -0.56122, - 0.46122, -0.36122, -0.26122, -0.16122, -0.06122, 0, 0.16122, 0.26122, 0.36122, 0.46122, 0.56122, 0.66122, 0.76122, 0.86122, 0.96122, 1.06122, 1.16122, or 1.21734.

[0267] d4 can be -0.49419, -0.39419, -0.29419, -0.19419, -0.09419, 0, 0.09419, 0.19419, 0.29419, 0.39419, 0.49419, 0.59419 or 0.66553.

[0268] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Beckman AMH detection platform, and the second platform is the Kangrun chemiluminescent AMH detection platform, and spline regression is used to convert the detection results of the Beckman AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0269] Specifically, the formula for converting Beckman AMH (X) to Kangrun chemiluminescence AMH (Y) is:

[0270] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*g1(Beckman AMH-1)+d5*g2(Beckman AMH-4)+e5*g3(Beckman AMH-13)

[0271] In the formula, if the Beckman AMH value is less than or equal to 1 ng / ml, then g1=0, g2=0, g3=0, and the formula is:

[0272] Kangrun chemiluminescence AMH=a5+b5*Beckmann AMH

[0273] If the Beckman AMH value is >1 ng / mL and <=4 ng / mL, then g1=1, g2=0, g3=0, and the formula is:

[0274] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)

[0275] If the Beckman AMH value is >4 ng / mL and <=13 ng / mL, then g1=1, g2=1, g3=0, and the formula is:

[0276] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)

[0277] If the Beckman AMH value is >13 ng / mL, then g1=1, g2=1, g3=1, and the formula is:

[0278] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)+e5*(Beckman AMH-13),

[0279] Wherein, a5 is selected from any value between -0.24101 and 0.26258; b5 is selected from any value between 0.72643 and 1.42037; c5 is selected from any value between -0.78107 and 0.04326; d5 is selected from any value between 0.29488 and 0.57880; e5 is selected from any value between -0.71455 and -0.47130;

[0280] Further, a5 can be -0.24101, -0.14101, -0.04101, 0, 0.14101, 0.24101 or 0.26258.

[0281] b5 can be 0.72643, 0.82643, 0.92643, 1.12643, 1.22643, 1.32643 or 1.42037.

[0282] c5 can be -0.78107, -0.68107, -0.58107, -0.48107, -0.38107, -0.28107, -0.18107, -0.08107, 0 or 0.04326.

[0283] d5 can be 0.29488, 0.39488, 0.49488 or 0.57880.

[0284] e5 can be -0.71455, -0.61455, -0.51455 or -0.47130.

[0285] Specifically, the formula for converting Kangrun chemiluminescence AMH (X) to Beckmann AMH (Y) is:

[0286] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * g1 (Kangrun chemiluminescence AMH-0.5) + d6 * g2 (Kangrun chemiluminescence AMH-13);

[0287] In the formula, if the Kangrun chemiluminescence AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0288] Beckman AMH=a6+b6*Kangrun chemiluminescence AMH

[0289] If the Kangrun chemiluminescent AMH value is >0.5 and <=13 ng / ml, then g1=1, g2=0, and the formula is:

[0290] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5)

[0291] If the Kangrun chemiluminescent AMH value is >13 ng / ml, then g1=1, g2=1, and the formula is:

[0292] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5) + d6 * (Kangrun chemiluminescence AMH-13);

[0293] Among them, a6 is selected from any value between -0.45082 and 0.54656; b6 is selected from any value between -0.35329 and 1.82169; c6 is selected from any value between -0.63742 and 1.56010; d6 is selected from any value between -0.52283 and -0.22757.

[0294] Further, a6 can be -0.45082, -0.35082, -0.25082, -0.15082, -0.05082, 0, 0.05082, 0.15082, 0.25082, 0.35082, 0.45082 or 0.54656.

[0295] b6 can be -0.35329, -0.25329, -0.15329, -0.05329, 0, 0.05329, 0.15329, 0.25329, 0.35329, 0.45329, 0.55329, 0.65329, 0.75329, 0.85329, 0.95329, 1.05329, 1.15329, 1.25329, 1.35329, 1.45329, 1.55329, 1.16329, 1.17329 or 1.82169.

[0296] c6 can be -0.63742, -0.53742, -0.43742, -0.33742, -0.23742, -0.13742, -0.03742, 0, 0.03742, 0.13742, 0.23742, 0.33742, 0.43742, 0.53742, 0.63742, 0.73742, 0.83742, 0.93742, 1.03742, 1.13742, 1.23742, 1.33742, 1.43742, or 1.56010.

[0297] d6 can be -0.52283, -0.42283, -0.32283 or -0.22757.

[0298] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is Kangrun chemiluminescent AMH detection platform, and the second platform is Kangrun enzyme-immune (Anshlab) AMH detection platform, spline regression is used to convert the detection results of Kangrun chemiluminescent AMH detection platform and Kangrun enzyme-immune (Anshlab) AMH detection platform;

[0299] Specifically, the formula for converting Kangrun chemiluminescence AMH (X) to Kangrun chemiluminescence AMH (Y) is:

[0300] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 * (Kangrun chemiluminescent AMH-2) + d7 * g2 * (Kangrun chemiluminescent AMH-4) + e7 * g3 * (Kangrun chemiluminescent AMH-7)

[0301] Wherein, a7 is selected from any value in the range of -0.31000 to 0.20149, b7 is selected from any value in the range of 0.73863 to 1.10064, c7 is selected from any value in the range of -0.08941 to 0.51000, d7 is selected from any value in the range of 0.08121 to 0.54582, and e7 is selected from any value in the range of -0.70493 to -0.42738;

[0302] Further, a7 may be -0.31000, -0.21000, -0.11000, -0.01000, 0, 0.01000, 0.11000 or 0.20149.

[0303] b7 can be 0.73863, 0.83863, 0.93863, 1.03863 or 1.10064.

[0304] c7 can be -0.08941, 0, 0.08941, 0.18941, 0.28941, 0.38941, 0.48941 or 0.51000.

[0305] d7 can be 0.08121, 0.18121, 0.28121, 0.38121, 0.48121 or 0.54582.

[0306] e7 can be -0.70493, -0.60493, -0.50493, or -0.42738.

[0307] Furthermore, if Kangrun chemiluminescent AMH <= 2 ng / ml, then g1 = 0, and the formula is:

[0308] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH;

[0309] If Kangrun chemiluminescent AMH>2ng / ml and <=4ng / ml, then g1=1, g2=0, and the formula is:

[0310] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2);

[0311] If Kangrun chemiluminescent AMH>4 ng / ml and <=7 ng / ml, then g1=1, g2=1, and the formula is:

[0312] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2) + d7 * g2 (Kangrun chemiluminescent AMH-4)

[0313] If Kangrun chemiluminescent AMH>7 ng / ml, then g1=1, g2=1, g3=1, and the formula is:

[0314] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7*Kangrun chemiluminescent AMH + c7*g1 (Kangrun chemiluminescent AMH-2) + d7*g2 (Kangrun chemiluminescent AMH-4) + e7*g3 (Kangrun chemiluminescent AMH-7).

[0315] In the present application, the sex hormone is estradiol E2, wherein the first platform is the Kangrun estradiol detection platform and the second platform is the Siemens estradiol detection platform, and spline regression is used to convert the detection results of Kangrun estradiol and Siemens estradiol;

[0316] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * g1 (Kangrun E2 test value - 660) + d8 * g2 (Kangrun E2 test value - 2030)

[0317] In the formula,

[0318] If the Kangrun E2 test value is <= 660pmol / L, then g1=0, g2=0, and the formula is:

[0319] Siemens E2 test value = a8 + b8 * Kangrun E2 test value

[0320] If the Kangrun E2 test value is >660pmol / L and <=2030pmol / L, then g1=1, g2=0, and the formula is:

[0321] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660)

[0322] If the Kangrun E2 test value is >2030pmol / L, then g1=1, g2=1, and the formula is:

[0323] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660) + d8 * (Kangrun E2 test value - 2030)

[0324] Among them, a7 is selected from any value between -93.069 and -6.91207; b7 is selected from any value between 1.310892 and 1.569808; c7 is selected from any value between -0.8957 and -0.52322; d7 is selected from any value between -0.08505 and 0.085318.

[0325] Further, a7 may be -93.069, -83.069, -73.069, -63.069, -53.069, -43.069, -33.069, -23.069, -13.069 or -6.91207.

[0326] b7 can be 1.310892, 1.410892, 1.510892 or 1.569808.

[0327] c7 can be -0.8957, -0.7957, -0.6957, -0.5957 or -0.52322.

[0328] d7 can be -0.08505, -0.07505, -0.06505, -0.05505, -0.04505, -0.03505, -0.02505, -0.01505, 0, 0.01505, 0.02505, 0.03505, 0.04505, 0.05505, 0.06505, 0.07505 or 0.085318.

[0329] The present application provides a system for converting hormone data detected by different detection methods, which includes:

[0330] A submodule for acquiring a first detection value for the hormone from a first platform;

[0331] A submodule for acquiring a second detection value for the hormone from a second platform;

[0332] A submodule for fitting the first detection value and the second detection value using Passing-Bablok regression;

[0333] A submodule for determining whether there is a systematic difference between the first detection value and the second detection value based on the passing-Bablok regression fitting result;

[0334] In the above submodule for determining whether there is a systematic difference between the first detection value and the second detection value based on the passing-Bablok regression fitting result, the following steps are performed:

[0335] If there is no systematic difference between the first detection value and the second detection value, the first detection value or the second detection value is converted based on the fitting formula of Passing-Bablok regression.

[0336] If there is a systematic difference between the first detection value and the second detection value, the first detection value and the second detection value are fitted based on the spline regression, and the first detection value or the second detection value is converted based on the fitting formula of the spline regression.

[0337] In the present application, judging whether there is a systematic difference between the first detection value and the second detection value based on the Passing-Bablok regression fitting result is to judge whether there is a systematic difference between the first detection value and the second detection value by calculating the 95% confidence interval of the intercept of the linear regression result.

[0338] In the present application, if the 95% confidence interval of the intercept includes 0, it is judged that there is no systematic difference between the first test value and the second test value;

[0339] If the 95% confidence interval of the intercept does not include 0, it is determined that there is a systematic difference between the first test value and the second test value.

[0340] In the present application, when fitting the first detection value and the second detection value based on spline regression, the nodes of the spline regression are confirmed by using inherent nodes and statistical indicators.

[0341] In the present application, the statistical index is the AIC (Akaike Information Criterion) or BIC (Bayesian Information Criterion) index.

[0342] In the present application, the hormone is a sex hormone, which is selected from the group consisting of hypothalamus-pituitary-gonadal axis hormones, including hypothalamic hormones, pituitary prolactin (PRL), follicle-stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH), inhibin B (inhibinB), estradiol (E2), testosterone (T), dehydroepiandrosterone sulfate (DHEA-S), estriol (E3), progesterone (P), preferably anti-Mullerian hormone (AMH) or estradiol (E2).

[0343] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Roche AMH detection platform and the second platform is the Kangrun chemiluminescent AMH detection platform, Passing-Bablok regression is used to convert the detection results of the Roche AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0344] Specifically, the formula for converting Kangrun chemiluminescent AMH (X) to Roche AMH (Y) is:

[0345] Roche AMH = a1 + b1 * Kangrun chemiluminescent AMH;

[0346] Wherein a1 is any value selected from -0.0035 to 0.0303; b1 is any value selected from 0.85 to 0.8844;

[0347] Further, a1 can be -0.0035, -0.0015, 0, 0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.010, 0.012, 0.014, 0.016, 0.018, 0.020, 0.022, 0.024, 0.026, 0.028, 0.030 or 0.0303.

[0348] b1 can be 0.85, 0.86, 0.87, 0.88, 0.881, 0.882, 0.883, 0.884, 0.8841, 0.8842, 0.8843, and 0.8844.

[0349] Specifically, the formula for converting Roche AMH (X) to Kangrun chemiluminescent AMH (Y) is:

[0350] Kangrun chemiluminescence AMH = a2 + b2 * Roche AMH;

[0351] Wherein a2 is any value selected from -0.0356 to 0.00393; b2 is any value selected from 1.1307 to 1.1765.

[0352] Further, a2 may be -0.0356, -0.0366, -0.0376, -0.0386 or -0.0393.

[0353] b2 can be 1.1307, 1.1407, 1.1507, 1.1607, 1.1707 or 1.1765.

[0354] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Beckman AMH detection platform, and the second platform is the Roche AMH detection platform, and spline regression is used to convert the detection results of the Beckman AMH detection platform and the Roche AMH detection platform;

[0355] Specifically, the formula for converting Roche AMH (X) to Beckman AMH (Y) is:

[0356] Beckman AMH = a3 + b3 * Roche AMH + c3 * g1 (Roche AMH-1) + d3 * g2 (Roche AMH-9),

[0357] In the formula, if Roche AMH <= 1 ng / mL, then g1 = 0, g2 = 0, and the formula is:

[0358] Beckman AMH = a3 + b3 * Roche AMH

[0359] If Roche AMH>1 ng / mL and <=9 ng / mL, then g1=1, g2=0, and the formula is:

[0360] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1)

[0361] If Roche AMH>9 ng / mL, then g1=1, g2=1, and the formula is:

[0362] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1) + d3 * (Roche AMH-9)

[0363] Wherein a3 is selected from any value between -0.28809 and 0.30461; b3 is selected from any value between 0.665236 and 1.430824; c3 is selected from any value between 0.073606 and 0.900514; d3 is selected from any value between -0.70344 and -0.42116;

[0364] Further, a3 can be -0.28809, -0.18809, 0, 0.18809, 0.28809 or 0.30461.

[0365] b3 can be 0.665236, 0.765236, 0.865236, 0.965236, 1 or 1.430824.

[0366] c3 can be 0.073606, 0.083606, 0.093606, 0.103606, 0.203606, 0.303606, 0.403606, 0.503606, 0.603606, 0.703606, 0.803606 or 0.900514.

[0367] d3 can be -0.70344, -0.60344, -0.50344, or -0.42116.

[0368] Specifically, the formula for converting Beckman AMH (X) to Roche AMH (Y) is:

[0369] Roche AMH = a4 + b4 * Beckmann AMH + c4 * g1 (Beckmann AMH-0.5) + d4 * g2 (Beckmann AMH-1)

[0370] In the formula, if the Beckman AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0371] Roche AMH = a4 + b4 * Beckmann AMH

[0372] If the Beckman AMH value is >0.5 ng / mL and <=1 ng / mL, then g1=1, g2=0, and the formula is:

[0373] Roche AMH = a4 + b4 * Beckmann AMH + c4 * (Beckmann AMH - 0.5)

[0374] If the Beckman AMH value is > 1 ng / mL, then g1 = 1, g2 = 1, and the formula is:

[0375] Roche AMH = a4 + b4*Beckmann AMH + c4*(Beckmann AMH-0.5) + d4*(Beckmann AMH-1);

[0376] a4 is selected from any value between -0.42466 and 0.41822; b4 is selected from any value between -0.11202 and 2.17363; c4 is selected from any value between -1.96122 and 1.21734; d4 is selected from any value between -0.49419 and 0.66553.

[0377] Further, a4 may be -0.42466, -0.32466, -0.22466, -0.12466, -0.02466, 0, 0.12466, 0.22466, 0.32466 or 0.41822.

[0378] b4 can be -0.11202, -0.01202, 0, 0.11202, 0.21202, 0.31202, 0.41202, 0.51202, 0.61202, 0.71202, 0.81202, 0.91202, 1.11202, 1.21202, 1.31202, 1.41202, 1.51202, 1.61202, 1.71202, 1.81202, 1.91202, 2.01202, 2.11202 or 2.17363.

[0379] c4 can be -1.96122, -1.86122, -1.76122, -1.66122, -1.56122, -1.46122, -1.36122, -1.26122, -1.16122, -1.06122, -0.96122, -0.86122, -0.76122, -0.66122, -0.56122, - 0.46122, -0.36122, -0.26122, -0.16122, -0.06122, 0, 0.16122, 0.26122, 0.36122, 0.46122, 0.56122, 0.66122, 0.76122, 0.86122, 0.96122, 1.06122, 1.16122, or 1.21734.

[0380] d4 can be -0.49419, -0.39419, -0.29419, -0.19419, -0.09419, 0, 0.09419, 0.19419, 0.29419, 0.39419, 0.49419, 0.59419 or 0.66553.

[0381] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Beckman AMH detection platform, and the second platform is the Kangrun chemiluminescent AMH detection platform, and spline regression is used to convert the detection results of the Beckman AMH detection platform and the Kangrun chemiluminescent AMH detection platform;

[0382] Specifically, the formula for converting Beckman AMH (X) to Kangrun chemiluminescence AMH (Y) is:

[0383] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*g1(Beckman AMH-1)+d5*g2(Beckman AMH-4)+e5*g3(Beckman AMH-13)

[0384] In the formula, if the Beckman AMH value is less than or equal to 1 ng / ml, then g1=0, g2=0, g3=0, and the formula is:

[0385] Kangrun chemiluminescence AMH=a5+b5*Beckmann AMH

[0386] If the Beckman AMH value is >1 ng / mL and <=4 ng / mL, then g1=1, g2=0, g3=0, and the formula is:

[0387] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)

[0388] If the Beckman AMH value is >4 ng / mL and <=13 ng / mL, then g1=1, g2=1, g3=0, and the formula is:

[0389] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)

[0390] If the Beckman AMH value is >13 ng / mL, then g1=1, g2=1, g3=1, and the formula is:

[0391] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)+e5*(Beckman AMH-13),

[0392] Wherein, a5 is selected from any value between -0.24101 and 0.26258; b5 is selected from any value between 0.72643 and 1.42037; c5 is selected from any value between -0.78107 and 0.04326; d5 is selected from any value between 0.29488 and 0.57880; e5 is selected from any value between -0.71455 and -0.47130;

[0393] Further, a5 can be -0.24101, -0.14101, -0.04101, 0, 0.14101, 0.24101 or 0.26258.

[0394] b5 can be 0.72643, 0.82643, 0.92643, 1.12643, 1.22643, 1.32643 or 1.42037.

[0395] c5 can be -0.78107, -0.68107, -0.58107, -0.48107, -0.38107, -0.28107, -0.18107, -0.08107, 0 or 0.04326.

[0396] d5 can be 0.29488, 0.39488, 0.49488 or 0.57880.

[0397] e5 can be -0.71455, -0.61455, -0.51455 or -0.47130.

[0398] Specifically, the formula for converting Kangrun chemiluminescence AMH (X) to Beckmann AMH (Y) is:

[0399] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * g1 (Kangrun chemiluminescence AMH-0.5) + d6 * g2 (Kangrun chemiluminescence AMH-13);

[0400] In the formula, if the Kangrun chemiluminescence AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0401] Beckman AMH=a6+b6*Kangrun chemiluminescence AMH

[0402] If the Kangrun chemiluminescent AMH value is >0.5 and <=13 ng / ml, then g1=1, g2=0, and the formula is:

[0403] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5)

[0404] If the Kangrun chemiluminescent AMH value is >13 ng / ml, then g1=1, g2=1, and the formula is:

[0405] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5) + d6 * (Kangrun chemiluminescence AMH-13);

[0406] Among them, a6 is selected from any value between -0.45082 and 0.54656; b6 is selected from any value between -0.35329 and 1.82169; c6 is selected from any value between -0.63742 and 1.56010; d6 is selected from any value between -0.52283 and -0.22757.

[0407] Further, a6 can be -0.45082, -0.35082, -0.25082, -0.15082, -0.05082, 0, 0.05082, 0.15082, 0.25082, 0.35082, 0.45082 or 0.54656.

[0408] b6 can be -0.35329, -0.25329, -0.15329, -0.05329, 0, 0.05329, 0.15329, 0.25329, 0.35329, 0.45329, 0.55329, 0.65329, 0.75329, 0.85329, 0.95329, 1.05329, 1.15329, 1.25329, 1.35329, 1.45329, 1.55329, 1.16329, 1.17329 or 1.82169.

[0409] c6 can be -0.63742, -0.53742, -0.43742, -0.33742, -0.23742, -0.13742, -0.03742, 0, 0.03742, 0.13742, 0.23742, 0.33742, 0.43742, 0.53742, 0.63742, 0.73742, 0.83742, 0.93742, 1.03742, 1.13742, 1.23742, 1.33742, 1.43742, or 1.56010.

[0410] d6 can be -0.52283, -0.42283, -0.32283 or -0.22757.

[0411] In the present application, the sex hormone is anti-Mullerian hormone (AMH), wherein the sex hormone is anti-Mullerian hormone (AMH), wherein the first platform is Kangrun chemiluminescent AMH detection platform, and the second platform is Kangrun enzyme-immune (Anshlab) AMH detection platform, spline regression is used to convert the detection results of Kangrun chemiluminescent AMH detection platform and Kangrun enzyme-immune (Anshlab) AMH detection platform;

[0412] Specifically, the formula for converting Kangrun chemiluminescence AMH (X) to Kangrun chemiluminescence AMH (Y) is:

[0413] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 * (Kangrun chemiluminescent AMH-2) + d7 * g2 * (Kangrun chemiluminescent AMH-4) + e7 * g3 * (Kangrun chemiluminescent AMH-7)

[0414] Wherein, a7 is selected from any value in the range of -0.31000 to 0.20149, b7 is selected from any value in the range of 0.73863 to 1.10064, c7 is selected from any value in the range of -0.08941 to 0.51000, d7 is selected from any value in the range of 0.08121 to 0.54582, and e7 is selected from any value in the range of -0.70493 to -0.42738;

[0415] Further, a7 may be -0.31000, -0.21000, -0.11000, -0.01000, 0, 0.01000, 0.11000 or 0.20149.

[0416] b7 can be 0.73863, 0.83863, 0.93863, 1.03863 or 1.10064.

[0417] c7 can be -0.08941, 0, 0.08941, 0.18941, 0.28941, 0.38941, 0.48941 or 0.51000.

[0418] d7 can be 0.08121, 0.18121, 0.28121, 0.38121, 0.48121 or 0.54582.

[0419] e7 can be -0.70493, -0.60493, -0.50493, or -0.42738.

[0420] Furthermore, if Kangrun chemiluminescent AMH <= 2 ng / ml, then g1 = 0, and the formula is:

[0421] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH;

[0422] If Kangrun chemiluminescent AMH>2ng / ml and <=4ng / ml, then g1=1, g2=0, and the formula is:

[0423] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2);

[0424] If Kangrun chemiluminescent AMH>4 ng / ml and <=7 ng / ml, then g1=1, g2=1, and the formula is:

[0425] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2) + d7 * g2 (Kangrun chemiluminescent AMH-4)

[0426] If Kangrun chemiluminescent AMH>7 ng / ml, then g1=1, g2=1, g3=1, and the formula is:

[0427] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7*Kangrun chemiluminescent AMH + c7*g1 (Kangrun chemiluminescent AMH-2) + d7*g2 (Kangrun chemiluminescent AMH-4) + e7*g3 (Kangrun chemiluminescent AMH-7).

[0428] In the present application, the sex hormone is estradiol E2, wherein the first platform is the Kangrun estradiol detection platform and the second platform is the Siemens estradiol detection platform, and spline regression is used to convert the detection results of Kangrun estradiol and Siemens estradiol;

[0429] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * g1 (Kangrun E2 test value - 660) + d8 * g2 (Kangrun E2 test value - 2030)

[0430] In the formula,

[0431] If the Kangrun E2 test value is <= 660pmol / L, then g1=0, g2=0, and the formula is:

[0432] Siemens E2 test value = a8 + b8 * Kangrun E2 test value

[0433] If the Kangrun E2 test value is >660pmol / L and <=2030pmol / L, then g1=1, g2=0, and the formula is:

[0434] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660)

[0435] If the Kangrun E2 test value is >2030pmol / L, then g1=1, g2=1, and the formula is:

[0436] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660) + d8 * (Kangrun E2 test value - 2030)

[0437] Among them, a7 is selected from any value between -93.069 and -6.91207; b7 is selected from any value between 1.310892 and 1.569808; c7 is selected from any value between -0.8957 and -0.52322; d7 is selected from any value between -0.08505 and 0.085318.

[0438] Further, a7 may be -93.069, -83.069, -73.069, -63.069, -53.069, -43.069, -33.069, -23.069, -13.069 or -6.91207.

[0439] b7 can be 1.310892, 1.410892, 1.510892 or 1.569808.

[0440] c7 can be -0.8957, -0.7957, -0.6957, -0.5957 or -0.52322.

[0441] d7 can be -0.08505, -0.07505, -0.06505, -0.05505, -0.04505, -0.03505, -0.02505, -0.01505, 0, 0.01505, 0.02505, 0.03505, 0.04505, 0.05505, 0.06505, 0.07505 or 0.085318.

[0442] Example 1

[0443] For the selection of samples, reference was made to the previous AMH data distribution of the Reproductive Center of Peking University Third Hospital, as shown in Table 1. The AMH concentration of the collected samples was routinely tested using the Kangrun detection method. From November 2021 to February 2022, the inventors of this application randomly selected 300 serum AMH samples from the Reproductive Center of Peking University Third Hospital. These samples were taken from female patients undergoing standard in vitro fertilization (IVF). The samples were collected from the hospital's in vitro fertilization center, and the collected samples were frozen at -80°C. These patients all signed written informed consent forms and agreed to donate their remaining serum for research after the original blood test. Table 1 summarizes all AMH test results detected using the Kangrun method, all expressed in nanograms / ml.

[0444] Table 1. AMH distribution of samples collected in Example 1 based on Kangrun detection

[0445]

[0446] Example 2

[0447] For the two tubes of frozen serum of each sample obtained in Example 1, one tube was tested for AMH results using the Roche method after thawing, and the other tube was tested using the Kangrun chemiluminescence method and the Beckman method on the same day after thawing, thereby obtaining the AMH results of the three methods (which can be respectively referred to as "Roche AMH", "Beckman AMH" and "Kangrun chemiluminescence AMH").

[0448] If the test result of the serum sample collected in Example 1 exceeds the upper limit of the measuring range of each method, the sample is diluted using the reagent provided by the manufacturer of the method and tested again.

[0449] Roche detection used the Elecsys (Roche) AMH kit on a Roche Cobas e 602 instrument, Beckman detection used the Beckman Access AMH kit on a Beckman DxI 800 instrument, and Kangrun detection used the Kangrun chemiluminescent AMH kit on a Kangrun Kaeser 6600 instrument.

[0450] The quality control of the AMH test results detected by the three methods is carried out using the quality control products (control samples) provided by each AMH test kit. The difference between the test value of the control sample and the given target value is expressed as a deviation. Deviation ± 12% and coefficient of variation (CV) ≤ 8% are within a controllable range, and the test results are acceptable. Table 2 below shows the precision and accuracy results of each AMH test. For each detection method, the CV value of the analysis is ≤ 5%, and the deviation is ± 7%, indicating that the three methods established in the embodiments of the present application are stable, and the test results meet the requirements.

[0451] Table 2. Comparison of the precision and accuracy of three AMH assays.

[0452]

[0453] Example 3

[0454] In view of the widespread use of the three different AMH detection kits mentioned above, in order to unify these data, it is necessary to find a method to accurately standardize the AMH data detected by the three methods.

[0455] Generally speaking, if the AMH test data of the Beckman method, the AMH test data of the Kangrun chemiluminescence method, and the test data of the Roche method conform to the overall relationship, the Passing-Bablok regression can be considered to transform the data of each of the three. Passing-Bablok regression uses a non-parametric method to fit the parameters a and b in the linear equation y=a+bx. If the 95% confidence interval (CI) of the intercept a does not contain 0 or the 95% confidence interval of the slope b does not contain 1, it indicates that there is a systematic or proportional difference between the two AMH test results.

[0456] If the detection results of the two platform methods do not satisfy the overall relationship, you can consider using spline regression to find the transformation between the detection data of the two platforms. Spline regression is a sheet-like polynomial function that is smoothly connected at the nodes. It can be a linear relationship (straight lines connected at the nodes), a quadratic relationship, or a cubic relationship. In spline regression, based on the relationship between the independent variable and the dependent variable, linear spline regression can be used to try to build a linear regression that is smoothly connected at the nodes.

[0457] In Example 3, a Bland-Altman plot was also attempted to examine the differences in systematic bias and variance in different numerical ranges. 2 ), Corrected R 2 , root mean square error (RMSE), Aker information criterion (AIC) and corrected AIC (AICc) were used to evaluate the fitting effect of the data. 2 and the corrected R 2 The higher the AIC, the better the performance of the fitting model. The lower the RMSE, AIC and AICc, the better the fitting effect. All statistical analyses in this example were performed using JMP PRO v16.0 (Cary. NC, USA).

[0458] (1) First step: Passing-Bablok regression between the three AMH test results

[0459] Assuming that Roche AMH and Kangrun chemiluminescent AMH conform to the overall relationship, the Passing-Bablok regression method was used to construct the linear relationship between the two assay results. Figure 1 A shows the fitting results of Roche AMH and Kangrun chemiluminescent AMH. Figure 1 In A, the dotted line reflects the theoretical fitting results between the two AMH determinations, and the solid line is the Passing-Bablok regression result. Figure 1 As can be seen in A, there is a deviation between the solid line and the dotted line, and this difference needs to be statistically tested.

[0460] A Bland-Altman plot was further drawn to explore the consistency of the Roche AMH and Kangrun chemiluminescent AMH assay results. Figure 1 As shown in B. The horizontal axis represents the average value of Kangrun chemiluminescent AMH and Roche AMH, and the vertical axis represents the difference between Kangrun chemiluminescent AMH and Roche AMH. Figure 1 The gray solid line in B is the reference line, i.e., the value of 0 (representing the average difference between the two AMH test results of 0), and the middle solid line and dotted line represent the mean value and 95% confidence interval of Kangrun chemiluminescent AMH minus Roche AMH. The outer gray frame line (i.e., the gray area in the figure) is the 95% distribution range of the difference (mean ± 1.96 × standard deviation), also known as the limit of agreement. Theoretically, the more consistent the two measurement results are, the closer the middle solid line is to the reference line. If 95% of the points are located in the gray virtual frame and are normally distributed around the middle solid line, the consistency of the two AMH determinations can be obtained. In the present embodiment, the number of observation points exceeding the gray line is 8.5%, and the data distribution is obviously non-normal, especially the high value part. Therefore, based on the analysis results of the above-described embodiment, it is believed that the two AMH determinations are inconsistent.

[0461] further Figure 1 C shows the mountain graph of Kangrun chemiluminescent AMH and Roche AMH. The vertical axis of the mountain graph represents the detection value of Kangrun method minus the detection value of Roche method, and the vertical axis represents the percentile of cumulative distribution. When the cumulative distribution probability is less than 50%, it is the cumulative distribution probability, and when it is greater than 50%, it is 1-cumulative distribution probability. If the two AMH test results are consistent, they should be symmetrically distributed with 0 as the center. Figure 1 As can be seen in C, the data of "Kangrun chemiluminescent AMH-Roche AMH" presents a right-skewed tail, indicating that most of the differences of "Kangrun chemiluminescent AMH-Roche AMH" are greater than 0, and the right tail is very long, indicating that some data points of "Kangrun chemiluminescent AMH-Roche AMH" are very high. On the other hand, when the cumulative distribution is <50%, the difference between the two AMH detection methods is small.

[0462] The intercept of converting the detection result of AMH from the Roche method to the detection result of the Kangrun chemiluminescence method is -0.0124, and its 95% confidence interval (CI) includes 0, indicating that there is no systematic difference between Kangrun AMH and Roche AMH. The slope is 1.1517, and 95% of the Cl does not include 1, indicating that there may be a proportional difference between the two AMH determination results. Further using Passing-Bablok regression, the inventors of the present application have carried out other two-paired conversions between the three AMH determination results. The intercept and slope of the two-to-two conversion fitting results are summarized in the following table 3.

[0463] The inventors further standardized the pairwise data according to the fitting formula shown in Table 3. Table 4 shows the selected 30 data, and all the transformed or standardized data using Passing-Bablok regression were calculated (the results are not listed in detail).

[0464] Table 3. Conversion of AMH detection between Roche, Beckman, and Conrotron chemiluminescence methods using Passing-Bablok regression.

[0465]

[0466]

[0467] Here, CI stands for confidence interval.

[0468] Based on the above analysis, it can be seen that the conversion of Kangrun chemiluminescent AMH to Roche AMH, and Roche AMH to Kangrun chemiluminescent AMH, the results of Passing-Bablok regression fitting show that the intercept includes 0, so it is confirmed that the conversion formula of Kangrun chemiluminescent AMH to Roche AMH, and Roche AMH to Kangrun chemiluminescent AMH is as follows:

[0469] The formula for converting Kangrun chemiluminescent AMH (X) to Roche AMH (Y) is:

[0470] Roche AMH = a1 + b1 * Kangrun chemiluminescent AMH;

[0471] Wherein a1 is any value selected from -0.0035 to 0.0303; b1 is any value selected from 0.85 to 0.8844.

[0472] The formula for converting Roche AMH (X) to Kangrun chemiluminescent AMH (Y) is:

[0473] Kangrun chemiluminescence AMH = a2 + b2 * Roche AMH;

[0474] Wherein a2 is any value selected from -0.0356 to 0.00393; b2 is any value selected from 1.1307 to 1.1765.

[0475] (2) Step 2: Spline regression between the three AMH test results

[0476] The Passing-Bablok regression in (1) of Example 3 assumes that the correlation between the two AMH tests conforms to the overall relationship. However, as shown in Table 4 below and the remaining data after transformation or standardization using Passing-Bablok regression, in the four transformations except the transformation of Kangrun chemiluminescent AMH to Roche AMH, and the transformation of Roche AMH to Kangrun chemiluminescent AMH, the intercept of 95% CI does not include 0, and the fitting effect of the low value part is obviously not good, indicating that there is no overall relationship between each two of them. Therefore, spline regression is further used for transformation in Example 3 (2).

[0477] Table 4. Selectivity displayed for 30 samples converted from the AMH assay using the Passing-Bablok regression method.

[0478]

[0479]

[0480] Taking Roche AMH converted into Beckman AMH as an example, the AMH numerical value measured by Roche is used as the independent variable x, and the AMH numerical value measured by Beckman is used as the dependent variable y. According to the relationship between the independent variable and the dependent variable in the data of the embodiment, a linear spline regression is further adopted, i.e., a linear regression with smooth connection at the node is constructed. In the present embodiment 3 (2), the selection of nodes mainly considers the size of the inherent nodes, scatter plot relationship and BIC index of the data in clinical significance. Based on this, the inventor sets the nodes of two spline regressions, i.e., 1 ng / ml and "9 ng / ml, based on a large amount of research.

[0481] Table 5 Spline regression fitting results of the relationship between Roche and Beckmann

[0482]

[0483] According to the results in Table 5, the corresponding relationship between Beckman AMH and Roche AMH is established as follows:

[0484] Beckman AMH = a3 + b3 * Roche AMH + c3 * g1 (Roche AMH-1) + d3 * g2 (Roche AMH-9),

[0485] In the formula, if Roche AMH <= 1 ng / mL, then g1 = 0, g2 = 0, and the formula is:

[0486] Beckman AMH = a3 + b3 * Roche AMH

[0487] If Roche AMH>1 ng / mL and <=9 ng / mL, then g1=1, g2=0, and the formula is:

[0488] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1)

[0489] If Roche AMH>9 ng / mL, then g1=1, g2=1, and the formula is:

[0490] Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1) + d3 * (Roche AMH-9)

[0491] Wherein a3 is selected from any value between -0.28809 and 0.30461; b3 is selected from any value between 0.665236 and 1.430824; c3 is selected from any value between 0.073606 and 0.900514; d3 is selected from any value between -0.70344 and -0.42116.

[0492] Table 6 shows the comparison of the fitting effects of spline regression and Passing-Bablok regression. It can be seen that the RMSE, AIC and AICc of spline regression are lower than those of Passing-Bablok regression, while R 2 and the corrected R 2 Table 7 shows the spline regression estimates for converting AMH results from the Roche method to the Beckman method.

[0493] Table 6. Comparison of spline regression and Passing-Bablok regression when switching from Roche AMH to Beckman AMH

[0494] Spline regression Passing-Bablok Returns RMSE 0.858 0.943 <![CDATA[R 2 ]]> 0.977 0.972 <![CDATA[Corrected R 2 > 0.977 0.972 AIC 207.138 260.130 AICc 207.349 260.214

[0495] Table 7. Switching of AMH assays from Roche AMH to Beckman AMH using spline regression

[0496] Estimated value P-Value intercept 0.00826 0.9565 <1 1.04803 <.0001 1-9 0.48706 0.0217 >9 -0.5623 <.0001

[0497] Furthermore, a method for converting Beckman AMH(x) to Roche AMH(y) is described based on the regression results.

[0498] Table 8 Spline regression fitting results of the relationship between Roche and Beckmann

[0499]

[0500] According to the results in Table 8, the corresponding relationship between Roche AMH (y) and Beckman AMH (x) is established as follows:

[0501] Roche AMH = a4 + b4 * Beckmann AMH + c4 * g1 (Beckmann AMH-0.5) + d4 * g2 (Beckmann AMH-1)

[0502] In the formula, if the Beckman AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is:

[0503] Roche AMH = a4 + b4 * Beckmann AMH

[0504] If the Beckman AMH value is >0.5 ng / mL and <=1 ng / mL, then g1=1, g2=0, and the formula is:

[0505] Roche AMH = a4 + b4 * Beckmann AMH + c4 * (Beckmann AMH - 0.5)

[0506] If the Beckman AMH value is > 1 ng / mL, then g1 = 1, g2 = 1, and the formula is:

[0507] Roche AMH = a4 + b4 * Beckmann AMH + c4 * (Beckmann AMH-0.5) + d4 * (Beckmann AMH-1)

[0508] a4 is selected from any value between -0.42466 and 0.41822; b4 is selected from any value between -0.11202 and 2.17363; c4 is selected from any value between -1.96122 and 1.21734; d4 is selected from any value between -0.49419 and 0.66553.

[0509] When taking Beckman AMH to Roche AMH as an example, the AMH numerical value measured by Beckman is used as the independent variable x, and the AMH numerical value measured by Roche is used as the dependent variable y. According to the relationship between the independent variable and the dependent variable in the data of the embodiment, a linear spline regression is further adopted, i.e., a linear regression with smooth connection at the node is constructed. In the present embodiment 3 (2), the selection of the node is based on the principle of "the smaller the change, the better the formula", based on this, the inventor sets the nodes of two spline regressions, i.e., 0.5 ng / ml and 1 ng / ml, based on a large amount of research.

[0510] Furthermore, the method of converting Beckman AMH(x) to Kangrun AMH(y) is described based on the regression results.

[0511] Table 9 Spline regression fitting results of the relationship between Kangrun and Beckman

[0512] parameter Standard error t-value P-value 95% lower limit 95% cap Intercept 0.128467 0.08 0.9331 -0.24101 0.26258 <＝1 0.177026 6.06 <.0001 0.72643 1.42037 >1-4 0.210290 -1.75 0.0805 -0.78107 0.04326 >4-13 0.072428 6.03 <.0001 0.29488 0.57880 >13 0.062054 -9.55 <.0001 -0.71455 -0.47130

[0513] According to the results in Table 9, the corresponding relationship between Beckman AMH (x) and Kangrun chemiluminescence AMH (y) is established as follows:

[0514] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*g1(Beckman AMH-1)+d5*g2(Beckman AMH-4)+e5*g3(Beckman AMH-13)

[0515] In the formula, if the Beckman AMH value is less than or equal to 1 ng / ml, then g1=0, g2=0, g3=0, and the formula is:

[0516] Kangrun chemiluminescence AMH=a5+b5*Beckmann AMH

[0517] If the Beckman AMH value is >1 ng / mL and <=4 ng / mL, then g1=1, g2=0, g3=0, and the formula is:

[0518] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)

[0519] If the Beckman AMH value is >4 ng / mL and <=13 ng / mL, then g1=1, g2=1, g3=0, and the formula is:

[0520] Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)

[0521] If the Beckman AMH value is >13 ng / mL, then g1=1, g2=1, g3=1, and the formula is:

[0522] Kangrun chemiluminescence AMH = a5 + b5*Beckman AMH + c5*(Beckman AMH-1) + d5*(Beckman AMH-4) + e5*(Beckman AMH-13).

[0523] Among them, a5 is selected from any value between -0.24101 and 0.26258; b5 is selected from any value between 0.72643 and 1.42037; c5 is selected from any value between -0.78107 and 0.04326; d5 is selected from any value between 0.29488 and 0.57880; e5 is selected from any value between -0.71455 and -0.47130.

[0524] When taking Beckman AMH versus Kangrun chemiluminescent AMH as an example, the AMH value determined by Beckman is used as the independent variable x, and the AMH value determined by Kangrun chemiluminescent is used as the dependent variable y. According to the relationship between the independent variable and the dependent variable in the data of the embodiment, a linear spline regression is further adopted, i.e., a linear regression with smooth connection at the node is constructed. In the present embodiment 3 (2), the selection of nodes is based on the principle of "the smaller the change, the better the formula", based on which, the inventors set three nodes of spline regression based on a large number of studies, i.e., 1 ng / ml, 4 ng / ml and 13 ng / ml.

[0525] Furthermore, based on the regression results, the method of converting Kangrun chemiluminescent AMH(x) to Beckmann AMH(y) is described.

[0526] Table 10 Spline regression fitting results of the relationship between Kangrun chemiluminescence and Beckman

[0527]

[0528] According to the results in Table 10, the corresponding relationship between Beckman AMH and Kangrun chemiluminescence AMH is established as follows:

[0529] Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * g1 (Kangrun chemiluminescence AMH-0.5) + d6 * g2 (Kangrun chemiluminescence AMH-13)

[0530] In the formula, if the Kangrun chemiluminescence AMH value is <= 0.5, then g1=0, g2=0, and the formula is:

[0531] Beckman AMH=a6+b6*Kangrun chemiluminescence AMH

[0532] If the Kangrun chemiluminescence AMH value is > 0.5 and <= 13, then g1 = 1, g2 = 0, and the formula is:

[0533] Beckman AMH=a6+b6*Kangrun chemiluminescence AMH+c6*(Kangrun AMH-0.5)

[0534] If the Kangrun chemiluminescence AMH value is >13, then g1=1, g2=1, and the formula is:

[0535] Beckman AMH = a6 + b6*Kangrun chemiluminescence AMH + c6* (Kangrun chemiluminescence AMH-0.5) + d6* (Kangrun AMH-13).

[0536] Among them, a6 is selected from any value between -0.45082 and 0.54656; b6 is selected from any value between -0.35329 and 1.82169; c6 is selected from any value between -0.63742 and 1.56010; d6 is selected from any value between -0.52283 and -0.22757.

[0537] When Kangrun chemiluminescent AMH is used as an example for Beckman AMH, the AMH value determined by Kangrun chemiluminescent is used as the independent variable x, and the AMH value determined by Beckman is used as the dependent variable y. According to the relationship between the independent variable and the dependent variable in the data of the embodiment, a linear spline regression is further adopted, that is, a linear regression with smooth connection at the node is constructed. In the present embodiment 3 (2), the selection of nodes is based on the principle of "the smaller the change, the better the formula", based on which, the inventors set three nodes of spline regression based on a large number of studies, i.e., 0.5 ng / ml and 13 ng / ml.

[0538] Based on a thorough analysis of Example 3, for cases where the intercept does not include 0, the formula of spline regression is used, and for the conversion between Roche AMH and Kang's chemiluminescent AMH, for cases where the intercept includes 0, the formula of Passing-Bablok regression is used.

[0539] Table 11. Four transformations using spline regression when the 95% CI of the intercept using Passing-Bablok regression does not include 0.

[0540]

[0541] PD01059

[0542]

[0543] Recently, AMH has been considered to be the driving force for regulating ovarian reserve and PCOS. However, there are large differences between different AMH detection methods, which has significantly hindered the application of AMH-related artificial intelligence tools. More importantly, the huge differences between detection methods have also led to multiple AMH examinations between different testing centers or hospitals. The inventors of this application first proposed the idea of ​​using Passing-Bablok+ spline regression to convert AMH concentration from one detection method to another, and through in-depth research, confirmed the most suitable method and system for data standardization or conversion of the three detection methods, which is convenient for practical application.

[0544] The idea of ​​the Passing-Bablok regression + spline regression method of the present application is to first apply Passing-Bablok regression to detect whether there are systematic differences, that is, to see whether the 95% CI of the intercept of the Passing-Bablok regression result includes 0. If there are no systematic differences, that is, the intercept includes the value 0, the Passing-Bablok regression method can be used to convert the AMH detection between the two platforms. If there are systematic differences, spline regression will be used. In the spline regression process, the AIC method can be applied to the selection of the number of nodes. The smaller the AIC, the smaller the variation. And the specific values ​​of these nodes are fully studied and adjusted in the embodiments based on the calculated conversion values ​​and their original values, and finally the most suitable inspection data are confirmed.

[0545] Three AMH detection methods were used in the study of this embodiment, and the applicant found that the detection values ​​of some individual samples using the Roche AMH detection method were much lower than those of the other two platforms. Interference caused by biotin is considered to be the key reason for the occasional low values ​​using the Roche AMH detection method. For the AMH test kit, Roche chose to use a biotinylated AMH secondary antibody, which introduced the possibility of biotin interference in the design, while the other two did not. A small number of patients who visit ART clinics often take health products or drugs containing biotin, such as vitamin B7, vitamin H, coenzyme R, etc. For people who take health products containing biotin, biotin will be present in their blood and bind to the avermectin magnetic beads, which will reduce the binding points of the avermectin magnetic beads and reduce the AMH measurement value of the Roche platform.

[0546] In this embodiment, the inventors found that there was no systematic difference between the Roche method and the Kangrun chemiluminescence method, the 95% CI of the intercept covered 0, and there was only a proportional difference. The three commercial AMH detection methods performed well in accuracy and precision. In our study, the conversion of the three detection methods allowed one AMH detection method to be changed to another without additional examination. Therefore, it may contribute to the application of AMH-related tools and reduce the cost of patients receiving artificial reproductive technology when visiting different clinics.

[0547] Example 4

[0548] For E2 testing, serum samples were donated by patients attending the Reproductive Center of Peking University Third Hospital. The subjects gave written informed consent to donate their remaining serum for research after the original blood test. This experiment was ethically approved by the Institutional Review Board of Peking University Third Hospital, China.

[0549] E2 Detection

[0550] The same serum samples were frozen and thawed and tested on two different manufacturers' SIEMENS Immulite2000 automated immunoassay systems (Siemens Medical Diagnostics, Shanghai, China) and KANGRUN Keaser6600 automated immunoassay systems (KANGRUN Inc., Guangzhou, Guangdong, China). The E2 assays on both automated systems were performed once according to the manufacturer's recommendations. All E2 analysis results were reported in pmol / L. The quality control of the E2 assay was provided by Bio-RAD Laboratories (Lyphochek Immunoassay Plus Control, Trilevel, Catalog No. 370, Lot No. 40370). The coefficient of variation of the three-level control was less than 10%.

[0551] 4.1 Passing-Bablok regression results

[0552] Passing-Bablok regression was used to construct the linear relationship between Siemens E2 test values ​​and Kangrun E2 test values. Figure 2 The fitting results are shown. Figure 2 In the figure, the black dotted line reflects the situation where the two measured values ​​are completely consistent, and the black straight line fitting line is the Passing-Bablok regression result. It can be seen from the figure that the linear relationship between the two measured values ​​is somewhat different from the dotted line, and this difference needs to be statistically tested.

[0553] Table 12 shows the analysis results of Passing-Bablok regression. The results show that the 95% confidence interval of the intercept estimate does not include 0, indicating that the intercept is statistically different from 0; the 95% confidence interval of the slope estimate does not include 1, indicating that the slope is statistically different from 1. The intercept mainly reflects the systematic difference. Since the 95% confidence interval does not include 0, it indicates that there is a systematic difference between the two measurement results. The slope mainly reflects the proportion difference. Since the 95% confidence interval does not include 1, it indicates that there may be a proportion difference between the two measurement results.

[0554] Table 12 Passing-Bablok regression results

[0555]

[0556] Figure 3 Bland-Altman plots are presented. Bland-Altman plots complement the Passing-Bablok regression to graphically explore the agreement between two measurements. Figure 3In the figure, the horizontal axis is the mean of the measured values ​​of Kangrun E2 test value and Siemens E2 test value, and the vertical axis is the difference between Kangrun E2 test value and Siemens E2 test value. The black solid line in the figure is the reference line, i.e., the value of 0. The middle solid line and dotted line represent the mean of the difference and its 95% confidence interval, respectively. The upper and lower lines of the gray box (i.e., the gray area in the figure) are the 95% distribution range of the difference (mean ± 1.96 × standard deviation), respectively. This range is also called the consistency limit. In theory, if the distribution of the difference follows a normal distribution, 95% of the differences should be within the consistency limit.

[0557] The higher the consistency of the two measured values, the closer the middle solid line representing the mean difference is to the reference line (the gray solid line representing the mean difference is 0). According to the number of data points outside the 95% consistency limit and the maximum difference within the consistency limit, and the clinical acceptability, the consistency of the two methods to be evaluated is evaluated. In this embodiment, the number of observations beyond the consistency limit is large, and most of them are negative values, indicating that Siemens has a large difference with Kangrun in many test values.

[0558] Figure 4 A mountain-shaped graph of Siemens E2 test value and Kangrun E2 test value is given. The horizontal axis of the mountain-shaped graph is the difference between Siemens E2 test value and Kangrun E2 test value, and the vertical axis is the transformation of the cumulative distribution graph, that is, when the cumulative distribution probability is less than 50%, it is the cumulative distribution probability, and when it is greater than 50%, it is 1-cumulative distribution probability. If the two results are consistent, they should be symmetrically distributed with 0 as the center. Figure 4 It can be seen that the "Siemens E2 test value - Kangrun E2 test value" is obviously left-biased, indicating that the difference between "Siemens E2 test value - Kangrun E2 test value" is mostly less than 0, and the left tail is long, indicating that the difference between "Siemens E2 test value - Kangrun E2 test value" is particularly low. Figure 3 From the Bland-Altman plot, we can see that the "Siemens E2 test value - Kangrun E2 test value" has more low values.

[0559] 4.2 Spline regression results

[0560] (1) First determine the number of nodes and node values

[0561] The determination of the number of nodes is mainly based on clinical and statistical indicators. The statistical indicators mainly use indicators such as BIC and AIC. The lower the BIC and AIC values, the better the model. Finally, based on the statistical indicators of different numbers of nodes and taking into account the actual clinical situation, the model with two nodes was determined. The two nodes were 660pmol / L and 2030pmol / L, respectively.

[0562] (2) Fitting the model according to the number of nodes

[0563] Fitting a linear spline regression model with 2 nodes, the statistical analysis results are shown in Table 13:

[0564] Table 13 Spline regression fitting results of the relationship between Kangrun and Siemens

[0565]

[0566] According to the results in Table 13, the corresponding relationship between the Siemens E2 test value (y) and the Kangrun E2 test value (x) is established as follows:

[0567] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * g1 (Kangrun E2 test value - 660) + d8 * g2 (Kangrun E2 test value - 2030)

[0568] In the formula,

[0569] If the Kangrun E2 test value is <= 660pmol / L, then g1=0, g2=0, and the formula is:

[0570] Siemens E2 test value = a8 + b8 * Kangrun E2 test value

[0571] If the Kangrun E2 test value is >660pmol / L and <=2030pmol / L, then g1=1, g2=0, and the formula is:

[0572] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660)

[0573] If the Kangrun E2 test value is >2030pmol / L, then g1=1, g2=1, and the formula is:

[0574] Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660) + d8 * (Kangrun E2 test value - 2030)

[0575] Among them, a8 is selected from any value between -93.069 and -6.91207; b8 is selected from any value between 1.310892 and 1.569808; c8 is selected from any value between -0.8957 and -0.52322; d8 is selected from any value between -0.08505 and 0.085318.

[0576] When the Kangrun E2 test value is compared with the Siemens E2 test value as an example, the E2 test value measured by Kangrun is used as the independent variable x, and the E2 test value measured by Siemens is used as the dependent variable y. According to the relationship between the independent variable and the dependent variable in the data of the embodiment, a linear spline regression is further adopted, that is, a linear regression with smooth connections at the nodes is constructed. In this embodiment 4, the selection of nodes is based on the principle of "the smaller the change, the better the formula", based on which the inventors set the number of spline regression nodes based on a large number of studies, namely 660pmol / L and 2030pmol / L.

[0577] (3) Evaluation of fitting model

[0578] Table 14 shows the comparison of the fitting effects of spline regression and Passing-Bablok regression. It can be seen that the root mean square error, AIC, and AICc of spline regression are lower than those of Passing-Bablok regression, while R 2 and adjust R 2 Both are higher than linear regression, indicating that spline regression has better fitting effect.

[0579] Table 14 Comparison of fitting indices between spline regression and linear regression

[0580]

[0581] (4) Spline regression prediction

[0582] Figure 5 A prediction comparison between Passing-Bablok regression and spline regression is given. It can be seen that the prediction of Passing-Bablok regression is obviously biased in the low value range (Kangrun E2 test value is less than 1000), while spline regression gives a more reasonable prediction value.

[0583] Example 5

[0584] For the two tubes of frozen serum of each sample obtained in Example 1, one tube was detected by the Kangrun chemiluminescence method after thawing, and the other tube was detected by the Kangrun enzyme immunoassay (Anshlab) AMH detection method.

[0585] Generally speaking, if the AMH data of Kangrun chemiluminescence detection and the AMH detection data of Kangrun enzyme immune (Anshlab) AMH detection method conform to the overall relationship, the Passing-Bablok regression can be considered to transform the two data. Passing-Bablok regression uses a non-parametric method to fit the parameters a and b in the linear equation y=a+bx. If the 95% confidence interval (CI) of the intercept a does not contain 0 or the 95% confidence interval of the slope b does not contain 1, it indicates that there is a systematic or proportional difference between the two AMH test results.

[0586] If the detection results of the two platform methods do not satisfy the overall relationship, you can consider using spline regression to find the transformation between the detection data of the two platforms. Spline regression is a sheet-like polynomial function that is smoothly connected at the nodes. It can be a linear relationship (straight lines connected at the nodes), a quadratic relationship, or a cubic relationship. In spline regression, based on the relationship between the independent variable and the dependent variable, linear spline regression can be used to try to build a linear regression that is smoothly connected at the nodes.

[0587] To further determine the number of nodes, linear spline regression was used for practical convenience.

[0588] The number of nodes is determined based on the BIC indicator. The lower the BIC, the better the model fit. We specified 1-5 nodes respectively. The results showed that the model fit was optimal when there were 3 nodes. The results are as follows: Figure 6 shown.

[0589] The three nodes are 2 ng / mL, 3.7 ng / mL, and 6.9 ng / mL. Considering the convenience of practical application, we round them off and set the three nodes to 2 ng / mL, 4 ng / mL, and 7 ng / mL. The data is divided into 4 segments, such as Figure 7 shown.

[0590] The following is a linear spline regression fit

[0591] The fitting effect of 3-node straight line spline regression is as follows:

[0592] R2 and corrected R2 were 0.9326 and 0.9322 respectively, and AIC and AICC were 713.816 and 713.936 respectively

[0593] Parameter estimates

[0594]

[0595] *Slope 1 represents the slope of the first segment of data, slope 2 represents the change in the slope of the second segment of data compared to the first segment of data; slope 3 represents the change in the slope of the third segment of data compared to the second segment of data; slope 4 represents the change in the slope of the fourth segment of data compared to the third segment of data. The formula summarized is as follows:

[0596] The formula for converting Kangrun chemiluminescence AMH (X) to Kangrun chemiluminescence AMH (Y) is:

[0597] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 * (Kangrun chemiluminescent AMH-2) + d7 * g2 * (Kangrun chemiluminescent AMH-4) + e7 * g3 * (Kangrun chemiluminescent AMH-7)

[0598] Wherein, a7 is selected from any value in the range of -0.31000 to 0.20149, b7 is selected from any value in the range of 0.73863 to 1.10064, c7 is selected from any value in the range of -0.08941 to 0.51000, d7 is selected from any value in the range of 0.08121 to 0.54582, and e7 is selected from any value in the range of -0.70493 to -0.42738;

[0599] further,

[0600] If Kangrun chemiluminescent AMH <= 2 ng / ml, then g1 = 0, and the formula is:

[0601] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH;

[0602] If Kangrun chemiluminescent AMH>2ng / ml and <=4ng / ml, then g1=1, g2=0, and the formula is:

[0603] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2);

[0604] If Kangrun chemiluminescent AMH>4 ng / ml and <=7 ng / ml, then g1=1, g2=1, and the formula is:

[0605] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2) + d7 * g2 (Kangrun chemiluminescent AMH-4)

[0606] If Kangrun chemiluminescent AMH>7 ng / ml, then g1=1, g2=1, g3=1, and the formula is:

[0607] Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7*Kangrun chemiluminescent AMH + c7*g1 (Kangrun chemiluminescent AMH-2) + d7*g2 (Kangrun chemiluminescent AMH-4) + e7*g3 (Kangrun chemiluminescent AMH-7).

[0608] Prediction effect Figure 8 Among them, the prediction effect of the first three segments of data less than 7 ng / mL is as follows Fig. 9 shown.

[0609] This application uses Passing-Bablok regression and spline regression to construct the conversion relationship between Kangrun and Siemens. The assumption of Passing-Bablok regression is that both test results are random. It is mainly to examine the consistency of the two test results, and compare the difference between the intercept and 0 and the difference between the slope and 1. If both tests are not statistically significant, the two test results can be considered to be consistent and the results can be interchanged. Otherwise, the two test results cannot be considered to be consistent, and the results cannot be interchanged. They need to be converted through other corresponding formulas. If the two test results have an obvious linear relationship, they can be converted through Passing-Bablok regression; if they are not linearly related, spline regression can be considered. Both Passing-Bablok regression and spline regression can give clear conversion formulas.

[0610] Although the embodiments of the present invention are described above, the present invention is not limited to the above specific embodiments and application fields, and the above specific embodiments are only illustrative and instructive, but not restrictive. A person skilled in the art can make many forms under the guidance of this specification and without departing from the scope of protection of the claims of the present invention, all of which belong to the protection of the present invention.

Claims

1. A method for converting hormone data detected by different detection methods, comprising: Obtaining a first detection value for the hormone from a first platform; Obtaining a second detection value for the hormone from a second platform; Passing-Bablok regression is used to fit the first detection value and the second detection value; Based on the Passing-Bablok regression fitting result, judging whether there is a systematic difference between the first detection value and the second detection value; wherein, judging whether there is a systematic difference between the first detection value and the second detection value based on the Passing-Bablok regression fitting result is to judge whether there is a systematic difference between the first detection value and the second detection value by calculating the 95% confidence interval of the intercept of the Passing-Bablok regression result; if the 95% confidence interval of the intercept includes 0, it is judged that there is no systematic difference between the first detection value and the second detection value; if the 95% confidence interval of the intercept does not include 0, it is judged that there is a systematic difference between the first detection value and the second detection value; If there is no systematic difference between the first detection value and the second detection value, the first detection value or the second detection value is converted based on the fitting formula of Passing-Bablok regression. If there is a systematic difference between the first detection value and the second detection value, the first detection value and the second detection value are fitted based on the spline regression, and the first detection value or the second detection value is converted based on the fitting formula of the spline regression; when the first detection value and the second detection value are fitted based on the spline regression, the nodes of the spline regression are confirmed by using inherent nodes and statistical indicators, and the statistical indicators are AIC (Akaike Information Criterion) or BIC (Bayesian Information Criterion) indicators; Wherein, the hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Roche AMH detection platform, and the second platform is the Kangrun chemiluminescent AMH detection platform, Passing-Bablok regression is used to convert the detection results of the Roche AMH detection platform and the Kangrun chemiluminescent AMH detection platform; The formula for converting Kangrun chemiluminescence AMH (X) to Roche AMH (Y) is: Roche AMH = a1 + b1 * Kangrun chemiluminescent AMH; Wherein a1 is any value selected from -0.0035 to 0.0303; b1 is any value selected from 0.85 to 0.8844; The formula for converting Roche AMH (X) to Kangrun chemiluminescent AMH (Y) is: Kangrun chemiluminescence AMH = a2 + b2 * Roche AMH; wherein a2 is any value selected from -0.0356 to 0.00393; b2 is any value selected from 1.1307 to 1.1765; or Wherein, the hormone is anti-Mullerian hormone (AMH), wherein when the first platform is the Beckman AMH detection platform and the second platform is the Roche AMH detection platform, spline regression is used to convert the detection results of the Beckman AMH detection platform and the Roche AMH detection platform; The formula for converting Roche AMH (X) to Beckman AMH (Y) is: Beckman AMH = a3 + b3 * Roche AMH + c3 * g1 (Roche AMH-1) + d3 * g2 (Roche AMH-9), In the formula, if Roche AMH <= 1 ng / mL, then g1 = 0, g2 = 0, and the formula is: Beckman AMH = a3 + b3 * Roche AMH If Roche AMH>1 ng / mL and <=9 ng / mL, then g1=1, g2=0, and the formula is: Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1) If Roche AMH>9 ng / mL, then g1=1, g2=1, and the formula is: Beckman AMH = a3 + b3 * Roche AMH + c3 * (Roche AMH-1) + d3 * (Roche AMH-9) wherein a3 is selected from any value between -0.28809 and 0.30461; b3 is selected from any value between 0.665236 and 1.430824; c3 is selected from any value between 0.073606 and 0.900514; d3 is selected from any value between -0.70344 and -0.42116; The formula for converting Beckman AMH (X) to Roche AMH (Y) is: Roche AMH = a4 + b4 * Beckmann AMH + c4 * g1 (Beckmann AMH-0.5) + d4 * g2 (Beckmann AMH-1) In the formula, if the Beckman AMH value is <= 0.5 ng / ml, then g1=0, g2=0, and the formula is: Roche AMH = a4 + b4 * Beckmann AMH If the Beckman AMH value is >0.5 ng / mL and <=1 ng / mL, then g1=1, g2=0, and the formula is: Roche AMH = a4 + b4 * Beckmann AMH + c4 * (Beckmann AMH - 0.5) If the Beckman AMH value is > 1 ng / mL, then g1 = 1, g2 = 1, and the formula is: Roche AMH = a4 + b4*Beckmann AMH + c4*(Beckmann AMH-0.5) + d4*(Beckmann AMH-1); a4 is selected from any value between -0.42466 and 0.41822; b4 is selected from any value between -0.11202 and 2.17363; c4 is selected from any value between -1.96122 and 1.21734; d4 is selected from any value between -0.49419 and 0.66553; or Wherein, the hormone is anti-Mullerian hormone (AMH), wherein the first platform is the Beckman AMH detection platform, and the second platform is the Kangrun chemiluminescent AMH detection platform, and spline regression is used to convert the detection results of the Beckman AMH detection platform and the Kangrun chemiluminescent AMH detection platform; The formula for converting Beckman AMH (X) to Kangrun chemiluminescence AMH (Y) is: Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*g1(Beckman AMH-1)+d5*g2(Beckman AMH-4)+e5*g3(Beckman AMH-13) In the formula, if the Beckman AMH value is less than or equal to 1 ng / ml, then g1=0, g2=0, g3=0, and the formula is: Kangrun chemiluminescence AMH=a5+b5*Beckmann AMH If the Beckman AMH value is >1 ng / mL and <=4 ng / mL, then g1=1, g2=0, g3=0, and the formula is: Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1) If the Beckman AMH value is >4 ng / mL and <=13 ng / mL, then g1=1, g2=1, g3=0, and the formula is: Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4) If the Beckman AMH value is >13 ng / mL, then g1=1, g2=1, g3=1, and the formula is: Kangrun chemiluminescence AMH=a5+b5*Beckman AMH+c5*(Beckman AMH-1)+d5*(Beckman AMH-4)+e5*(Beckman AMH-13), Wherein, a5 is selected from any value between -0.24101 and 0.26258; b5 is selected from any value between 0.72643 and 1.42037; c5 is selected from any value between -0.78107 and 0.04326; d5 is selected from any value between 0.29488 and 0.57880; e5 is selected from any value between -0.71455 and -0.47130; Among them, the formula for converting Kangrun chemiluminescence AMH (X) to Beckman AMH (Y) is: Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * g1 (Kangrun chemiluminescence AMH-0.5) + d6 * g2 (Kangrun chemiluminescence AMH-13); In the formula, if the Kangrun chemiluminescence AMH value is less than or equal to 0.5 ng / ml, then g1=0, g2=0, and the formula is: Beckman AMH=a6+b6*Kangrun chemiluminescence AMH If the Kangrun chemiluminescent AMH value is >0.5 and <=13 ng / ml, then g1=1, g2=0, and the formula is: Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5) If the Kangrun chemiluminescent AMH value is >13 ng / ml, then g1=1, g2=1, and the formula is: Beckman AMH = a6 + b6 * Kangrun chemiluminescence AMH + c6 * (Kangrun chemiluminescence AMH-0.5) + d6 * (Kangrun chemiluminescence AMH-13); Wherein, a6 is selected from any value between -0.45082 and 0.54656; b6 is selected from any value between -0.35329 and 1.82169; c6 is selected from any value between -0.63742 and 1.56010; d6 is selected from any value between -0.52283 and -0.22757; or Wherein, the hormone is anti-Mullerian hormone (AMH), wherein the first platform is Kangrun chemiluminescent AMH detection platform, and the second platform is Kangrun enzyme-immune (Anshlab) AMH detection platform, spline regression is used to convert the detection results of Kangrun chemiluminescent AMH detection platform and Kangrun enzyme-immune (Anshlab) AMH detection platform; The formula for converting Kangrun chemiluminescence AMH (X) to Kangrun enzyme immune (Anshlab) AMH (Y) is: Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 * (Kangrun chemiluminescent AMH-2) + d7 * g2 * (Kangrun chemiluminescent AMH-4) + e7 * g3 * (Kangrun chemiluminescent AMH-7) Wherein, a7 is selected from any value in the range of -0.31000 to 0.20149, b7 is selected from any value in the range of 0.73863 to 1.10064, c7 is selected from any value in the range of -0.08941 to 0.51000, d7 is selected from any value in the range of 0.08121 to 0.54582, and e7 is selected from any value in the range of -0.70493 to -0.42738; further, If Kangrun chemiluminescent AMH <= 2 ng / ml, then g1 = 0, and the formula is: Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH; If Kangrun chemiluminescent AMH>2ng / ml and <=4ng / ml, then g1=1, g2=0, and the formula is: Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2); If Kangrun chemiluminescent AMH>4 ng / ml and <=7 ng / ml, then g1=1, g2=1, and the formula is: Kangrun enzyme immunoassay (Anshlab) AMH = a7 + b7 * Kangrun chemiluminescent AMH + c7 * g1 (Kangrun chemiluminescent AMH-2) + d7 * g2 (Kangrun chemiluminescent AMH-4) If Kangrun chemiluminescent AMH>7 ng / ml, then g1=1, g2=1, g3=1, and the formula is: Kangrun Enzyme Immunoassay (Anshlab) AMH = a7 + b7* Kangrun Chemiluminescent AMH + c7*g1 (Kangrun Chemiluminescent AMH-2) + d7*g2 (Kangrun Chemiluminescent AMH-4) + e7*g3 (Kangrun Chemiluminescent AMH-7); or Wherein, the hormone is estradiol E2, wherein, when the first platform is the Kangrun estradiol detection platform and the second platform is the Siemens estradiol detection platform, spline regression is used to convert the detection results of Kangrun estradiol and Siemens estradiol; Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * g1 (Kangrun E2 test value - 660) + d8 * g2 (Kangrun E2 test value - 2030) In the formula, If the Kangrun E2 test value is <= 660pmol / L, then g1=0, g2=0, and the formula is: Siemens E2 test value = a8 + b8 * Kangrun E2 test value If the Kangrun E2 test value is >660pmol / L and <=2030pmol / L, then g1=1, g2=0, and the formula is: Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660) If the Kangrun E2 test value is >2030pmol / L, then g1=1, g2=1, and the formula is: Siemens E2 test value = a8 + b8 * Kangrun E2 test value + c8 * (Kangrun E2 test value - 660) + d8 * (Kangrun E2 test value - 2030) Among them, a7 is selected from any value between -93.069 and -6.91207; b7 is selected from any value between 1.310892 and 1.569808; c7 is selected from any value between -0.8957 and -0.52322; d7 is selected from any value between -0.08505 and 0.085318.

Images