Artificial intelligence methods and systems for determining endometriosis status

Abstract

Artificial Intelligence methods and systems use a training dataset comprising clinical data and, optionally, biomarker data, to train a machine learning algorithm to produce a classification model that classifies endometriosis status in a subject. This can include a presence / absence, stage, risk level, or likelihood of endometriosis for the subject. The determination of status may be used to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

Description

ARTIFICIAL INTELLIGENCE METHODS AND SYSTEMS FOR DETERMINING ENDOMETRIOSIS STATUSBACKGROUND

[0001] Methods of artificial intelligence, including machine learning, are being brought to bear to analyze diagnostic questions. This is due, in part, to the ability of these tools to handle large datasets and to interpret natural language.

[0002] Endometriosis is a common condition affecting women of pubescent and reproductive age. Its diagnosis currently involves generally invasive procedures.

[0003] The disease is thought to be caused by endometrial tissue which migrates from its normal position lining the uterus to other parts of the body, primarily within the abdominal cavity. The ovaries and gut wall are commonly affected. The displaced endometrial tissue, like that in its normal position, grows and declines according to the menstrual cycle as a result of the actions of the ovarian hormones. Endometriosis may cause many symptoms including, but not limited to, abdominal pain, gastrointestinal upset, excessive bleeding, infertility and menstrual disturbance.

[0004] Women experiencing recurrent pelvic pain or infertility may be suspected to have endometriosis. Endometriosis, an inflammatory disorder in which endometrial cells proliferate outside the uterus, affects nearly 10% of reproductive age women. It is seen in 50-60% of reproductive aged women with chronic pelvic pain and in up to 50% of women with infertility. Despite its prevalence, endometriosis often goes undiagnosed for years. The average time from the onset of symptoms to a correct diagnosis can range from 5-10 years. The disease can be difficult to recognize based on patients’ descriptions of symptoms, especially at early stages, and the definitive diagnosis of the condition presently requires laparoscopic examination, a surgical procedure.

[0005] Women with endometriosis can present with a number of symptoms including pelvic pain, infertility, dysmenorrhea, dyspareunia and abnormal menstruation. However, many other conditions share one or more of these symptoms. This makes differential diagnosis of endometriosis a challenge. CA125 has been used as a biomarker for endometriosis. However, for CA125 tests, the area under the curve is about 0.6. Increases in AUC would improve performance of any test in the clinic. Laparoscopy is the current “gold standard” approach forvisual confirmation of endometriosis pathology and collection of lesion tissue for histological analysis.SUMMARY

[0006] Provided herein are artificial intelligence and machine learning methods for determining endometriosis status in a subject. The method involves creating a model for endometriosis status by assembling a data structure including, for a plurality of subjects, some of whom have endometriosis, and some of whom do not, responses to questionnaires including items concerning women’s health. Optionally, the data structure can include measures of one or more biomarkers from the subjects. The data structure is used to train a machine learning algorithm to classify endometriosis status of a subject. These include, for example, use of only clinical data, use of clinical and protein biomarker data, use of clinical and miRNA data, use of clinical, protein and miRNA data, and clinical and other types of biomarker data.

[0007] This disclosure contemplates using a variety of data sources in preparation of a training dataset of test dataset to be used in machine learning methods.

[0008] In another aspect, the model is used to classify the endometriosis status of a subject, using responses to the questionnaire and, optionally, biomarker data.

[0009] This disclosure addresses, among other things, a need in the art for minimally- invasive, accurate and more efficient methods of detecting, diagnosing, and monitoring endometriosis. Noninvasive procedures typically do not involve any incisions into the body or removal of tissue, e.g., physical examination or responses to questionnaire items. Minimally invasive procedures can involve small incision or slight penetration of body cavities, e.g., collection of blood samples. Highly invasive procedures can involve significant incisions, extensive tissue manipulation and a high level of body intrusion, such as many kinds of surgery.

[0010] In one aspect, a method of non-invasively testing or screening for endometriosis is provided, comprising: (a) obtaining clinical data comprising a plurality of responses provided by a female subject to a questionnaire comprising a series of targeted questions or surveys; (b) processing the clinical data into a data structure or format that is conducive or compatible for processing with biomarker data; and (c) applying a trained machine learning model to at least the processed clinical data, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the subject, wherein the set of metrics is useable to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0011] In some instances of the method, (a) further comprises obtaining the biomarker data from a biological sample collected from the subject, and (b) further comprises processing the biomarker data into another data structure or format that is conducive or compatible for processing with the clinical data. In some instances of the method, (c) comprises applying the trained machine learning model to (1) the processed clinical data and (2) the processed biomarker data to improve an accuracy of the set of metrics.

[0012] In some cases, the data structure can comprise “biomarker data,” that is, data pertaining to molecules in a biological sample. The biological sample can comprise a fluid sample, a blood sample, a saliva sample, or a menstrual blood sample. In some cases, the biomarker data comprises miRNA data associated with one or more miRNAs. The one or more miRNAs may be selected from the group consisting of hsa-miR-30b-5p, hsa-miR-19b-3p, hsa- miR-199a-5p, hsa-miR-125b-5p, hsa-let-7a-5p, hsa-miR-199a-3p, hsa-miR-99b-5p, hsa-miR- 20a-5p, hsa-miR-23a-3p, hsa-miR-339-3p, hsa-miR-24-3p, hsa-miR-145-5p, hsa-miR-3615, hsa- miR-185-5p, hsa-miR-25-3p, hsa-miR-484, hsa-miR-142-3p, hsa-miR-3613-5p, hsa-miR-15b- 5p, hsa-miR-451a, hsa-miR-186-5p, hsa-miR-18a-5p, hsa-miR-501-3p, hsa-miR-141-3p, hsa- miR-29b-3p, hsa-miR-22-3p, hsa-miR-150-5p, hsa-let-7d-5p, hsa-miR-122-5p, hsa-miR-182-5p, hsa-miR-98-5p, hsa-let-7b-5p, hsa-miR-342-3p, hsa-miR-628-3p, hsa-miR-155-5p, hsa-miR- 130a-3p, hsa-miR-26a-5p, hsa-miR-324-5p, hsa-let-7e-5p, and hsa-let-7f-5p. In some cases, the biomarker data comprises protein quantification data associated with one or more proteins. In some cases, the one or more proteins comprises CA-125.

[0013] In some cases, the series of targeted questions or surveys are designed to collect clinical data from the subject, e.g., data relating to demographics, reproductive history, medical history, abdominal and pelvic surgical history, symptoms that the subject is currently experiencing, pain level and severity, surgical data and surgical procedures performed, histopathology reports, concomitant medications, or fasting status. In some instances of the method, (b) further comprises extracting a set of clinically relevant factors from the list, for analysis by the trained machine learning model. In some cases, the set of clinically relevant factors relates to one or more of the following: a pain severity score experienced by the subject in a previous month; age of first menarche; previous pregnancy (if applicable); body mass index (BMI); fibroids (if applicable and medical history); ovarian cysts (if applicable and medical history); symptom duration over past 10 years; heavy or irregular periods (current symptom experienced); infertility or history of infertility (if applicable), dysmenorrhea (current symptomexperienced); pelvic pain (current symptom experienced); painful intercourse, urination, bowel movement (current symptom experienced); or family history of endometriosis (if applicable). In some cases, at least some of the targeted questions or surveys are designed to establish or define a common terminology, framework or workflow for identifying, detecting or assessing dysmenorrhea and its symptoms, by healthcare providers and for patients. In some cases, the plurality of responses comprises one or more scaled scores. The one or more scaled scores may relate to a pain level or severity. In some cases, the plurality of responses comprises one or more binary yes / no responses. In some cases, the plurality of responses comprises one or more textual responses that are indicative of symptom(s), pain or discomfort that the subject is experiencing. In some cases, the one or more textual responses comprises free-form text. In some cases, the plurality of responses comprises one or more multiple-choice question responses. In some cases, the plurality of responses comprises one or more numerical values. At least one of the numerical values may relate to a severity / scale, time duration, frequency and / or date(s) of one or more physical conditions or symptoms experienced by the subject. In some cases, the series of targeted questions or surveys are presented to the subject via a graphical user interface (GUI), and wherein the clinical data is obtained when the subject inputs the plurality of responses via the GUI.

[0014] In some cases, the trained machine learning model is trained using a training dataset comprising of both clinical data and biomarker data from a plurality of test subjects. In some cases, the clinical data is collected when the plurality of test subjects respond to the series of targeted questions or surveys. In some cases, the clinical data is collected from the plurality of test subjects at a plurality of time points spanning from pre-surgery to post-surgery. In some cases, the training dataset further comprises a surgical confirmation or histopathology report of whether each of the plurality of test subjects has endometriosis or non-endometriosis, performed within a time period after the test subjects have completed the series of targeted questions or surveys. In some cases, the trained artificial intelligence or machine learning model is trained on the clinical data and the miRNA data. In some cases, the trained machine learning model is trained on the clinical data and the protein quantification data. In some cases, the trained machine learning model is trained on the clinical data, the miRNA data, and the protein quantification data. In some cases, the trained machine learning model is configured to fit the clinical data and the biomarker data from the training dataset. In some cases, the trained machine learning model is configured to (a) collectively process (1) categorical and / or binary variables within the clinical data and (2) continuous expressions or variables in the biomarker data, and (b)assign a first set of weights to the categorical and / or binary variables and a second set of weights to the continuous expressions or variables. In some cases, the trained machine learning model comprises one or more of the following: a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient -boosted classifier, or another supervised or unsupervised machine learning algorithm. In some cases, the linear regression comprises LASSO (least absolute shrinkage and selection operator). In some cases, a performance metric of the trained machine learning model is improved by at least 10% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data. In some cases, a performance metric of the trained machine learning model is improved by at least 20% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data. In some cases, a performance metric of the trained machine learning model is improved by at least 30% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data. In some cases, the performance metric comprises Area Under the Curve (AUC), and wherein the trained machine learning model has an AUC greater than about 0.75, and the other machine learning model has an AUC less than about 0.75. In some cases, the trained machine learning model has an AUC greater than about 0.8. In some cases, the other machine learning model has an AUC ranging from about 0.6 to about 0.7. In some cases, the performance metric further comprises one or more of the following: sensitivity, specificity, positive predictive value (PPV), or negative predictive value (NPV). In some cases, the trained machine learning model has a sensitivity of at least about 70%, a specificity of at least about 70%, a PPV of at least about 80%, and an NPV of at least about 68%. In some cases, the method further comprises providing the set of metrics in a report to a healthcare provider or laboratory. In some cases, the method further comprises providing the report via a telehealth platform for establishing one or more pathways for effecting the specific course of treatment or prophylaxis for endometriosis for the subject. In some embodiments, the classifier has one or more of a sensitivity, specificity, AUC, positive predictive value or negative predictive value of at least any of 70%, 80%, 90%, 95% or 98%. Accordingly, a classifier using only clinical data has a higher AUC (e.g., at least 0.8) than CA125 alone (AUC about 0.6).

[0015] In another aspect, a data analytics module is provided comprising one or more processors that are configured to execute a set of instructions for performing any of the methods or steps, as described above and elsewhere herein. In some instances, the data analytics module is implemented in a cloud-based environment.

[0016] A further aspect is directed to a method comprising: aggregating a plurality of training datasets from a plurality of sources to generate a combined dataset, wherein the plurality of training datasets comprise (1) clinical data from one or more clinical sources and (2) biomarker data from biological samples processed by one or more laboratory facilities for a plurality of test subjects; using the combined dataset to train a machine learning model; applying the trained machine learning model to at least clinical data associated with a patient, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the patient; and using in part the set of metrics to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0017] In some cases, the clinical data comprises structured data. In some cases, the structured data comprises categorical, textual, binary, or numerical values. In some cases, the biomarker data comprises continuous expressions or variables associated with one or more biomarkers. In some instances of the method, aggregating the plurality of training datasets comprises applying one or more pre-processing or formatting steps to the clinical data and the biomarker data, such that the combined dataset is configured for training the machine learning model. In some cases, the one or more pre-processing or formatting steps comprise using an algorithm to remove outliers from the clinical data and / or impute missing data fields in the clinical data. In some cases, the one or more pre-processing or formatting steps comprise using an algorithm to remove outliers from the biomarker data and / or normalize the biomarker data. In some cases, the one or more clinical sources comprise electronic health records (EHRs) or electronic medical records (EMRs). In some cases, the one or more clinical sources comprise databases maintained by one or more healthcare providers. In some cases, the clinical data comprises a plurality of responses provided by the test subjects or the patient to a questionnaire comprising a series of targeted questions or surveys. In some cases, the series of targeted questions or surveys are designed to collect information the test subjects or the patient from a list comprising of: demographics, reproductive history, medical history, abdominal and pelvic surgical history, symptoms currently being experienced, pain level and severity, surgical data and surgical procedures performed, histopathology reports, concomitant medications, or fasting status. In some cases, a set of clinically relevant factors is extracted from the list to train the machine learning model, or used in part by the trained machine learning model to generate the report. In some cases, the set of clinically relevant factors relates to one or more of the following: a pain severity score experienced by the test subjects or the patient in a previous month; age of first menarche; previous pregnancy (if applicable); body mass index (BMI);fibroids (if applicable and medical history); ovarian cysts (if applicable and medical history); symptom duration over past 10 years; heavy or irregular periods (current symptom experienced); infertility or history of infertility (if applicable), dysmenorrhea (current symptom experienced); pelvic pain (current symptom experienced); painful intercourse, urination, bowel movement (current symptom experienced); or family history of endometriosis (if applicable).

[0018] In some cases, at least some of the plurality of training datasets are aggregated using an application programming interface (API) or a web-based interface. In some cases, wherein at least some of the plurality of training datasets are stored in a cloud database. In some cases, the machine learning model is trained in a cloud-based environment. In some cases, the trained machine learning model is applied to at least the clinical data in a cloud-based environment or a local computing environment.

[0019] In some cases, the biomarker data comprises at least one of miRNA data or protein quantification data. In some cases, the miRNA data is associated with one or more miRNAs selected from the group consisting of hsa-miR-30b-5p, hsa-miR-19b-3p, hsa-miR-199a-5p, hsa- miR-125b-5p, hsa-let-7a-5p, hsa-miR-199a-3p, hsa-miR-99b-5p, hsa-miR-20a-5p, hsa-miR-23a- 3p, hsa-miR-339-3p, hsa-miR-24-3p, hsa-miR-145-5p, hsa-miR-3615, hsa-miR-185-5p, hsa- miR-25-3p, hsa-miR-484, hsa-miR-142-3p, hsa-miR-3613-5p, hsa-miR-15b-5p, hsa-miR-451a, hsa-miR-186-5p, hsa-miR-18a-5p, hsa-miR-501-3p, hsa-miR-141-3p, hsa-miR-29b-3p, hsa-miR- 22-3p, hsa-miR-150-5p, hsa-let-7d-5p, hsa-miR-122-5p, hsa-miR-182-5p, hsa-miR-98-5p, hsa- let-7b-5p, hsa-miR-342-3p, hsa-miR-628-3p, hsa-miR-155-5p, hsa-miR-130a-3p, hsa-miR-26a- 5p, hsa-miR-324-5p, hsa-let-7e-5p, and hsa-let-7f-5p. In some cases, the protein quantification data is associated with one or more proteins comprising of CA-125. In some cases, the trained machine learning model comprises one or more of the following: a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient-boosted classifier, or another supervised or unsupervised machine learning algorithm.

[0020] Another aspect of the present disclosure provides a non-transitory computer readable medium comprising machine executable code that, upon execution by one or more computer processors, implements any of the methods above or elsewhere herein.

[0021] Another aspect of the present disclosure provides a system comprising one or more computer processors and computer memory coupled thereto. The computer memory comprisesmachine executable code that, upon execution by the one or more computer processors, implements any of the methods above or elsewhere herein.

[0022] Additional aspects and advantages of the present disclosure will become readily apparent to those skilled in this art from the following detailed description, wherein only illustrative embodiments of the present disclosure are shown and described. As will be realized, the present disclosure is capable of other and different embodiments, and its several details are capable of modifications in various obvious respects, all without departing from the disclosure. Accordingly, the drawings and description are to be regarded as illustrative in nature, and not as restrictive.INCORPORATION BY REFERENCE

[0023] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference. To the extent publications and patents or patent applications incorporated by reference contradict the disclosure contained in the specification, the specification is intended to supersede and / or take precedence over any such contradictory material.BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings (also “Figure” and “FIG.” herein), of which:

[0025] FIG. 1 illustrates a sample processing and analysis workflow, in accordance with some embodiments;

[0026] FIG. 2 illustrates a screening model that utilizes in part clinical data and miRNA data, in accordance with some embodiments;

[0027] FIG. 3 illustrates charts showing the performance characteristics of the screening model of FIG. 2, in accordance with some embodiments;

[0028] FIG. 4 illustrates a screening model that utilizes in part clinical data and protein quantification data, in accordance with some embodiments;

[0029] FIG. 5 illustrates charts showing the performance characteristics of the screening model of FIG. 4, in accordance with some embodiments;

[0030] FIG. 6 illustrates a workflow for collection of clinical data for training the screening model, in accordance with some embodiments. Study procedures include blood and saliva sample collection and collection of clinical data. The clinical portion of the model can be developed using the data from these detailed questions that follow. These subjects have confirmation of endometriosis or non-endometriosis via laparoscopy within weeks of completing the questionnaires.

[0031] FIG. 7 illustrates the time points at which various data (including clinical data and sample data) are collected, for purposes for training and validation, in accordance with some embodiments;

[0032] FIGs. 8-20 illustrate examples of a series of targeted questions or surveys that are used in a questionnaire, to generate the clinical data for training the model, as well as for assessment of new patients, in accordance with some embodiments;

[0033] FIG. 8 — Demographics (Pre-Surgery Visit)

[0034] FIG. 9 — Eligibility (Pre-Surgery Visit)

[0035] FIG. 10 — Reproductive History (Pre-Surgery Visit)

[0036] FIG. 11 — Medical History (Pre-Surgery Visit)

[0037] FIG. 12 — Abdominal & Pelvic Surgical History (Pre-Surgery Visit)

[0038] FIG. 13 — Current Symptoms (Pre-Surgery Visit)

[0039] FIG. 14 — EHP-5 (Pre-Surgery & Post-Surgery Visits)

[0040] FIG. 15 — Pain Symptom Score (Pre-Surgery & Post-Surgery Visits), including “Worst pelvic pain score during last period,” and “Worst pelvic pain score during last month.”

[0041] FIG. 16 — Vital Signs (Pre-Surgery Visit)

[0042] FIG. 17 - EPHect Form (Surgical Procedure)

[0043] FIG. 18 - Diagnosis (Surgical Procedure)

[0044] FIG. 19 — Subject Status (Post-Surgery Visit)

[0045] FIG. 20 — ConMeds (Pre-Surgery & Post-Surgery Visits)

[0046] FIG. 21 illustrates a model building workflow in accordance with some embodiments;

[0047] FIG. 22A illustrates performance charts for different types of models, in accordance with some embodiments;

[0048] FIG. 22B illustrates an Endo / Non-Endo distribution on CA-125 protein, in accordance with some embodiments;

[0049] FIG. 23 illustrates a platform in accordance with some embodiments; and

[0050] FIG. 24 schematically illustrates a computer system that is programmed or otherwise configured to implement methods provided herein.

[0051] FIG. 25 displays a plot of univariate predictive significance (in order from greatest to least significance from left to right) for particular clinical data features used in algorithms described herein. Explanations for the features, listed from left to right, are:V1_DMGR_AGE: AgeV1_VTLS_BMI: BMIC1_CALC_MHRPESP_FBRDS: Has the subject been diagnosed with any of the following conditions? (fibroids checked)V1_RPRH_PRVPRGYN: Has the subject ever been pregnant before?C1_CALC_SYMPTOM_PAIN1: Which of the following signs / symptoms is the subject experiencing? (Painful periods / dysmenorrhea checked)C1_CALC_RACE: RaceC1_CALC_SYMPTOM_PAIN2A: Which of the following signs / symptoms is the subject experiencing? (non-menstrual pelvic pain checked)V1_EPHS_LIFE: On Endometriosis Health Profile Questionnaire; Felt as though your symptoms are ruining your life?V1_EHPS_UNDSTAND: On Endometriosis Health Profile Questionnaire; Felt others do not understand what you’re going through?C1_CALC_SYMPTOM_PAIN2B: Which of the following signs / symptoms is the subject experiencing? (painful intercourse, painful urination, and / or painful bowel movements checked)C1_CALC_SYMPTOM_INFERT: Which of the following signs / symptoms is the subject experiencing? (infertility checked)Vl_EHP5_sum: EHP-5 questionnaire overall score (individual answers scored as: never = 0, rarely = 1, sometimes = 2, often = 3, always = 4). For overall score, add the total ofthe points for the questionnaire and divide by the total possible number of points (20) and multiply xlOO.C1_CALC_SYMDUR_ORD: How long has the subject been experiencing the above signs and symptoms?V1_PNSS_PAINLP: Pain Symptom Score Questionnaire; Worst pelvic pain score during last periodV1_RPRH_FAMILY_HISTORY: Does the subject report having a family history of endometriosis?V1_RPRH_INFERTYN: Has the subject experienced infertility?V1_EHP5_WALK: On Endometriosis Health Profile Questionnaire; Found it difficult to walk because of the pain?Vl_EHP5_M00D: On Endometriosis Health Profile Questionnaire; Had mood swings? C1_CALC_MHRPESP_OVRCYS: Ovarian Cysts: Has the subject been diagnosed with any of the following conditions? (ovarian cysts checked)V1_PNSS_PAINLM: Pain Symptom Score Questionnaire; Worst pelvic pain score during the last monthC1_CALC_MHRPESP_ADNMYSS: Adenomyosis: Has the subject been diagnosed with any of the following medical conditions? (adenomyosis checked)V1_EHP5_ APPEAR: On Endometriosis Health Profile Questionnaire; Felt your appearance has been affected?V1_RPRH_AGE_MENARCHE: Age of menarche (first period) C1_CALC_ETHNICITY: Ethnicity

[0052] FIG. 26 depicts a chart showing various conditions that can be confused with endometriosis based on which symptoms are being experienced by the subject. These conditions include: (1) Pelvic Inflammatory Disease (PID), Appendicitis, Ovarian cysts, Ovarian torsion, Ectopic pregnancy, Celiac disease, Irritable bowel syndrome (IBS), Fibromyalgia, or Malignancy (for pelvic pain); (2) Male factor, Anovulation, Luteal phase defect, Tubal factors, Ovarian insufficiency, Fibroids, PID, Hyper- or Hypothyroidism, or Polycystic ovarian syndrome (PCOS) (for infertility); (3) Adenomyosis, Leiomyomas, Fibroids, Ovarian cysts, PID, IBS, Endometrial polyps, Cervical stenosis, Musculoskeletal, or Increased prostaglandins (for dysmenorrhea); (4) Vaginismus, Vaginal atrophy, Vulvodynia, Vaginitis and vulvovaginitis, Pelvic adhesions, Trauma, Uterine prolapse, Cystitis, PID, or Uterine fibroids (for dyspareunia); and (5) Uterinepolyps, Fibroids, PID, PCOS, Cancer, Hyper- or hypothyroidism, Pituitary disorders, Excessive exercise, Medication side effect, or Abnormal pregnancy (for abnormal menstruation).DETAILED DESCRIPTION

[0053] While various embodiments of the invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions may occur to those skilled in the art without departing from the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed.

[0054] Whenever the term “at least,” “greater than,” or “greater than or equal to” precedes the first numerical value in a series of two or more numerical values, the term “at least,” “greater than” or “greater than or equal to” applies to each of the numerical values in that series of numerical values. For example, greater than or equal to 1, 2, or 3 is equivalent to greater than or equal to 1, greater than or equal to 2, or greater than or equal to 3.

[0055] Whenever the term “no more than,” “less than,” or “less than or equal to” precedes the first numerical value in a series of two or more numerical values, the term “no more than,” “less than,” or “less than or equal to” applies to each of the numerical values in that series of numerical values. For example, less than or equal to 3, 2, or 1 is equivalent to less than or equal to 3, less than or equal to 2, or less than or equal to 1.

[0056] The term “any of’ between a modifier and a sequence means that the modifier modifies each member of the sequence. So, for example, the phrase “at least any of 1, 2, or 3” means “at least 1, at least 2, or at least 3”.

[0057] The term “about” refers to a range that is 5% plus or minus from a stated numerical value within the context of the particular usage.

[0058] The term “real time” or “real-time,” as used interchangeably herein, generally refers to an event (e.g., an operation, a process, a method, a technique, a computation, a calculation, an analysis, a visualization, an optimization, etc.) that is performed using recently obtained (e.g., collected or received) data. In some cases, a real time event may be performed almost immediately or within a short enough time span, such as within at least 0.0001 millisecond (ms), 0.0005 ms, 0.001 ms, 0.005 ms, 0.01 ms, 0.05 ms, 0.1 ms, 0.5 ms, 1 ms, 5 ms, 0.01 seconds, 0.05 seconds, 0.1 seconds, 0.5 seconds, 1 second, or more. In some cases, a real time event may be performed almost immediately or within a short enough time span, such as within at most 1second, 0.5 seconds, 0.1 seconds, 0.05 seconds, 0.01 seconds, 5 ms, 1 ms, 0.5 ms, 0.1 ms, 0.05 ms, 0.01 ms, 0.005 ms, 0.001 ms, 0.0005 ms, 0.0001 ms, or less.

[0059] Described herein are non-invasive or minimally-invasive, accurate and more efficient methods of detecting, diagnosing, and monitoring endometriosis.Subjects

[0060] The methods and compositions described herein are applicable to human and nonhuman subjects, including veterinary subjects. Preferred subjects are "patients” — living humans that are receiving medical care for a disease or condition (e.g., endometriosis), or who are suspected of having such disease or condition or who are at risk of having such disease or condition. This includes persons with no defined illness who are being investigated for signs of pathology (e.g., endometriosis).

[0061] Preferred patients or subjects for the methods and compositions described herein are female patients that are at pubescent or post-pubescent ages, pre-menopausal, peri-menopausal, menopausal, or post-menopausal (as endometriosis may persist after menopause). As such, in general, the methods and compositions provided herein may be useful for female subjects within a large range of ages, generally over the age of 10 or over the age of 18. Often, the subject is under the age of 49. In some cases, the subject is premenopausal. In some cases, the subject is premenopausal and over 18. In some cases, the subject is premenopausal and under 49. The subject may be at any phase of the menstrual cycle, e.g., luteal or proliferative.

[0062] In some cases, a subject may be at risk of having endometriosis. A subject is “at risk” of endometriosis is their risk of endometriosis is greater than average risk for the population in which risk was measured, e.g., in the 51stpercentile, the 75thpercentile, the 80thpercentile, the 90thpercentile or the 95thpercentile. A subject at risk of having endometriosis, may, for example, have a family history of endometriosis, symptoms of endometriosis, or a past medical history of endometriosis. The endometriosis may be any stage of endometriosis. In some cases, the subject has, or is suspected of having, Stage I, II, III, or IV endometriosis. In some cases, the subject has, or is suspected of having, endometrioma. In some cases, the subject has endometriosis at any stage (e.g., Stage I-IV) but a method provided herein detects endometriosis as a general condition in the subject, without specifying the stage. In some cases, the subject has early-stage endometriosis. In some cases, the subject as Stage I / II endometriosis. In some cases, the subject has Stage III / IV endometriosis.

[0063] In some cases, a subject may be suspected of having endometriosis. Such a subject may display no symptoms of endometriosis. But in other cases, such subject may display symptoms of endometriosis such as dysmenorrhea, pain with bowel movements or urination, deep dyspareunia, chronic lower abdominal pain, chronic lower back pain, adnexal masses, infertility, or excessive bleeding. In some cases, the subject may be suspected of having endometriosis due to the results of a previous or concurrent test for endometriosis. In some cases, a subject is suspected of having endometriosis due to multiple factors. For example, the subject may be suspected of having endometriosis due to the presence of symptoms in an overall clinical context consistent with endometriosis.

[0064] In some cases, a subject may have, or be suspected of having, a non-endometriosis condition. In general, as used herein, the term “non-endometriosis condition” refers to an abnormal reproductive condition that is not endometriosis. Nonlimiting examples of a non- endometriosis condition include fibroids, leiomyomas, cysts, dermoid cysts, serous cystadenomas, cystadenomas, ovarian cysts, mucous cystadenomas, pelvic infection, teratoma, and / or paratubal cysts. In some cases, such subject may display symptoms of endometriosis such as dysmenorrhea, pain with bowel movements or urination, deep dyspareunia, chronic lower abdominal pain, chronic lower back pain, adnexal masses, infertility, or excessive bleeding. In some cases, the non-endometriosis condition is benign. In some cases, the non-endometriosis condition is malignant.

[0065] In some cases, the subject may be receiving a hormonal treatment. The hormonal treatment may include GnRH (gonadotrophin releasing hormone) agonists (with or without estrogen / progesterone replacement therapy or tibolone treatment) and antagonists, levonorgestrel-releasing intrauterine devices (e.g., Mirena), danazol, antiprogesterones, gestrinone, aromatase inhibitors, selective estrogen receptor modulators (SERMs), or selective progesterone receptor modulators (SPRMs).Differential Diagnosis

[0066] Endometriosis shares many signs and symptoms in common with other women’s conditions, including conditions related to reproductive health. These include for example, pelvic pain, infertility, dysmenorrhea, dyspareunia and abnormal menstruation. The methods of this disclosure allow differential diagnosis of endometriosis from among other conditions sharing one or more of these signs or symptoms. Such a diagnosis can inform a path of treatment, as such treatments may be different for these other conditions.Treatment

[0067] In some embodiments, the methods of the present disclosure include assigning or administering treatment to a patient having, at risk of developing, or suspected of having endometriosis. By detecting the clinical status of the patient using the classifiers described herein, the appropriate treatment can be assigned or administered to a patient suffering from endometriosis. These treatments can include, but are not limited to, hormone therapy, chemotherapy, immunotherapy, and surgical treatment. Similarly, the methods of the current disclosure can be used to assign or administer treatment to a patient with reduced fertility due to endometriosis. In this fashion, by determining the degree to which the patient's fertility has been reduced, through the detection of biomarkers found herein, the appropriate treatment can be assigned or administered. Relevant treatments include, but are not limited to, hormone therapy, chemotherapy, immunotherapy, and surgical treatment.

[0068] The present disclosure includes therapies (e.g., drug, surgical) for the treatment or prevention of endometriosis. A non-limitative list of known methods and materials for endometriosis treatment include, but are not limited to, pain killers, hormonal treatments, chemotherapy, and surgical treatments. In some cases, a patient or subject in which endometriosis is detected by a method provided herein, may undergo additional tests or procedures to confirm the endometriosis. For example, the patient or subject may have surgery (e.g., laparoscopic surgery) to further diagnose or characterize the endometriosis, e.g., the stage of endometriosis; or to treat the endometriosis. Pain killers used for the treatment of endometriosis include both simple analgesics, such as paracetamol, COX-2 inhibitors, aspirin, and other non-steroidal anti-inflammatory drugs well known in the art, and narcotic analgesics, such as morphine, codeine, oxycodone, and others well known in the art. Hormonal treatments include, but are not limited to, oral contraceptives, progestins (such as Dydrogesterone, Medroxyprogesterone acetate, Depot medroxyprogesterone acetate, Norethisterone, Levonorgestrel), progesterone and progesterone-like substances, GnRH agonists (such as leuprorelin, buserelin, goserelin, histrelin, deslorelin, nafarelin, and triptorelin), androgens and synthetic androgens like Danazol, GnRH agonists (e.g., Elagolix) (with or without estrogen / progesterone replacement therapy or tibolone treatment), levonorgestrel-releasing intrauterine devices (e.g., Mirena), danazol, antiprogesterones, gestrinone, selective estrogen receptor modulators (SERMs), or selective progesterone receptor modulators (SPRMs), and aromatase inhibitors. Surgical treatments include, but are not limited to, laparoscopic surgery,hysterectomy, and oophorectomy. Other treatments particularly well suited for use in the present disclosure are well known in the art. In some embodiments, the patient can be treated using a statin, including but not limited to, atorvastatin, cerivastatin, fluvastatin, lovastatin, mevastatin, pitavastatin, pravastatin, rosuvastatin and simvastatin.

[0069] In some aspects, the present disclosure provides for administration of a treatment to the subject to treat the endometriosis detected herein based on the classification generated using the algorithms described herein. Treatments for endometriosis include, but are not limited to, pain killers (e.g., NSAIDs), hormonal treatments, chemotherapy, and surgical treatments. Pain killers used for the treatment of endometriosis include both simple analgesics, such as paracetamol, COX-2 inhibitors, aspirin, and other non-steroidal anti-inflammatory drugs well known in the art, and narcotic analgesics, such as morphine, codeine, and oxycodone. Hormonal treatments include, but are not limited to, oral contraceptives, progestins, such as Dydrogesterone, Medroxyprogesterone acetate, Depot medroxyprogesterone acetate, Norethisterone, Levonorgestrel, and others well known in the art, progesterone and progesterone- like substances, GnRH agonists, such as leuprorelin, buserelin, goserelin, histrelin, deslorelin, nafarelin, triptorelin, and leuprolin, androgens and synthetic androgens like Danazol, GnRH antagonists, and aromatase inhibitors. Surgical treatments include, but are not limited to, laparoscopic surgery, hysterectomy, and oophorectomy.

[0070] In some aspects, a GnRH antagonist is administered to treat the endometriosis identified herein. A variety of antagonists of GnRH suitable for clinical administration, both peptide (goserelin acetate, buserelin, histrelin, deslorelin, nafarelin, and triptorelin, leuproreolin) and non-peptide (Elagolix / ABT-620, NBI-56418, see for e.g., Taylor et al. N Engl J Med. 2017 Jul 6;377(l):28-40), are available for second-line treatment of endometriosis in individuals with refractory endometriosis.

[0071] In one aspect, a method of non-invasively testing or screening for endometriosis is provided, comprising: (a) obtaining clinical data comprising a plurality of responses provided by a female subject to a questionnaire comprising a series of targeted questions or surveys; (b) processing the clinical data into a data structure or format that is conducive or compatible for processing with biomarker data; and (c) applying a trained machine learning model to at least the processed clinical data, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the subject, wherein the set of metrics is useable to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0072] In some instances of the method, (a) further comprises obtaining the biomarker data from a biological sample collected from the subject, and (b) further comprises processing the biomarker data into another data structure or format that is conducive or compatible for processing with the clinical data. In some instances of the method, (c) comprises applying the trained machine learning model to (1) the processed clinical data and (2) the processed biomarker data to improve an accuracy of the set of metrics.

[0073] In some cases, the biological sample comprises blood or saliva. In some cases, the biomarker data comprises miRNA data associated with one or more miRNAs. The one or more miRNAs may be selected from the group consisting of hsa-miR-30b-5p, hsa-miR-19b-3p, hsa- miR-199a-5p, hsa-miR-125b-5p, hsa-let-7a-5p, hsa-miR-199a-3p, hsa-miR-99b-5p, hsa-miR- 20a-5p, hsa-miR-23a-3p, hsa-miR-339-3p, hsa-miR-24-3p, hsa-miR-145-5p, hsa-miR-3615, hsa- miR-185-5p, hsa-miR-25-3p, hsa-miR-484, hsa-miR-142-3p, hsa-miR-3613-5p, hsa-miR-15b- 5p, hsa-miR-451a, hsa-miR-186-5p, hsa-miR-18a-5p, hsa-miR-501-3p, hsa-miR-141-3p, hsa- miR-29b-3p, hsa-miR-22-3p, hsa-miR-150-5p, hsa-let-7d-5p, hsa-miR-122-5p, hsa-miR-182-5p, hsa-miR-98-5p, hsa-let-7b-5p, hsa-miR-342-3p, hsa-miR-628-3p, hsa-miR-155-5p, hsa-miR- 130a-3p, hsa-miR-26a-5p, hsa-miR-324-5p, hsa-let-7e-5p, and hsa-let-7f-5p. In some cases, the biomarker data comprises protein quantification data associated with one or more proteins. In some cases, the one or more proteins comprises CA-125.Clinical Data

[0074] In some cases, the series of targeted questions or surveys are designed to collect information of the subject from a list comprising of: demographics, reproductive history, medical history, abdominal and pelvic surgical history, symptoms that the subject is currently experiencing, pain level and severity, surgical data and surgical procedures performed, histopathology reports, concomitant medications, or fasting status. In some instances of the method, (b) further comprises extracting a set of clinically relevant factors from the list, for analysis by the trained machine learning model. In some cases, the set of clinically relevant factors relates to one or more of the following: a pain severity score experienced by the subject in a previous month; age of first menarche; previous pregnancy (if applicable); body mass index (BMI); fibroids (if applicable and medical history); ovarian cysts (if applicable and medical history); symptom duration over past 10 years; heavy or irregular periods (current symptom experienced); infertility or history of infertility (if applicable), dysmenorrhea (current symptom experienced); pelvic pain (current symptom experienced); painful intercourse, urination, bowelmovement (current symptom experienced); or family history of endometriosis (if applicable). In some cases, at least some of the targeted questions or surveys are designed to establish or define a common terminology, framework or workflow for identifying, detecting or assessing dysmenorrhea and its symptoms, by healthcare providers and for patients. In some cases, the plurality of responses comprises one or more scaled scores. The one or more scaled scores may relate to a pain level or severity. In some cases, the plurality of responses comprises one or more binary yes / no responses. In some cases, the plurality of responses comprises one or more textual responses that are indicative of symptom(s), pain or discomfort that the subject is experiencing. In some cases, the one or more textual responses comprises free-form text. In some cases, the plurality of responses comprises one or more multiple-choice question responses. In some cases, the plurality of responses comprises one or more numerical values. At least one of the numerical values may relate to a severity / scale, time duration, frequency and / or date(s) of one or more physical conditions or symptoms experienced by the subject. In some cases, the series of targeted questions or surveys are presented to the subject via a graphical user interface (GUI), and wherein the clinical data is obtained when the subject inputs the plurality of responses via the GUI.

[0075] Clinical data can be collected through the use of a questionnaire that comprises a plurality of items for response. The plurality of items in a questionnaire for generating training data can include, for example, at least any of five, 10, 50, 100, 500, or 1000. The classifier generated using responses to these items may include all or a subset of the items. Accordingly, the questionnaire used in the testing phase also can include at least any of five, 10, 50, 100, 500, or 1000 items. In particular, a questionnaire can include any of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 items selected from, Demographics (e.g., date of birth, race, ethnicity), Eligibility (e.g., informed consent), Reproductive History (e.g., age of first menarche, number of pregnancies, number of live births, number of C-sections, number of live births, experience of infertility, fertility treatment, family history of endometriosis), Medical History (e.g., prior medical conditions), Abdominal & Pelvic Surgical History, Current Symptoms (e.g., related to dysmenorrhea), Endometriosis Health Profile 5 (EHP-5) items (e.g., items related to pain, control and powerlessness, emotional well-being, social support, self-image), Pain Symptom Score (e.g., “Worst pelvic pain score during last month.”), Vital Signs (e.g., height, weight, BMI), Endometriosis Phenome and Biobanking Harmonisation Project (EPHect) Form (e.g., items related to pain, depression and anxiety, menstrual history, fertility, medical and surgical history, medication use, personal information), Diagnosis (e.g., diagnosis of endometriosis), SubjectStatus (e.g., completion of questionnaire), and Concomitant medications (e.g., medication name, dose, frequency, route, indication, start date, end date).Model Building and Classification

[0076] Datasets collected by the methods disclosed herein can be used in building models to predict endometriosis status in a subject.

[0077] A categorical variable is a variable that can be defined by two or more, typically nonoverlapping, categorical states. For example, the categorical variable can be a pathological condition, and the categorical states can be status of the condition, such as different diagnoses, disease stage, risk of developing, likelihood of having, prognosis. Categorical states include status of the categorical variable or diagnosis of the pathological condition.

[0078] A measurement of a variable can be any combination of numbers and words. A measure can be any scale, including nominal (e.g., name, class or category, e.g., composite), ordinal (e.g., hierarchical order of categories), interval (distance between members of an order), ratio (interval compared to a meaningful “0”), or a cardinal number measurement that counts the number of things in a set. Measurements of a variable on a nominal scale indicate a name or category, e.g., category into which the sequencing read is classified. Measurements of a variable on an ordinal scale produce a ranking, such as “first”, “second”, “third”. Measurements on a ratio scale indicate, for example, any measure on a pre-defined scale, absolute number of reads, normalized or estimated numbers, as well as statistical measurements such as frequency, mean, median, standard deviation, or quantile. Measurements that involve quantification are typically determined at the ratio scale level.

[0079] Accordingly, in the case of endometriosis as a pathological condition, the status of the condition can be a nominal (e.g., present / absent; positive / negative; Stage I, Stage II, Stage III or Stage IV), an ordinal (e.g., likelihood (e.g., high, medium, low)), or a ratio (e.g., risk percentage / probability). Determining a categorical state of a pathological condition can also be referred to as “diagnosing” or “making a diagnosis.” The state can be expressed as any measure as described herein, e.g., a binary, a category (e.g., likely, neutral, and unlikely), or a probability (e.g., a continuous variable). Determining a categorical state of a pathological condition can also be referred to as “diagnosing” or “making a diagnosis.”

[0080] Training data sets for the creation of diagnostic models typically involve measures from a plurality of subjects (more generally, referred to as “instances”) from each of a pluralityof categorical states for a categorical variable. For example, the dataset can include at least five subjects, at least 10 subjects, at least 50 subjects, at least 100 subjects, or at least 200 subjects from each categorical state. Within these parameters, the total number of samples can be, for example, at least 20 subjects, at least 50 subjects, at least 100 subjects, at least 200 subjects, or at least 400 subjects. So for example, the dataset could comprise 25 subjects from a first categorical state and 75 subjects from a second categorical state, etc.

[0081] In certain embodiments, a plurality of instances among the collection of instances in the training set can be assigned to each of a plurality of different categorical states for a categorical variable. So, for example, given 100 instances, 50 might be assigned to a first categorical state and 50 might be assigned to a second categorical state for a categorical variable.

[0082] Using data from a plurality of subjects classified into the categorical states to be predicted, the machine learning algorithm generates a classifier that predicts, for an individual subjects, the categorical state, based on abundance data collected from the subjects.

[0083] In some cases, the trained machine learning model is trained using a training dataset comprising of both clinical data and biomarker data from a plurality of test subjects. In some cases, the clinical data is collected when the plurality of test subjects respond to the series of targeted questions or surveys. In some cases, the clinical data is collected from the plurality of test subjects at a plurality of time points spanning from pre-surgery to post-surgery. In some cases, the training dataset further comprises a surgical confirmation or histopathology report of whether each of the plurality of test subjects has endometriosis or non-endometriosis, performed within a time period after the test subjects have completed the series of targeted questions or surveys. In some cases, the trained machine learning model is trained on the clinical data and the miRNA data. In some cases, the trained machine learning model is trained on the clinical data and the protein quantification data. In some cases, the trained machine learning model is trained on the clinical data, the miRNA data, and the protein quantification data. In some cases, the trained machine learning model is configured to fit the clinical data and the biomarker data from the training dataset. In some cases, the trained machine learning model is configured to (a) collectively process (1) categorical and / or binary variables within the clinical data and (2) continuous expressions or variables in the biomarker data, and (b) assign a first set of weights to the categorical and / or binary variables and a second set of weights to the continuous expressions or variables.

[0084] In a classification or diagnostic step, a biological sample is collected from a subject is collected. With or without prior freezing, RNA is extracted from the sample within the specified time periods after collection as described herein. One or more forms of RNA, e.g., miRNA, are measured to produce a dataset. Using a classifier as described above, an operator can classify the categorical state of a particular categorical variable of a subject based on the dataset. The classifier can classify conditions according to any classification scheme useful to the operator.

[0085] Artificial Intelligence is a field concerned with using machines, e.g., computers, to mimic intelligence, including learning, reasoning, problem-solving, perception, and language understanding. Machine learning is a subset of artificial intelligence concerned with learning from data. Accordingly, references to machine learning are included in the concept of artificial intelligence, e.g., an artificial intelligence agent.Artificial Intelligence / Machine Learning

[0086] Datasets can be analyzed using traditional statistical methods, such as, multivariate analysis of variance multivariate regression, principal component analysis, and discriminant analysis. However, more complex datasets for example, those including multidimensional measures of each instance and a large number of instances, require computerized methods for analysis and, in particular, machine learning.

[0087] Machine learning algorithms are trained on training datasets to generate models that predict the categorical state of a sample based on subject data, such as expression data. Predicted categorical state can be translated into recommendations to a subject for actions to be taken, for example, health interventions, such as therapeutic treatments.

[0088] Methods of generating models to predict a categorical state can involve providing a training dataset on which a machine learning algorithm can be trained to develop one or more models to predict or infer a categorical state. The training dataset will include a plurality of training examples, typically for each of a plurality of subjects and typically in the form of a vector. Each training example will include a plurality of features or data points and, for each feature, data, e.g., in the form of numbers or descriptors. Where learning is to be supervised, the data will include a classification of the sample or subject into a category of a categorical variable to be inferred. For example, the categorical variable may be “endometriosis diagnosis” and the categories or classifications of this variable can be “present” and “absent”. For machine learning, the training examples can have at least 5, at least 10, at least 100, at least 1000, at least 10,000, different features. The features selected are those on which prediction will be based.

[0089] In some cases, the trained machine learning model comprises one or more of the following: a support vector machine (SVM), discriminant analyses (e.g., Bayesian classifier or Fischer analysis),, a linear regression, a quantile regression, a logistic regression, a random forest, recursive partitioning, a neural network, a gradient-boosted classifier, or another supervised or unsupervised machine learning algorithm. In some cases, the linear regression comprises LASSO (least absolute shrinkage and selection operator). In some cases, a performance metric of the trained machine learning model is improved by at least 10% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data. In some cases, a performance metric of the trained machine learning model is improved by at least 20% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data. In some cases, a performance metric of the trained machine learning model is improved by at least 30% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data.

[0090] In another aspect, a data analytics module is provided comprising one or more processors that are configured to execute a set of instructions for performing any of the methods or steps, as described above and elsewhere herein. In some instances, the data analytics module is implemented in a cloud-based environment.

[0091] A further aspect is directed to a method comprising: aggregating a plurality of training datasets from a plurality of sources to generate a combined dataset, wherein the plurality of training datasets comprise (1) clinical data from one or more clinical sources and (2) biomarker data from biological samples processed by one or more laboratory facilities for a plurality of test subjects; using the combined dataset to train a machine learning model; applying the trained machine learning model to at least clinical data associated with a patient, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the patient; and using in part the set of metrics to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0092] In some cases, the clinical data comprises structured data. In some cases, the structured data comprises categorical, textual, binary, or numerical values. In some cases, the biomarker data comprises continuous expressions or variables associated with one or more biomarkers. In some instances of the method, aggregating the plurality of training datasets comprises applying one or more pre-processing or formatting steps to the clinical data and the biomarker data, such that the combined dataset is configured for training the machine learningmodel. In some cases, the one or more pre-processing or formatting steps comprise using an algorithm to remove outliers from the clinical data and / or impute missing data fields in the clinical data. In some cases, the one or more pre-processing or formatting steps comprise using an algorithm to remove outliers from the biomarker data and / or normalize the biomarker data. In some cases, the one or more clinical sources comprise electronic health records (EHRs) or electronic medical records (EMRs). In some cases, the one or more clinical sources comprise databases maintained by one or more healthcare providers. In some cases, the clinical data comprises a plurality of responses provided by the test subjects or the patient to a questionnaire comprising a series of targeted questions or surveys. In some cases, the series of targeted questions or surveys are designed to collect information the test subjects or the patient from a list comprising of: demographics, reproductive history, medical history, abdominal and pelvic surgical history, symptoms currently being experienced, pain level and severity, surgical data and surgical procedures performed, histopathology reports, concomitant medications, or fasting status. In some cases, a set of clinically relevant factors is extracted from the list to train the machine learning model, or used in part by the trained machine learning model to generate the report. In some cases, the set of clinically relevant factors relates to one or more of the following: a pain severity score experienced by the test subjects or the patient in a previous month; age of first menarche; previous pregnancy (if applicable); body mass index (BMI); fibroids (if applicable and medical history); ovarian cysts (if applicable and medical history); symptom duration over past 10 years; heavy or irregular periods (current symptom experienced); infertility or history of infertility (if applicable), dysmenorrhea (current symptom experienced); pelvic pain (current symptom experienced); painful intercourse, urination, bowel movement (current symptom experienced); or family history of endometriosis (if applicable).

[0093] In some cases, at least some of the plurality of training datasets are aggregated using an application programming interface (API) or a web-based interface. In some cases, wherein at least some of the plurality of training datasets are stored in a cloud database. In some cases, the machine learning model is trained in a cloud-based environment. In some cases, the trained machine learning model is applied to at least the clinical data in a cloud-based environment or a local computing environment.

[0094] In some cases, the biomarker data comprises at least one of miRNA data or protein quantification data. In some cases, the miRNA data is associated with one or more miRNAs selected from the group consisting of hsa-miR-30b-5p, hsa-miR-19b-3p, hsa-miR-199a-5p, hsa-miR-125b-5p, hsa-let-7a-5p, hsa-miR-199a-3p, hsa-miR-99b-5p, hsa-miR-20a-5p, hsa-miR-23a- 3p, hsa-miR-339-3p, hsa-miR-24-3p, hsa-miR-145-5p, hsa-miR-3615, hsa-miR-185-5p, hsa- miR-25-3p, hsa-miR-484, hsa-miR-142-3p, hsa-miR-3613-5p, hsa-miR-15b-5p, hsa-miR-451a, hsa-miR-186-5p, hsa-miR-18a-5p, hsa-miR-501-3p, hsa-miR-141-3p, hsa-miR-29b-3p, hsa-miR- 22-3p, hsa-miR-150-5p, hsa-let-7d-5p, hsa-miR-122-5p, hsa-miR-182-5p, hsa-miR-98-5p, hsa- let-7b-5p, hsa-miR-342-3p, hsa-miR-628-3p, hsa-miR-155-5p, hsa-miR-130a-3p, hsa-miR-26a- 5p, hsa-miR-324-5p, hsa-let-7e-5p, and hsa-let-7f-5p. In some cases, the protein quantification data is associated with one or more proteins comprising of CA-125. In some cases, the trained machine learning model comprises one or more of the following: a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient-boosted classifier, or another supervised or unsupervised machine learning algorithm.

[0095] In some cases, a first algorithm can be used or configured to process clinical data, and a second algorithm can be used or configured to process biomarker data. The first and second algorithms may be same or different. In other cases, a combined model can be configured to process both clinical data and biomarker data. The combined model can be generated, for example by combining or fusing the first and second algorithms. The combined model can be configured to integrate and process clinical data with biomarker data. For example, the combined model is capable of integrating clinical data (which may be in a structured format such as categorical / yes-no / scored) with the continuous expression data for blood markers (e.g. miRNAs or CA-125) to produce a classification / prediction result. In some cases, a machine learning model can be used to fit the clinical data / factors and biomarker data. The model can be configured to handle categorical, binary or continuous variables together and assign weights accordingly.

[0096] The present disclosure further provides methods for using artificial intelligence methods, such as natural language processing (“NLP”) models or large language models (“LLMs”), to generate any of the questionnaires described herein, and / or to analyze subjects’ responses to the questionnaires. Outputs of the NLP models or LLMs can be provided to one or more machine learning algorithms described herein, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for a female subject.

[0097] In some embodiments, the questionnaire may include a list of algorithmically generated questions. The questions may be generated using, for example, a natural languageprocessing (NLP) algorithm or an LLM. The questions may be adaptively modified based at least in part on health profiles or historical data of the subjects.

[0098] Examples of NLP algorithms include sentiment analysis, semantic analysis, vectorspace semantics, relation extraction, language modeling, word / document embeddings and clustering, topic modeling, discourse analysis, syntactic analysis, and dialogue analysis. In some embodiments, the NLP model can be selected from the group consisting of a sentiment model, a statistical language model, a topic model, a syntactic model, an embedding model, or a dialog or discourse model.

[0099] In some instances, the NLP models or LLMs can be used to perform semantic analysis on a subject’s verbal responses to a questionnaire. Semantic analysis, as disclosed herein, may refer to analysis of spoken language from the subject’s responses to assessment questions or captured conversations. The analysis may be of words or phrases, and may be configured to account for primary queries or follow-up queries. In the case of captured humanhuman conversations, the analysis may also apply to the speech of the other party. As used herein, the terms “semantic analysis” and “natural language processing (NLP)”may be used interchangeably.

[0100] In some cases, any of the methods described herein may be used to generate an assessment of endometriosis, without requiring the presence or intervention of a human clinician. In other embodiments, the methods described herein may be able to be used to augment or enhance clinician-provided assessments, or aid a clinician in providing an assessment of endometriosis.

[0101] Methods and systems of the present disclosure can be implemented by way of one or more algorithms. An algorithm can be implemented by way of software upon execution by a central processing unit. The algorithm can, for example, analyze textual or verbal responses from subjects using natural language processing.

[0102] It should be noted that throughout this disclosure a series of terms may be used interchangeably, and this usage is not intended to limit the scope of the disclosure in any manner. For example, the terms “patient” or “subject” may be employed interchangeably to refer to an individual being screened for endometriosis. Likewise, “semantic analysis” and “NLP” may be used interchangeably to reference natural language processing models and elements. In a similar manner, “stakeholders” may be employed to refer to a wide variety of interested third parties who are not the patient being screened. These stakeholders may include physicians, health careproviders, care team members, insurance companies, research organizations, and family / relatives of the patient, hospitals, crisis centers and the like. It should thus be understood that when another label is employed, such as “physician”, the intention in this disclosure is to reference any number of stakeholders.

[0103] FIG. 21 illustrates a model building workflow in accordance with some embodiments. The workflow may include data processing / normalization / filtering steps as shown. In some cases, the model can be constructed with a statistical software using for example a glmnet package. In some instances, the model can be trained using subject data (e.g. 250 or more test subjects). The variables can be selected, and the performances can be estimated / evaluated based on out of bag sampling. In some cases, the model can be formulated as linear regression models with intercept and coefficients assigned for each factor.

[0104] FIG. 22A illustrates performance charts for different types of models, in accordance with some embodiments. Specifically, the charts show AUCs with different types of models. Referring to FIG. 22A, models without clinical data may be capable of achieving about 0.6 AUC, whereas the AUC of the model with clinical data is about 0.82.

[0105] FIG. 22B illustrates an Endometriosis / Non- Endometriosis distribution on CA-125 protein, in accordance with some embodiments. Based on the Endometriosis / NonEndometriosis distribution on CA-125, the model performance is likely to be about 50% accuracy. However, it is noted that the model performance can be improved to AUC of 0.78 by including clinical factors / data.

[0106] The model performance can be determined by the combination of multiple markers / factors. Markers / factors can be included in the model by being complementary with each other, or can be excluded by correlation with others. Examples of factors in the clinical data that were found to be of high relevance include family history, symptoms of pains, infertility, BMI, and others which are frequently included in the models.

[0107] In some cases, for a specific model, a weight can be computed for each factor, including clinical factor, miRNA, and CA-125 factor. A result can be calculated by the sum of the products of weights and their corresponding factor values, as in a linear model. The factors and weights may be same or different, in different models. In some cases, the factors and weights may be variable, and can be tailored to optimize model performance.

[0108] In some instances, the current clinical data only model may be capable of achieving an AUC of 0.685 at best. By adding CA-125 to the model, the AUC can be improved to 0.784. The AUC can be further improved to 0.82 by adding miRNAs to the model. Compared to the clinical factors, CA-125 and miRNA data may be more objective and often reflect the biological process / status. The functional / pathway analysis of the miRNAs may help to understand the mechanism of action of endometriosis and provide insights for the treatment development.

[0109] Referring to FIG. 2, a model may be run on blood plasma and may be based on three components: (1) clinical data collected from questionnaire; (2) miRNA data, and (3) normalizers. Examples of normalizers that may be used may include RNASeq hsa-miR-23a-3p, hsa-miR-30b- 5p, hsa-miR-151a-3p, hsa-miR-484, RD10021 hsa-miR-103a-3p, hsa-miR-484, hsa-miR-23a-3p, hsa-miR-28-5p, hsa-miR-30b-5p, hsa-miR-151a-3p, hsa-miR-484, hsa-miR-30b-5p, cel-miR-39- 3p, hsa-miR-23a-3p, hsa-miR-718, hsa-miR-151a-3p, hsa-miR-1307-3p, hsa-miR-28-5p, hsa- miR-130b-3p, hsa-miR-151a-3p, hsa-miR-23a-3p, hsa-miR-28-5p, hsa-miR-30b-5p, hsa-miR- 718, hsa-miR-484, hsa-miR-191-5p, hsa-miR-28-3p, hsa-miR-361-3p, hsa-miR-423-3p, hsa- miR-30b-5p, hsa-miR-484, cel-miR-39-3p, hsa-miR-23a-3p, hsa-miR-30b-5p, hsa-miR-151a-3p, hsa-miR-484, hsa-miR-28-5p, hsa-miR-103a-3p, cel-miR-39-3p, hsa-miR-718, hsa-miR-130b- 3p, hsa-miR-1307-3p, hsa-miR-191-5p, hsa-miR-28-3p, hsa-miR-361-3p, or hsa-miR-423-3pComputer Systems

[0110] In an aspect, the present disclosure provides computer systems that are programmed or otherwise configured to implement methods of the disclosure. FIG. 24 shows a computer system 2401 that is programmed or otherwise configured to implement a method of non- invasively testing or screening for endometriosis. The computer system 2401 may be configured to, for example, (a) obtain clinical data comprising a plurality of responses provided by a female subject to a questionnaire comprising a series of targeted questions or surveys; (b) process the clinical data into a data structure or format that is conducive or compatible for processing with biomarker data; and (c) apply a trained machine learning model to at least the processed clinical data, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the subject, wherein the set of metrics is useable to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject. In some instances, the computer system 2401 may be configured to, for example, aggregate a plurality of training datasets from a plurality of sources to generate a combined dataset, wherein the plurality of training datasets comprise (1) clinical data from one or more clinical sources and (2) biomarkerdata from biological samples processed by one or more laboratory facilities for a plurality of test subjects; use the combined dataset to train a machine learning model; apply the trained machine learning model to at least clinical data associated with a patient, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the patient; and use in part the set of metrics to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0111] The computer system 2401 can be an electronic device of a user or a computer system that is remotely located with respect to the electronic device. The electronic device can be a mobile electronic device.

[0112] The computer system 2401 may include a central processing unit (CPU, also "processor" and "computer processor" herein) 2405, which can be a single core or multi core processor, or a plurality of processors for parallel processing. The computer system 2401 also includes memory or memory location 2410 (e.g., random-access memory, read-only memory, flash memory), electronic storage unit 2415 (e.g., hard disk), communication interface 2420 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 2425, such as cache, other memory, data storage and / or electronic display adapters. The memory 2410, storage unit 2415, interface 2420 and peripheral devices 2425 are in communication with the CPU 2405 through a communication bus (solid lines), such as a motherboard. The storage unit 2415 can be a data storage unit (or data repository) for storing data. The computer system 2401 can be operatively coupled to a computer network ("network") 2430 with the aid of the communication interface 2420. The network 2430 can be the Internet, an internet and / or extranet, or an intranet and / or extranet that is in communication with the Internet. The network 2430 in some cases is a telecommunication and / or data network. The network 2430 can include one or more computer servers, which can enable distributed computing, such as cloud computing. The network 2430, in some cases with the aid of the computer system 2401, can implement a peer-to- peer network, which may enable devices coupled to the computer system 2401 to behave as a client or a server.

[0113] The CPU 2405 can execute a sequence of machine-readable instructions, which can be embodied in a program or software. The instructions may be stored in a memory location, such as the memory 2410. The instructions can be directed to the CPU 2405, which can subsequently program or otherwise configure the CPU 2405 to implement methods of the presentdisclosure. Examples of operations performed by the CPU 2405 can include fetch, decode, execute, and writeback.

[0114] The CPU 2405 can be part of a circuit, such as an integrated circuit. One or more other components of the system 2401 can be included in the circuit. In some cases, the circuit is an application specific integrated circuit (ASIC).

[0115] The storage unit 2415 can store files, such as drivers, libraries and saved programs. The storage unit 2415 can store user data, e.g., user preferences and user programs. The computer system 2401 in some cases can include one or more additional data storage units that are located external to the computer system 2401 (e.g., on a remote server that is in communication with the computer system 2401 through an intranet or the Internet).

[0116] The computer system 2401 can communicate with one or more remote computer systems through the network 2430. For instance, the computer system 2401 can communicate with a remote computer system of a user (e.g., an end user). Examples of remote computer systems include personal computers (e.g., portable PC), slate or tablet PC's (e.g., Apple® iPad, Samsung® Galaxy Tab), telephones, Smart phones (e.g., Apple® iPhone, Android-enabled device, Blackberry®), or personal digital assistants. The user can access the computer system 2401 via the network 2430.

[0117] Methods as described herein can be implemented by way of machine (e.g., computer processor) executable code stored on an electronic storage location of the computer system 2401, such as, for example, on the memory 2410 or electronic storage unit 2415. The machine executable or machine-readable code can be provided in the form of software. During use, the code can be executed by the processor 2405. In some cases, the code can be retrieved from the storage unit 2415 and stored on the memory 2410 for ready access by the processor 2405. In some situations, the electronic storage unit 2415 can be precluded, and machine-executable instructions are stored on memory 2410.

[0118] The code can be pre-compiled and configured for use with a machine having a processor adapted to execute the code, or can be compiled during runtime. The code can be supplied in a programming language that can be selected to enable the code to execute in a precompiled or as-compiled fashion.

[0119] Aspects of the systems and methods provided herein, such as the computer system 2401, can be embodied in programming. Various aspects of the technology may be thought of as"products" or "articles of manufacture" typically in the form of machine (or processor) executable code and / or associated data that is carried on or embodied in a type of machine readable medium. Machine-executable code can be stored on an electronic storage unit, such as memory (e.g., read-only memory, random-access memory, flash memory) or a hard disk. "Storage" type media can include any or all of the tangible memory of the computers, processors or the like, or associated modules thereof, such as various semiconductor memories, tape drives, disk drives and the like, which may provide non-transitory storage at any time for the software programming. All or portions of the software may at times be communicated through the Internet or various other telecommunication networks. Such communications, for example, may enable loading of the software from one computer or processor into another, for example, from a management server or host computer into the computer platform of an application server. Thus, another type of media that may bear the software elements includes optical, electrical and electromagnetic waves, such as used across physical interfaces between local devices, through wired and optical landline networks and over various air-links. The physical elements that carry such waves, such as wired or wireless links, optical links or the like, also may be considered as media bearing the software. As used herein, unless restricted to non-transitory, tangible "storage" media, terms such as computer or machine "readable medium" refer to any medium that participates in providing instructions to a processor for execution.

[0120] Hence, a machine readable medium, such as computer-executable code, may take many forms, including but not limited to, a tangible storage medium, a carrier wave medium or physical transmission medium. Non-volatile storage media including, for example, optical or magnetic disks, or any storage devices in any computer(s) or the like, may be used to implement the databases, etc. shown in the drawings. Volatile storage media include dynamic memory, such as main memory of such a computer platform. Tangible transmission media include coaxial cables; copper wire and fiber optics, including the wires that comprise a bus within a computer system. Carrier- wave transmission media may take the form of electric or electromagnetic signals, or acoustic or light waves such as those generated during radio frequency (RF) and infrared (IR) data communications. Common forms of computer-readable media therefore include for example: a floppy disk, a flexible disk, hard disk, magnetic tape, any other magnetic medium, a CD-ROM, DVD or DVD-ROM, any other optical medium, punch cards paper tape, any other physical storage medium with patterns of holes, a RAM, a ROM, a PROM and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave transporting data or instructions, cables or links transporting such a carrier wave, or any other medium fromwhich a computer may read programming code and / or data. Many of these forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to a processor for execution.

[0121] The computer system 2401 can include or be in communication with an electronic display 2435 that comprises a user interface (UI) 2440. A user or entity can also interact with various elements in the UI. Examples of UI's include, without limitation, a graphical user interface (GUI) and web-based user interface.

[0122] Methods and systems of the present disclosure can be implemented by way of one or more algorithms. An algorithm can be implemented by way of software upon execution by the central processing unit 2405. For example, the algorithm may be configured to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the subject based at on the processed clinical data, wherein the set of metrics is useable to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0123] Example of clinical study and evaluation protocol (shown on the following pages)Digital processing device

[0124] The methods, kits, and systems disclosed herein may include a digital processing device or use of the same. In further embodiments, the digital processing device includes one or more hardware central processing units (CPU) that carry out the device’s functions. In still further embodiments, the digital processing device further comprises an operating system configured to perform executable instructions. In some embodiments, the digital processing device is optionally connected to a computer network. In further embodiments, the digital processing device is optionally connected to the Internet such that it accesses the World Wide Web. In still further embodiments, the digital processing device is optionally connected to a cloud computing infrastructure. In other embodiments, the digital processing device is optionally connected to an intranet. In other embodiments, the digital processing device is optionally connected to a data storage device.

[0125] In accordance with the description herein, suitable digital processing devices include, by way of non-limiting examples, server computers, desktop computers, laptop computers, notebook computers, sub-notebook computers, netbook computers, netpad computers, set-top computers, handheld computers, Internet appliances, mobile smartphones, tablet computers, personal digital assistants, video game consoles, and vehicles. Those of skill in the art willrecognize that many smartphones are suitable for use in the system described herein. Those of skill in the art will also recognize that select televisions, video players, and digital music players with optional computer network connectivity are suitable for use in the system described herein. Suitable tablet computers include those with booklet, slate, and convertible configurations, known to those of skill in the art.

[0126] The digital processing device may generally include an operating system configured to perform executable instructions. The operating system may be, for example, software, including programs and data, which manages the device’ s hardware and provides services for execution of applications. Those of skill in the art will recognize that suitable server operating systems include, by way of non-limiting examples, FreeBSD, OpenBSD, NetBSD®, Linux, Apple® Mac OS X Server®, Oracle® Solaris®, Windows Server®, and Novell® NetWare®. Those of skill in the art will recognize that suitable personal computer operating systems include, by way of non-limiting examples, Microsoft® Windows®, Apple® Mac OS X®, UNIX®, and UNIX-like operating systems such as GNU / Linux®. In some embodiments, the operating system is provided by cloud computing. Those of skill in the art will also recognize that suitable mobile smart phone operating systems include, by way of non-limiting examples, Nokia® Symbian® OS, Apple® iOS®, Research In Motion® BlackBerry OS®, Google® Android®, Microsoft® Windows Phone® OS, Microsoft® Windows Mobile® OS, Linux®, and Palm® WebOS®.

[0127] The device generally includes a storage and / or memory device. The storage and / or memory device may be one or more physical apparatuses used to store data or programs on a temporary or permanent basis. In some embodiments, the device is volatile memory and requires power to maintain stored information. In some embodiments, the device is non-volatile memory and retains stored information when the digital processing device is not powered. In further embodiments, the non-volatile memory comprises flash memory. In some embodiments, the nonvolatile memory comprises dynamic random-access memory (DRAM). In some embodiments, the non-volatile memory comprises ferroelectric random-access memory (FRAM). In some embodiments, the non-volatile memory comprises phase-change random access memory (PRAM). In other embodiments, the device is a storage device including, by way of non-limiting examples, CD-ROMs, DVDs, flash memory devices, magnetic disk drives, magnetic tapes drives, optical disk drives, and cloud computing-based storage. In further embodiments, the storage and / or memory device is a combination of devices such as those disclosed herein.

[0128] A display to send visual information to a user may generally be initialized. Examples of displays include a cathode ray tube (CRT, a liquid crystal display (LCD), a thin film transistor liquid crystal display (TFT-LCD, an organic light emitting diode (OLED) display. In various further embodiments, on OLED display is a passive-matrix OLED (PMOLED) or active-matrix OLED (AMOLED) display. In some embodiments, the display may be a plasma display, a video projector or a combination of devices such as those disclosed herein.

[0129] The digital processing device may generally include an input device to receive information from a user. The input device may be, for example, a keyboard, a pointing device including, by way of non-limiting examples, a mouse, trackball, track pad, joystick, game controller, or stylus; a touch screen, or a multi-touch screen, a microphone to capture voice or other sound input, a video camera to capture motion or visual input or a combination of devices such as those disclosed herein.Non-transitory computer readable storage medium

[0130] The methods, kits, and systems disclosed herein may include one or more non- transitory computer readable storage media encoded with a program including instructions executable by the operating system to perform and analyze the test described herein; preferably connected to a networked digital processing device. The computer readable storage medium may be a tangible component of a digital that may be optionally removable from the digital processing device. The computer readable storage medium includes, by way of non-limiting examples, CD-ROMs, DVDs, flash memory devices, solid state memory, magnetic disk drives, magnetic tape drives, optical disk drives, cloud computing systems and services, and the like. In some instances, the program and instructions are permanently, substantially permanently, semipermanently, or non-transitorily encoded on the media.

[0131] A non-transitory computer-readable storage media may be encoded with a computer program including instructions executable by a processor to create or use a classification system. The storage media may comprise (a) a database, in a computer memory, of one or more clinical features of two or more control samples, wherein (i) the two or more control samples may be from two or more subjects; and (ii) the two or more control samples may be differentially classified based on a classification system comprising three or more classes; (b) a first software module configured to compare the one or more clinical features of the two or more control samples; and (c) a second software module configured to produce a classifier set based on the comparison of the one or more clinical features.

[0132] At least two of the classes may be selected from endometriosis, non-endometriosis, and healthy.Web application

[0133] In some embodiments, a computer program includes a web application. In light of the disclosure provided herein, those of skill in the art will recognize that a web application, in various embodiments, utilizes one or more software frameworks and one or more database systems. In some embodiments, a web application is created upon a software framework such as Microsoft® .NET or Ruby on Rails (RoR). In some embodiments, a web application utilizes one or more database systems including, by way of non-limiting examples, relational, non-relational, object oriented, associative, and XML database systems. In further embodiments, suitable relational database systems include, by way of non-limiting examples, Microsoft® SQL Server, mySQL™, and Oracle®. Those of skill in the art will also recognize that a web application, in various embodiments, is written in one or more versions of one or more languages. A web application may be written in one or more markup languages, presentation definition languages, client-side scripting languages, server-side coding languages, database query languages, or combinations thereof. In some embodiments, a web application is written to some extent in a markup language such as Hypertext Markup Language (HTML), Extensible Hypertext Markup Language (XHTML), or extensible Markup Language (XML). In some embodiments, a web application is written to some extent in a presentation definition language such as Cascading Style Sheets (CSS). In some embodiments, a web application is written to some extent in a client-side scripting language such as Asynchronous Javascript and XML (AJAX), Flash® Actionscript, Javascript, or Silverlight®. In some embodiments, a web application is written to some extent in a server-side coding language such as Active Server Pages (ASP), ColdFusion®, Perl, Java™, JavaServer Pages (JSP), Hypertext Preprocessor (PHP), Python™, Ruby, Tel, Smalltalk, WebDNA®, or Groovy. In some embodiments, a web application is written to some extent in a database query language such as Structured Query Language (SQL). In some embodiments, a web application integrates enterprise server products such as IBM® Lotus Domino®. In some embodiments, a web application includes a media player element. In various further embodiments, a media player element utilizes one or more of many suitable multimedia technologies including, by way of non-limiting examples, Adobe® Flash®, HTML 5, Apple® QuickTime®, Microsoft® Silverlight®, Java™, and Unity®.Mobile application

[0134] In some embodiments, a computer program includes a mobile application provided to a mobile digital processing device. In some embodiments, the mobile application is provided to a mobile digital processing device at the time it is manufactured. In other embodiments, the mobile application is provided to a mobile digital processing device via the computer network described herein.

[0135] In view of the disclosure provided herein, a mobile application may be created by techniques known to those of skill in the art using hardware, languages, and development environments known to the art. Those of skill in the art will recognize that mobile applications are written in several languages. Suitable programming languages include, by way of nonlimiting examples, C, C++, C#, Objective-C, Java™, Javascript, Pascal, Object Pascal, Python™, Ruby, VB.NET, WML, and XHTML / HTML with or without CSS, or combinations thereof.

[0136] Suitable mobile application development environments are available from several sources. Commercially available development environments include, by way of non-limiting examples, AirplaySDK, alcheMo, Appcelerator®, Celsius, Bedrock, Flash Lite, .NET Compact Framework, Rhomobile, and WorkLight Mobile Platform. Other development environments are available without cost including, by way of non-limiting examples, Lazarus, MobiFlex, MoSync, and Phonegap. Also, mobile device manufacturers distribute software developer kits including, by way of non-limiting examples, iPhone and iPad (iOS) SDK, Android™ SDK, BlackBerry® SDK, BREW SDK, Palm® OS SDK, Symbian SDK, webOS SDK, and Windows® Mobile SDK.

[0137] Those of skill in the art will recognize that several commercial forums are available for distribution of mobile applications including, by way of non-limiting examples, Apple® App Store, Android™ Market, BlackBerry® App World, App Store for Palm devices, App Catalog for webOS, Windows® Marketplace for Mobile, Ovi Store for Nokia® devices, Samsung® Apps, and Nintendo® DSi Shop.Standalone application

[0138] In some embodiments, a computer program includes a standalone application, which is a program that is run as an independent computer process, not as an add-on to an existing process, e.g., not a plug-in. Those of skill in the art will recognize that standalone applicationsare often compiled. A compiler is a computer program(s) that transforms source code written in a programming language into binary object code such as assembly language or machine code. Suitable compiled programming languages include, by way of non-limiting examples, C, C++, Objective-C, COBOL, Delphi, Eiffel, Java™, Lisp, Python™, Visual Basic, and VB .NET, or combinations thereof. Compilation is often performed, at least in part, to create an executable program. In some embodiments, a computer program includes one or more executable complied applications.Web browser plug-in

[0139] In some embodiments, the computer program includes a web browser plug-in. In computing, a plug-in is one or more software components that add specific functionality to a larger software application. Makers of software applications support plug-ins to enable third- party developers to create abilities which extend an application, to support easily adding new features, and to reduce the size of an application. When supported, plug-ins enable customizing the functionality of a software application. For example, plug-ins are commonly used in web browsers to play video, generate interactivity, scan for viruses, and display particular file types. Those of skill in the art will be familiar with several web browser plug-ins including, Adobe® Flash® Player, Microsoft® Silverlight®, and Apple® QuickTime®. In some embodiments, the toolbar comprises one or more web browser extensions, add-ins, or add-ons. In some embodiments, the toolbar comprises one or more explorer bars, tool bands, or desk bands.

[0140] In view of the disclosure provided herein, those of skill in the art will recognize that several plug-in frameworks are available that enable development of plug-ins in various programming languages, including, by way of non-limiting examples, C++, Delphi, Java™, PHP, Python™, and VB .NET, or combinations thereof.

[0141] Web browsers (also called Internet browsers) are software applications, designed for use with network-connected digital processing devices, for retrieving, presenting, and traversing information resources on the World Wide Web. Suitable web browsers include, by way of nonlimiting examples, Microsoft® Internet Explorer®, Mozilla® Firefox®, Google® Chrome, Apple® Safari®, Opera Software® Opera®, and KDE Konqueror. In some embodiments, the web browser is a mobile web browser. Mobile web browsers (also called microbrowsers, minibrowsers, and wireless browsers) are designed for use on mobile digital processing devices including, by way of non-limiting examples, handheld computers, tablet computers, netbook computers, subnotebook computers, smartphones, music players, personal digital assistants(PDAs), and handheld video game systems. Suitable mobile web browsers include, by way of non-limiting examples, Google® Android® browser, RIM BlackBerry® Browser, Apple® Safari®, Palm® Blazer, Palm® WebOS® Browser, Mozilla® Firefox® for mobile, Microsoft® Internet Explorer® Mobile, Amazon® Kindle® Basic Web, Nokia® Browser, Opera Software® Opera® Mobile, and Sony® PSP™ browser.Software modules

[0142] The methods, kits, and systems disclosed herein may include software, server, and / or database modules, or use of the same. In view of the disclosure provided herein, software modules are created by techniques known to those of skill in the art using machines, software, and languages known to the art. The software modules disclosed herein are implemented in a multitude of ways. In various embodiments, a software module comprises a file, a section of code, a programming object, a programming structure, or combinations thereof. In further various embodiments, a software module comprises a plurality of files, a plurality of sections of code, a plurality of programming objects, a plurality of programming structures, or combinations thereof. In various embodiments, the one or more software modules comprise, by way of nonlimiting examples, a web application, a mobile application, and a standalone application. In some embodiments, software modules are in one computer program or application. In other embodiments, software modules are in more than one computer program or application. In some embodiments, software modules are hosted on one machine. In other embodiments, software modules are hosted on more than one machine. In further embodiments, software modules are hosted on cloud computing platforms. In some embodiments, software modules are hosted on one or more machines in one location. In other embodiments, software modules are hosted on one or more machines in more than one location.Databases

[0143] The methods, kits, and systems disclosed herein may comprise one or more databases, or use of the same. In view of the disclosure provided herein, those of skill in the art will recognize that many databases are suitable for storage and retrieval of information pertaining to miRNA or ncRNA expression profiles, sequencing data, classifiers, classification systems, therapeutic regimens, or a combination thereof. In various embodiments, suitable databases include, by way of non-limiting examples, relational databases, non-relational databases, object- oriented databases, object databases, entity-relationship model databases, associative databases, and XML databases. In some embodiments, a database is internet-based. In further embodiments,a database is web-based. In still further embodiments, a database is cloud computing-based. In other embodiments, a database is based on one or more local computer storage devices.Data transmission

[0144] The methods, kits, and systems disclosed herein may be used to transmit one or more reports. The one or more reports may comprise information pertaining to the classification and / or identification of one or more samples from one or more subjects. The one or more reports may comprise information pertaining to a disease status (e.g., endometriosis or non-endometriosis). The one or more reports may comprise information pertaining to therapeutic regimens for use in treating endometriosis in a subject in need thereof. The one or more reports may be transmitted to a subject or a medical representative of the subject. The medical representative of the subject may be a physician, physician’s assistant, nurse, or other medical care provider. The medical representative of the subject may be a family member of the subject. A family member of the subject may be a parent, guardian, child, sibling, aunt, uncle, cousin, or spouse. The medical representative of the subject may be a legal representative of the subject.EXEMPLARY EMBODIMENTS

[0145] 1. A method of non-invasively or minimally invasively testing or screening for endometriosis, comprising:(a) obtaining clinical data comprising a plurality of responses provided by a subject to a questionnaire comprising a series of targeted questions or surveys;(b) processing the clinical data into a data structure or format that is conducive or compatible for processing with biomarker data; and(c) applying a trained machine learning model to at least the processed clinical data, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the subject, wherein the set of metrics is useable to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0146] 2. The method of embodiment 1 , wherein (a) further comprises obtaining the biomarker data from a biological sample collected from the subject, and (b) further comprises processing the biomarker data into another data structure or format that is conducive or compatible for processing with the clinical data.

[0147] 3. The method of embodiment 2, wherein (c) comprises applying the trained machine learning model to (1) the processed clinical data and (2) the processed biomarker data to improve an accuracy of the set of metrics.

[0148] 4. The method of embodiments 2 or 3, wherein the biomarker data comprises miRNA data associated with one or more miRNAs.

[0149] 5. The method of embodiment 4, wherein the one or more miRNAs are selected from the group consisting of hsa-miR-30b-5p, hsa-miR-19b-3p, hsa-miR-199a-5p, hsa-miR-125b-5p, hsa-let-7a-5p, hsa-miR-199a-3p, hsa-miR-99b-5p, hsa-miR-20a-5p, hsa-miR-23a-3p, hsa-miR- 339-3p, hsa-miR-24-3p, hsa-miR-145-5p, hsa-miR-3615, hsa-miR-185-5p, hsa-miR-25-3p, hsa- miR-484, hsa-miR-142-3p, hsa-miR-3613-5p, hsa-miR-15b-5p, hsa-miR-451a, hsa-miR-186-5p, hsa-miR-18a-5p, hsa-miR-501-3p, hsa-miR-141-3p, hsa-miR-29b-3p, hsa-miR-22-3p, hsa-miR- 150-5p, hsa-let-7d-5p, hsa-miR-122-5p, hsa-miR-182-5p, hsa-miR-98-5p, hsa-let-7b-5p, hsa- miR-342-3p, hsa-miR-628-3p, hsa-miR-155-5p, hsa-miR-130a-3p, hsa-miR-26a-5p, hsa-miR- 324-5p, hsa-let-7e-5p, and hsa-let-7f-5p.

[0150] 6. The method of embodiments 2 or 3, wherein the biomarker data comprises protein quantification data associated with one or more proteins.

[0151] 7. The method of embodiment 6, wherein the one or more proteins comprises CA-125.

[0152] 8. The method of any of embodiments 2 through 7, wherein the biological sample comprises blood or saliva.

[0153] 9. The method of embodiment 1 , wherein the series of targeted questions or surveys are designed to collect information of the subject from a list comprising of: demographics, reproductive history, medical history, abdominal and pelvic surgical history, symptoms that the subject is currently experiencing, pain level and severity, surgical data and surgical procedures performed, histopathology reports, concomitant medications, or fasting status.

[0154] 10. The method of embodiment 9, wherein (b) further comprises extracting a set of clinically relevant factors from the list, for analysis by the trained machine learning model.

[0155] 11. The method of embodiment 10, wherein the set of clinically relevant factors relates to one or more of the following: a pain severity score experienced by the subject in a previous month; age of first menarche; previous pregnancy (if applicable); body mass index (BMI); fibroids (if applicable and medical history); ovarian cysts (if applicable and medical history); symptom duration over past 10 years; heavy or irregular periods (current symptom experienced); infertility or history of infertility (if applicable), dysmenorrhea (current symptomexperienced); pelvic pain (current symptom experienced); painful intercourse, urination, bowel movement (current symptom experienced); or family history of endometriosis (if applicable).

[0156] 12. The method of embodiment 1, wherein at least some of the targeted questions or surveys are designed to establish or define a common terminology, framework or workflow for identifying, detecting or assessing dysmenorrhea and its symptoms, by healthcare providers and for patients.

[0157] 13. The method of embodiment 1, wherein the plurality of responses comprises one or more scaled scores.

[0158] 14. The method of embodiment 13, wherein the one or more scaled scores relate to a pain level or severity.

[0159] 15. The method of embodiment 1, wherein the plurality of responses comprises one or more binary yes / no responses.

[0160] 16. The method of embodiment 1, wherein the plurality of responses comprises one or more textual responses that are indicative of symptom(s), pain or discomfort that the subject is experiencing.

[0161] 17. The method of embodiment 16, wherein the one or more textual responses comprises free-form text.

[0162] 18. The method of embodiment 1, wherein the plurality of responses comprises one or more multiple-choice question responses.

[0163] 19. The method of embodiment 1, wherein the plurality of responses comprises one or more numerical values.

[0164] 20. The method of embodiment 19, wherein at least one of the numerical values relate to a severity / scale, time duration, frequency and / or date(s) of one or more physical conditions or symptoms experienced by the subject.

[0165] 21. The method of embodiment 1, wherein the series of targeted questions or surveys are presented to the subject via a graphical user interface (GUI), and wherein the clinical data is obtained when the subject inputs the plurality of responses via the GUI.

[0166] 22. The method of any of embodiments 2 through 7, wherein the trained machine learning model is trained using a training dataset comprising of both clinical data and biomarker data from a plurality of test subjects.

[0167] 23. The method of embodiment 22, wherein the clinical data is collected when the plurality of test subjects respond to the series of targeted questions or surveys.

[0168] 24. The method of embodiments 22 or 23, wherein the clinical data is collected from the plurality of test subjects at a plurality of time points spanning from pre-surgery to postsurgery.

[0169] 25. The method of any of embodiments 22 through 24, wherein the training dataset further comprises a surgical confirmation or histopathology report of whether each of the plurality of test subjects has endometriosis or non-endometriosis, performed within a time period after the test subjects have completed the series of targeted questions or surveys.

[0170] 26. The method of embodiment 22, wherein the trained machine learning model is trained on the clinical data and the miRNA data.

[0171] 27. The method of embodiment 22, wherein the trained machine learning model is trained on the clinical data and the protein quantification data.

[0172] 28. The method of embodiment 22, wherein the trained machine learning model is trained on the clinical data, the miRNA data, and the protein quantification data.

[0173] 29. The method of embodiment 22, wherein the trained machine learning model is configured to fit the clinical data and the biomarker data from the training dataset.

[0174] 30. The method of embodiment 22, wherein the trained machine learning model is configured to (a) collectively process (1) categorical and / or binary variables within the clinical data and (2) continuous expressions or variables in the biomarker data, and (b) assign a first set of weights to the categorical and / or binary variables and a second set of weights to the continuous expressions or variables.

[0175] 31. The method of any of embodiments 22 through 30, wherein the trained machine learning model comprises one or more of the following: a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient-boosted classifier, or another supervised or unsupervised machine learning algorithm.

[0176] 32. The method of embodiment 31, wherein the linear regression comprises LASSO(least absolute shrinkage and selection operator).

[0177] 33. The method of embodiment 22, wherein a performance metric of the trained machine learning model is improved by at least 10% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data.

[0178] 34. The method of embodiment 22, wherein a performance metric of the trained machine learning model is improved by at least 20% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data.

[0179] 35. The method of embodiment 22, wherein a performance metric of the trained machine learning model is improved by at least 30% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data.

[0180] 36. The method of any of embodiments 33 through 35, wherein the performance metric comprises Area Under the Curve (AUC), and wherein the trained machine learning model has an AUC greater than about 0.75, and the other machine learning model has an AUC less than about 0.75.

[0181] 37. The method of embodiment 36, wherein the trained machine learning model has an AUC greater than about 0.8.

[0182] 38. The method of embodiment 36, wherein the other machine learning model has anAUC ranging from about 0.6 to about 0.7.

[0183] 39. The method of any of embodiments 33 through 35, wherein the performance metric further comprises one or more of the following: sensitivity, specificity, positive predictive value (PPV), or negative predictive value (NPV).

[0184] 40. The method of embodiment 39, wherein the trained machine learning model has a sensitivity of at least about 70%, a specificity of at least about 70%, a PPV of at least about 80%, and an NPV of at least about 68%.

[0185] 41. The method of embodiment 1, further comprising: providing the set of metrics in a report to a healthcare provider or laboratory.

[0186] 42. The method of embodiment 42, further comprising: providing the report via a telehealth platform for establishing one or more pathways for effecting the specific course of treatment or prophylaxis for endometriosis for the subject.

[0187] 43. A data analytics module comprising one or more processors that are configured to execute a set of instructions for performing the methods or steps in any one of embodiments 1 through 42.

[0188] 44. The data analytics module of embodiment 43, wherein said data analytics module is implemented in a cloud-based environment.

[0189] 45. A method comprising: aggregating a plurality of training datasets from a plurality of sources to generate a combined dataset, wherein the plurality of training datasets comprise (1) clinical data from one or more clinical sources and (2) biomarker data from biological samples processed by one or more laboratory facilities for a plurality of test subjects; using the combined dataset to train a machine learning model; applying the trained machine learning model to at least clinical data associated with a patient, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the patient; and using in part the set of metrics to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

[0190] 46. The method of embodiment 45, wherein the clinical data comprises structured data.

[0191] 47. The method of embodiment 46, wherein the structured data comprises categorical, textual, binary, or numerical values.

[0192] 48. The method of embodiment 45, wherein the biomarker data comprises continuous expressions or variables associated with one or more biomarkers.

[0193] 49. The method of any of embodiments 45 through 48, wherein aggregating the plurality of training datasets comprises applying one or more pre-processing or formatting steps to the clinical data and the biomarker data, such that the combined dataset is configured for training the machine learning model.

[0194] 50. The method of embodiment 49, wherein the one or more pre-processing or formatting steps comprise using an algorithm to remove outliers from the clinical data and / or impute missing data fields in the clinical data.

[0195] 51. The method of embodiment 49, wherein the one or more pre-processing or formatting steps comprise using an algorithm to remove outliers from the biomarker data and / or normalize the biomarker data.

[0196] 52. The method of embodiment 45, wherein the one or more clinical sources comprise electronic health records (EHRs) or electronic medical records (EMRs).

[0197] 53. The method of embodiment 45, wherein the one or more clinical sources comprise databases maintained by one or more healthcare providers.

[0198] 54. The method of any of embodiment 45, wherein the clinical data comprises a plurality of responses provided by the test subjects or the patient to a questionnaire comprising a series of targeted questions or surveys.

[0199] 55. The method of embodiment 54, wherein the series of targeted questions or surveys are designed to collect information the test subjects or the patient from a list comprising of: demographics, reproductive history, medical history, abdominal and pelvic surgical history, symptoms currently being experienced, pain level and severity, surgical data and surgical procedures performed, histopathology reports, concomitant medications, or fasting status.

[0200] 56. The method of embodiment 55, wherein a set of clinically relevant factors is extracted from the list to train the machine learning model, or used in part by the trained machine learning model to generate the report.

[0201] 57. The method of embodiment 56, wherein the set of clinically relevant factors relates to one or more of the following: a pain severity score experienced by the test subjects or the patient in a previous month; age of first menarche; previous pregnancy (if applicable); body mass index (BMI); fibroids (if applicable and medical history); ovarian cysts (if applicable and medical history); symptom duration over past 10 years; heavy or irregular periods (current symptom experienced); infertility or history of infertility (if applicable), dysmenorrhea (current symptom experienced); pelvic pain (current symptom experienced); painful intercourse, urination, bowel movement (current symptom experienced); or family history of endometriosis (if applicable).

[0202] 58. The method of embodiment 45, wherein at least some of the plurality of training datasets are aggregated using an application programming interface (API) or a web-based interface.

[0203] 59. The method of embodiment 45, wherein at least some of the plurality of training datasets are stored in a cloud database.

[0204] 60. The method of embodiment 45, wherein the machine learning model is trained in a cloud-based environment.

[0205] 61. The method of embodiment 45, wherein the trained machine learning model is applied to at least the clinical data in a cloud-based environment or a local computing environment.

[0206] 62. The method of any of embodiments 45 through 61, wherein the biomarker data comprises at least one of miRNA data or protein quantification data.

[0207] 63. The method of embodiment 62, wherein the miRNA data is associated with one or more miRNAs selected from the group consisting of hsa-miR-30b-5p, hsa-miR-19b-3p, hsa- miR-199a-5p, hsa-miR-125b-5p, hsa-let-7a-5p, hsa-miR-199a-3p, hsa-miR-99b-5p, hsa-miR- 20a-5p, hsa-miR-23a-3p, hsa-miR-339-3p, hsa-miR-24-3p, hsa-miR-145-5p, hsa-miR-3615, hsa- miR-185-5p, hsa-miR-25-3p, hsa-miR-484, hsa-miR-142-3p, hsa-miR-3613-5p, hsa-miR-15b- 5p, hsa-miR-451a, hsa-miR-186-5p, hsa-miR-18a-5p, hsa-miR-501-3p, hsa-miR-141-3p, hsa- miR-29b-3p, hsa-miR-22-3p, hsa-miR-150-5p, hsa-let-7d-5p, hsa-miR-122-5p, hsa-miR-182-5p, hsa-miR-98-5p, hsa-let-7b-5p, hsa-miR-342-3p, hsa-miR-628-3p, hsa-miR-155-5p, hsa-miR- 130a-3p, hsa-miR-26a-5p, hsa-miR-324-5p, hsa-let-7e-5p, and hsa-let-7f-5p.

[0208] 64. The method of embodiment 62, wherein the protein quantification data is associated with one or more proteins comprising of CA-125.

[0209] 65. The method of any of embodiments 45 through 64, wherein the trained machine learning model comprises one or more of the following: a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient-boosted classifier, or another supervised or unsupervised machine learning algorithm.

[0210] 66. The method of any of the previous embodiments, wherein the set of metrics is useable to develop and effect a specific course of treatment for endometriosis for the subject.

[0211] 67. The method of any of the previous embodiments wherein the method is a method of detecting and treating endometriosis in the subject comprising treating the subject with a treatment for endometriosis when the machine learning model detects the presence or likelihood of endometriosis in the subject.

[0212] 68. The method of the previous embodiment, wherein the treatment for endometriosis comprises treatment comprises a hormonal treatment, a statin, a non-steroidal anti-inflammatory drug (NSAID), an oral contraceptive, a progestin, a gonadotrophin releasing (GnRH) agonist, a GnRH antagonist, an androgen, an antiprogesterone, a selective estrogen receptor modulator (SERM), a selective progesterone receptor modulator (SPRM), atorvastatin, cerivastatin, fluvastatin, lovastatin, mevastatin, pitavastatin, pravastatin, rosuvastatin, simvastatin, paracetamol, a COX-2 inhibitor, or aspirin.

[0213] 69. A computer-implemented method of detecting endometriosis in a subject, the computer-implemented method comprising:(a) obtaining clinical data comprising a plurality of responses provided by the subject to a questionnaire comprising a series of targeted questions, wherein the subject has endometriosis symptoms but has not been diagnosed with endometriosis;(b) processing the clinical data into a data structure to obtain processed clinical data compatible with a trained machine learning model; and(c) applying the trained machine learning model to the processed clinical data, to generate a set of metrics based on a statistical significance of the clinical data comprising the plurality of responses and wherein, based on the set of metrics, the trained machine learning model detects a presence, absence or likelihood of endometriosis in the subject.

[0214] 70. The computer-method of any of the previous embodiments, wherein the set of metrics is useable to develop and effect a specific course of treatment for endometriosis for the subject.

[0215] 71. The computer-implemented method of the previous embodiments, wherein the computer-implemented method is a method of detecting and treating endometriosis in the subject comprising treating the subject with a treatment for endometriosis when the machine learning model detects the presence or likelihood of endometriosis in the subject.

[0216] 72. The computer-implemented method of the previous embodiment, wherein the treatment for endometriosis comprises treatment comprises a hormonal treatment, a statin, a non-steroidal anti-inflammatory drug (NSAID), an oral contraceptive, a progestin, a gonadotrophin releasing (GnRH) agonist, a GnRH antagonist, an androgen, an antiprogesterone, a selective estrogen receptor modulator (SERM), a selective progesterone receptor modulator (SPRM), atorvastatin, cerivastatin, fluvastatin, lovastatin, mevastatin, pitavastatin, pravastatin, rosuvastatin, simvastatin, paracetamol, a COX-2 inhibitor, or aspirin.

[0217] 73. The method or computer-implemented method of any of the previous embodiments, wherein the questionnaire includes at least one question related to: demographics, reproductive history, medical history, abdominal and pelvic surgical history, symptoms that the subject is currently experiencing, pain level and severity, surgical data and surgical procedures performed, histopathology reports, concomitant medications, or fasting status.

[0218] 74. The method of embodiment 9, wherein (b) further comprises extracting a set of clinically relevant factors from the list, for analysis by the trained machine learning model.

[0219] 75. The method of embodiment 10, wherein the set of clinically relevant factors relates to one or more of the following: a pain severity score experienced by the subject in a previous month; age of first menarche; previous pregnancy (if applicable); body mass index (BMI); fibroids (if applicable and medical history); ovarian cysts (if applicable and medical history); symptom duration over past 10 years; heavy or irregular periods (current symptom experienced); infertility or history of infertility (if applicable), dysmenorrhea (current symptom experienced); pelvic pain (current symptom experienced); painful intercourse, urination, bowel movement (current symptom experienced); or family history of endometriosis (if applicable).

[0220] 76. The method of any of the previous embodiments, wherein at least one of clinically relevant factors group A are ranked with a greater weight or importance value than at least one of clinically relevant factors in group B, wherein group A comprises: age, body mass index, age, fibroids, previous pregnancy, race, dysmenorrhea, non-menstrual pelvic pain, painful intercourse, painful bowel movements, painful urination, fasting status, infertility, weight, Pain severity last period, ovarian cysts, family history of endometriosis, length of time or duration of endometriosis symptoms, infertility, abdominal symptoms, PNS-NM, surgery for abnormal uterine bleeding / chronic pelvic pain / ovarian cysts / bowel issues; and wherein group B comprises: abdominal symptoms, pain severity last month, surgery for uterine fibroids, symptom duration, fasting, adenomyosis, surgery for chronic pelvic pain, endometrioma, cycle Daycrd, polycystic ovarian syndrome, symptom none, hormone treatment, surgery for abnormal uterine bleeding, currently bleeding, age of first menarche, interstitial cystitis, family history, ethnicity, surgery for ovarian cysts, irritable bowel syndrome, and Stage ORD.

[0221] 77. The method of any one of the previous embodiments, wherein at least one of the clinically relevant factors is age, race, ethnicity, height, weight, body mass index, pregnancy history, infertility, family history of endometriosis, signs and symptoms, pain, pain severity,dysmenorrhea, non-menstrual pelvic pain, painful intercourse, pain severity last month, and pain severity last period.

[0222] 78. The method of any one of the previous embodiments, wherein at least one of the clinically relevant factors is age, race, ethnicity, height, weight, body mass index, pregnancy history, infertility, family history of endometriosis, signs and symptoms, pain, pain severity, dysmenorrhea, non-menstrual pelvic pain, painful intercourse, pain severity last month, and pain severity last period are weighted relatively heavily compared to other clinically relevant features in the machine learning model.

[0223] 79. The method of embodiment 78, wherein group A comprises: age, body mass index, age, fibroids, previous pregnancy, race, dysmenorrhea, non-menstrual pelvic pain, painful intercourse, painful bowel movements, painful urination, fasting status, infertility, body weight, or Pain severity last period.

[0224] 80. The method of any of the previous embodiments, wherein at least some of the targeted questions or surveys are designed to establish or define a common terminology, framework or workflow for identifying, detecting or assessing dysmenorrhea and its symptoms, by healthcare providers and for patients.

[0225] 81. The method of any of the previous embodiments, wherein the plurality of responses comprises one or more scaled scores.

[0226] 82. The method of any of the previous embodiments, wherein the one or more scaled scores relate to a pain level or severity.

[0227] 83. The method of any of the previous embodiments, wherein the plurality of responses comprises one or more binary yes / no responses.

[0228] 84. The method of any of the previous embodiments, wherein the plurality of responses comprises one or more textual responses that are indicative of symptom(s), pain or discomfort that the subject is experiencing.

[0229] 85. The method of any of the previous embodiments, wherein the one or more textual responses comprises free-form text.

[0230] 86. The method of any of the previous embodiments, wherein the plurality of responses comprises one or more multiple-choice question responses.

[0231] 87. The method of any of the previous embodiments, wherein the plurality of responses comprises one or more numerical values.

[0232] 88. The method of any of the previous embodiments, wherein at least one of the numerical values relate to a severity / scale, time duration, frequency and / or date(s) of one or more physical conditions or symptoms experienced by the subject.

[0233] 89. The method of any of the previous embodiments, wherein the series of targeted questions or surveys are presented to the subject via a graphical user interface (GUI), and wherein the clinical data is obtained when the subject inputs the plurality of responses via the GUI.

[0234] 90. The method of any of the previous embodiments, wherein the trained machine learning model is trained using a training dataset comprising of both clinical data and biomarker data from a plurality of test subjects.

[0235] 91. The method of any of the previous embodiments, wherein the clinical data is collected when the plurality of test subjects respond to the series of targeted questions or surveys.

[0236] 92. The method of any of the previous embodiments, wherein the clinical data is collected from the plurality of test subjects at a plurality of time points spanning from presurgery to post-surgery.

[0237] 93. The method of any of the previous embodiments, wherein the training dataset further comprises a surgical confirmation or histopathology report of whether each of the plurality of test subjects has endometriosis or non-endometriosis, performed within a time period after the test subjects have completed the series of targeted questions or surveys.

[0238] 94. The method of any of the previous embodiments, wherein the trained machine learning model is trained on the clinical data and the miRNA data.

[0239] 95. The method of any of the previous embodiments, wherein the trained machine learning model is trained on the clinical data and the protein quantification data.

[0240] 96. The method of any of the previous embodiments, wherein the trained machine learning model is trained on the clinical data, the miRNA data, and the protein quantification data.

[0241] 97. The method of any of the previous embodiments, wherein the trained machine learning model is configured to fit the clinical data and the biomarker data from the training dataset.

[0242] 98. The method of any of the previous embodiments, wherein the trained machine learning model is configured to (a) collectively process (1) categorical and / or binary variables within the clinical data and (2) continuous expressions or variables in the biomarker data, and (b) assign a first set of weights to the categorical and / or binary variables and a second set of weights to the continuous expressions or variables.

[0243] 99. The method of any of the previous embodiments, wherein the trained machine learning model comprises one or more of the following: a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient-boosted classifier, or another supervised or unsupervised machine learning algorithm.

[0244] 100. The method of any of the previous embodiments, wherein the linear regression comprises LASSO (least absolute shrinkage and selection operator).

[0245] 101. The method of any of the previous embodiments, wherein a performance metric of the trained machine learning model is improved by at least 10% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data.

[0246] 102. The method of any of the previous embodiments, wherein a performance metric of the trained machine learning model is improved by at least 20% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data.

[0247] 103. The method of any of the previous embodiments, wherein a performance metric of the trained machine learning model is improved by at least 30% compared to another machine learning model that is trained on only one of the biomarker data or the clinical data.

[0248] 104. The method of any of the previous embodiments, wherein the performance metric comprises Area Under the Curve (AUC), and wherein the trained machine learning model has an AUC greater than about 0.75, and the other machine learning model has an AUC less than about 0.75.

[0249] 105. The method of any of the previous embodiments, wherein the trained machine learning model has an AUC greater than about 0.8.

[0250] 106. The method of any of the previous embodiments, wherein the other machine learning model has an AUC ranging from about 0.6 to about 0.7.

[0251] 107. The method of any of the previous embodiments, wherein the performance metric further comprises one or more of the following: sensitivity, specificity, positive predictive value (PPV), or negative predictive value (NPV).

[0252] 108. The method of any of the previous embodiments, wherein the trained machine learning model has a sensitivity of at least about 70%, a specificity of at least about 70%, a PPV of at least about 80%, and an NPV of at least about 68%.

[0253] 109. The method of any of the previous embodiments, further comprising: providing the set of metrics in a report to a healthcare provider or laboratory.

[0254] 110. The method of any of the previous embodiments, further comprising: providing the report via a telehealth platform for establishing one or more pathways for effecting the specific course of treatment or prophylaxis for endometriosis for the subject.

[0255] 111. The method of any of the previous embodiments, wherein the machine learning model further includes ca-125 data and the presence of endometriosis is detected based less than 40%, 30%, 20%, 15%, 10%, 5%, or 1% on the ca-125 data.

[0256] 112. The method of any of the previous embodiments, wherein the machine learning model further includes ca-125 data and wherein ca-125 is weighted such as that it makes up less than the 40%, 30%, 20%, 15%, 10%, 5%, or 1% of the analysis.

[0257] 113. The method of any of the previous embodiments, wherein the subject presents with pelvic pain, and the machine learning model discriminates between endometriosis and at least one of the following indications: pelvic inflammatory disease (PID), appendicitis, ovarian cysts, ovarian torsion, ectopic pregnancy, celiac disease, irritable bowel syndrome, fibromyalgia, or malignancy.

[0258] 114. The method of any of the previous embodiments, wherein the subject presents with infertility, and the machine learning model discriminates between endometriosis and at least one of the following indications: male factor, anovulation, luteal phase defect, tubal factors, ovarian insufficiency, fibroids, pelvic inflammatory disease (PID), hyper or hypothyroidism, or polycystic ovarian syndrome (PCOS).

[0259] 115. The method of any of the previous embodiments, wherein the subject presents with dysmenorrhea or painful periods, and the machine learning model discriminates between endometriosis and at least one of the following indications: pelvic inflammatory disease (PID), adenomyosis, leiomyomas, fibroids, ovarian cysts, irritable bowel syndrome, endometrial polyps, cervical stenosis, musculoskeletal disorders or issues, or increased prostaglandins.

[0260] 116. The method of any of the previous embodiments, wherein the subject presents with dyspareunia or painful sex, and the machine learning model discriminates between endometriosis and at least one of the following indications: vaginismus, vaginal atrophy, vulvodynia, vaginitis, vulvovaginitis, pelvic adhesions, trauma, uterine prolapse, cystitis, PID, or uterine fibroids.

[0261] 117. The method of any of the previous embodiments, wherein the subject presents with abnormal menstruation, and the machine learning model discriminates between endometriosis and at least one of the following indications: uterine polyps, fibroids, pelvic inflammatory disease (PID), polycystic ovarian syndrome (PCOS), cancer, hyper or hypothyroidism, pituitary disorders, excessive exercise, medications or abnormal pregnancy.

[0262] 118. An artificial intelligence method for developing a model for classifying endometriosis status in a subject, comprising:(a) providing a data structure comprising, for a plurality of subjects, at least (1) responses to a plurality of items in a questionnaire; and(b) training a machine learning algorithm to generate a model that uses responses to one or more of the items to classify endometriosis status of a subject.

[0263] 119. The method of embodiment 118, wherein the data structure further comprises(2) data indicating endometriosis status of the subject.

[0264] 120. The method of embodiment 118 or embodiment 119, wherein the data structure further comprises (3) biomarker data.

[0265] 121. The method of embodiment 120, wherein the biomarker data comprises miRNA data.

[0266] 122. The method of embodiment 120 or embodiment 121, wherein the biomarker data comprises protein biomarker data (e.g., CA125).

[0267] 123. The method of embodiment 118, wherein the data structure does not include biomarker data.

[0268] 124. The method of any of embodiments 118-123, wherein the plurality of subjects comprises at least five subjects, at least 10 subjects, at least 50 subjects, at least 100 subjects, or at least 200 subjects in which endometriosis is present; and at least 10 subjects, at least 50 subjects, at least 100 subjects, or at least 200 subjects in which endometriosis is absent.

[0269] 125. The method of embodiment 124, wherein the total number of subjects is at least20 subjects, at least 50 subjects, at least 100 subjects, at least 200 subjects, or at least 400 subjects.

[0270] 126. The method of any of embodiments 118-125, wherein the subjects comprise premenopausal females over the age of 18.

[0271] 127. The method of any of embodiments 118-126, wherein the questionnaire comprises items involving information relating to one, two, three, four, five, six, seven, eight, nine, 10, 11, 12, 13, 14, 15, 16 or more of: demography, medical history, abdominal and pelvic surgical history, reproductive history, current symptoms, Endometriosis Health Profile (EHP-5) items, pain symptom score, vital signs, Endometriosis Phenome and Biospecimen Harmonisation Project items, surgical diagnosis, post surgical status, and medications, symptoms that the subject is currently experiencing, surgical data and surgical procedures performed, histopathology reports, concomitant, and fasting status.

[0272] 128. The method of any of embodiments 118-127, wherein the number of variables(i.e., features) in the data structure is at least 100, at least 500 or at least 1000.

[0273] 129. The method of any of embodiments 118-127, wherein the number of variables in the data structure is at least 2-times, at least 10 times or at least 100 times the number of subjects.

[0274] 130. The method of any of embodiments 118-128, wherein the machine learning algorithm is selected from a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient-boosted classifier.

[0275] 131. The method of any of embodiments 118-130, wherein providing the data structure comprises receiving responses to a plurality of items in the questionnaire from the plurality of subjects; and entering the responses into the data structure.

[0276] 132. The method of any of embodiments 118-131, wherein providing the data structure comprises receiving responses to a plurality of items in the questionnaire for a subject from health professional; and entering the responses into the data structure.

[0277] 133. The method of any of embodiments 118-132, wherein responses are provided through a web portal.

[0278] 134. The method of any of embodiments 118-133, wherein the responses are analyzed by a large language model.

[0279] 135. The method of any of embodiments 118-134, wherein providing the data structure comprises obtaining a biological sample from a subject, measuring one or more biomarkers in the biological sample, and entering the measurement into the data structure.

[0280] 136. The method of any of embodiments 118-135, wherein presence or absence of endometriosis in the subject is determined by a health professional.

[0281] 137. The method of any of embodiments 118-136, wherein the endometriosis status of the subject classified by the model is selected from presence / absence, stage of endometriosis, risk of developing endometriosis, likelihood of having endometriosis.

[0282] 138. An artificial intelligence method for classifying endometriosis status in a subject, comprising:(a) providing a data structure comprising, for a subject, at least, (I) responses to a plurality of items in a questionnaire; and(b) executing a model on the data structure to classify endometriosis status of the subject.

[0283] 139. The method of embodiment 138, wherein the subject presents with one or more signs or symptoms selected from pelvic pain, infertility, dysmenorrhea, dyspareunia and abnormal menstruation, and classifying comprises providing a differential diagnosis of endometriosis.

[0284] 140. The method of embodiment 138, wherein the data structure further comprises(II) biomarker data.

[0285] 141. The method of embodiment 140, wherein the biomarker data comprises miRNA data.

[0286] 142. The method of embodiment 140 or embodiment 141, wherein the biomarker data comprises protein biomarker data (e.g., CA125).

[0287] 143. The method of any of embodiments 138-140, wherein the data structure does not include biomarker data.

[0288] 144. The method of any of embodiments 138-143, wherein endometriosis status is selected from presence or absence, stage, risk level, or likelihood of endometriosis.

[0289] 145. The method of any of embodiments 138-144, wherein the classifier comprises as decision tree.

[0290] 146. The method of any of embodiments 138-145, wherein the model is a model of any of embodiments 118-137.

[0291] 147. The method of any of embodiments 138-146, wherein the model uses at least any of 5, 10, 20, 50 or 100.

[0292] 148. The method of any of embodiments 138-147, further comprising reporting the classification to the subject or to a healthcare provider.

[0293] 149. A method comprising:(a) determining that a subject has or is at risk of having endometriosis by the method of any of embodiments 138-148; and(b) treating the subject for endometriosis.

[0294] 150. The method of embodiment 149, wherein the treatment is selected from one or more of hormonal treatment, surgery, laparoscopic surgery, a statin, a non-steroidal antiinflammatory drug (NSAID), an oral contraceptive, a progestin, a gonadotrophin releasing (GnRH) agonist, a GnRH antagonist, an androgen, an antiprogesterone, a selective estrogen receptor modulator (SERM), a selective progesterone receptor modulator (SPRM), atorvastatin, cerivastatin, fluvastatin, lovastatin, mevastatin, pitavastatin, pravastatin, rosuvastatin, simvastatin, paracetamol, a COX-2 inhibitor, and aspirin.

[0295] 151. A computer system configured to:(a) obtain at least clinical data comprising a plurality of responses provided by a female subject to a questionnaire comprising a collection of items;(b) process the clinical data into a data structure; and(c) apply a trained machine learning model to at least the processed clinical data, to generate a classifier that classified endometriosis status of the female subject.

[0296] 152. The computer system of embodiment 151,i) a computer comprising a monitor; ii) a remote server in communication with the computer; iii) wherein the server presents a webpage from a website to the monitor comprising items from a questionnaire and accepts responses from a user; iv) wherein the system comprises a classifier configured to classify endometriosis status of a subject based on responses from the user and, optionally, report the endometriosis status to a webpage on the website.

[0297] 153. The computer system of embodiment 151 or 152, wherein the questionnaire comprises items involving information relating to one, two, three, four, five, six, seven, eight, nine, 10, 11, 12, 13, 14, 15, 16 or more of: demography, medical history, abdominal and pelvic surgical history, reproductive history, current symptoms, Endometriosis Health Profile (EHP-5) items, pain symptom score, vital signs, Endometriosis Phenome and Biospecimen Harmonisation Project items, surgical diagnosis, post surgical status, and medications, symptoms that the subject is currently experiencing, surgical data and surgical procedures performed, histopathology reports, concomitant, and fasting status.

[0298] 154. A method comprising: a) displaying a website comprising a webpage that displays items on a questionnaire; b) receiving responses to the items through the website; c) using a classifier algorithm, predicting an endometriosis status based on the responses; and, optionally, d) reporting the predicted status through the website.

[0299] 155. The method of embodiment 154, wherein the questionnaire comprises items involving information relating to one, two, three, four, five, six, seven, eight, nine, 10, 11, 12, 13, 14, 15, 16 or more of: demography, medical history, abdominal and pelvic surgical history, reproductive history, current symptoms, Endometriosis Health Profile (EHP-5) items, pain symptom score, vital signs, Endometriosis Phenome and Biospecimen Harmonisation Project items, surgical diagnosis, post surgical status, and medications, symptoms that the subject is currently experiencing, surgical data and surgical procedures performed, histopathology reports, concomitant, and fasting status.

[0300] 156. A method comprising: a) obtaining from a subject a biological sample;b) obtaining from a subject responses to items in a questionnaire, wherein responses to the items serve as markers of endometriosis;

[0301] c) measuring one or more biomarkers in the biological sample; d) generating a data structure comprising the responses and the measures; e) using a classifier, predicting a status for endometriosis for the subject.

[0302] 157. The method of embodiment 156, wherein the items in the questionnaire include one or more of:Demographics (e.g., date of birth, race, ethnicity),Eligibility (e.g., informed consent)Reproductive History (e.g., age of first menarche, number of pregnancies, number of live births, number of C-sections, number of live births, experience of infertility, fertility treatment, family history of endometriosis),Medical History (e.g., prior medical conditions),Abdominal & Pelvic Surgical History,Current Symptoms (e.g., related to dysmenorrhea),Endometriosis Health Profile 5 (EHP-5) items (e.g., items related to pain, control and powerlessness, emotional well-being, social support, self-image)Pain Symptom Score (e.g., “Worst pelvic pain score during last month.”),Vital Signs (e.g., height, weight, BMI),Endometriosis Phenome and Biobanking Harmonisation Project (EPHect) Form (e.g., items related to pain, depression and anxiety, menstrual history, fertility, medical and surgical history, medication use, personal information)Diagnosis (e.g., diagnosis of endometriosis),Subject Status (e.g., completion of questionnaire),Concomitant medications (e.g., medication name, dose, frequency, route, indication, start date, end date).EXAMPLES

[0303] EMPOWER herein is an acronym for a clinical study carried out by DotLabs™. In many instances, although surgeons may be familiar with endometriosis in EMPOWER, the definition of dysmenorrhea in EMPOWER is not defined, as the physicians involved understand the symptom well and how to identify it. In the broader population using the test, providers may not be as familiar with the term dysmenorrhea and / or how to identify or diagnose it. To maintainconsistency on the terminology, the present disclosure provides for the construction of a straightforward questionnaire or instructions on assessing for dysmenorrhea with help from EMPOWER physicians and key opinion leaders (KOLs) to be utilized as part of the test requisition form.

[0304] DotEndo™ Test OVERVIEW

[0305] DotEndo™ as described in the present disclosure is a noninvasive test used to test or evaluate for endometriosis, utilizing the methods, models and systems described herein. The test can be powered by a machine learning algorithm that combines patient clinical information with protein testing from a blood sample, as described elsewhere herein.

[0306] Data training set: The training set has been developed using the clinical information and protein quantification results obtained from women aged 18 to 49 enrolled in the EMPOWER clinical trial, a multi-center, prospective, observational study in patients undergoing laparoscopy, laparotomy or other pelvic surgical procedure for endometriosis, infertility or other benign gynecological condition such as pelvic pain, benign pelvic masses or, abnormal uterine bleeding.

[0307] Intended use: DotEndoTM is a first line tool intended to be used in women of reproductive age - 18 through 49 years of age - with unexplained pelvic pain, unexplained infertility, and / or suspected endometriosis. The test, performed in laboratory (or laboratories), incorporates the levels of biomarkers and clinical data as validated in surgically-confirmed patients into an algorithm that generates a test result that, if high risk or positive, is suggestive of the presence of endometriosis. This information, in conjunction with standard clinical assessment, may be used to screen for endometriosis.

[0308] Time-to-results: DotEndo™ test results can be available in a few business days after DotLab™ receives complete clinical information and any biomarker values, such as CA125. Results will be sent back to the ordering laboratory for review.

[0309] Performance: DotEndo™ initially perform above 70% sensitivity and 70% specificity.

[0310] Workflow: The provider and / or patient complete the test requisition form which comprises both demographic and clinical information. The questions have controlled vocabulary, and the questions are estimated to require no more than 10 minutes to complete via a graphical user interface (GUI). The provider can order DotEndo™ through the existing laboratory partnertest ordering interface. Alternatively, paper test requisition can be taken to the blood draw station. The patient has blood drawn (a serum sample) at a phlebotomy service center or in their provider’s office. The blood is sent along with the test requisition form to the designated laboratory partner. Alternatively, the patient may receive by mail a blood self-collection kit from DotLab™. The patient may safely self-collect a capillary blood sample at home without any direct interaction, and mails the specimen to the designated laboratory using the packaging materials provided with the kit. The sample is accessioned into the laboratory partner system. A unique barcode is assigned, and protein testing is completed. The testing laboratory forwards the completed test requisition form (including the unique barcode) and the protein results to DotLab™ via secure email, fax or web interface. DotLab™ accessions the data in house to ensure positive sample tracking. DotLab™ then runs a machine-learning algorithm using the questionnaire and protein quantification results, as described elsewhere herein. A DotLab ™branded patient report is returned to the lab partner for distribution to the healthcare provider. An alternative is for DotLab™ to provide just the result, not on a report.

[0311] DotEndo™ proteins of interest may include one or more of the following. CA-125: Discriminatory power (cases vs controls) for the first 250 patients in EMPOWER has been conclusively tested. CRP: Inflammation-related protein (Irungu et al., 2019; Kocbek et al. 2015; Abrao et al 1997). Immunoglobulines: IgG and IgM ACL; anti-CA autoantibodies (D'Cruz et al. 1996; Abrao et al 1997). Interleukins: IL-1R2, IL-l-beta, IL-6, IL-8, IL-15, IL-18 and IL-33 (Irungu et al., 2019; Nematian et al 2018, Miller et al., 2017; Mihalyi et al 2010, Kocbek et al. 2015). Hormones: Leptin, progesterone, oestradiol (Irungu et al., 2019; Kocbek et al. 2015; Ozhan et al. 2014). CA-15.3, CA19-9 and CAI, (Tuten et al 2013; Zhang et al 2006). TNFa (Nematian et al 2018, Mihalyi et al 2010). Others: Enolase, Glycodelin-A, ficolin-2, biglycan, Zn-alpha2-glycoprotein, ICAM-1, Gro-alpha, MCP-1, MIF, PDPK1, SICAM-1, CCL5, TMP2, TNC, VCAM-1, YKC-40 (Irungu et al., 2019; Liu and Qiu, 2018; Kocbek et al. 2015; Ozhan et al. 2014; Tuten et al., 2014; Signorile and Baldi 2014; Abrao et al 1997).

[0312] Machine learning information for the DotEndo™ test

[0313] Goal: DotEndo ™uses a scalable and automatic system to provide end-to-end data analysis and interpretation solutions from clinical data collection to clinical reports.

[0314] Platform description Option 1

[0315] The platform can be cloud-based and HIPAA-compliant including several component systems, for example as shown in FIG. 23. Data collection: A web interface can be used to allowusers to enter answers to clinical questions using controlled vocabularies. A test order can be generated in a LIMS system. Patient demographics, physician order name and information can be collected. The system can include an interface to directly communicate with collaborators for them to upload the clinical information (inclusive of the CAI 25 test result). The system can be configured to query LIMS, e.g. SQL Server database to retrieve data needed for analysis. The post analysis result can be available for review / reporting in LIMS. Data management: The data collected can be stored in a database and file system, for example S3 bucket in AWS cloud. Analytical System: The EMPOWER data can be used to train the machine learning algorithm. The system can be configured to retrieve patients’ clinical information / data from the databases and LIMS system, analyze the data, and generate the diagnostic results. Reporting system: A reporting system can be used to generate the clinical reports and communicate with patients / physicians / collaborators / EMR / EHR systems. An alternative is to utilize the LIS system for reporting. If a partner is going to issue the final report, the answer (risk score for endometriosis) can be transmitted back to the partner via the web interface / API.

[0316] Platform Description Option 2

[0317] The lab partner sends TRF (with answers to all clinical questions) and the CA125 result. The data can be entered into. A verification / double check mechanism can be created to reduce data entry errors, can be configured to export all of the data to SQL server (at a fixed or variable frequency). The DotLab™ machine learning algorithm can access all or any necessary information from the SQL server. The machine learning algorithm can run in batches (e.g. daily). A file with results from the batch can be generated. The lab staff can access the file and enter the result into. The report is generated, and is reviewed and signed off by appropriately qualified personnel. A process can be defined for entering data in a desired format to the machine learning algorithm to access.

[0318] Analytical system: The machine learning model can be trained and further optimized. The workflow can be automated. The clinical information can be analyzed to determine the weight of each attribute. The clinical information as well as the question list can be refined to optimize the model.

[0319] While preferred embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. It is not intended that the invention be limited by the specific examples provided within the specification. While the invention has been described with reference to theaforementioned specification, the descriptions and illustrations of the embodiments herein are not meant to be construed in a limiting sense. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the invention. Furthermore, it shall be understood that all aspects of the invention are not limited to the specific depictions, configurations or relative proportions set forth herein which depend upon a variety of conditions and variables. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention. It is therefore contemplated that the invention shall also cover any such alternatives, modifications, variations or equivalents. It is intended that the following claims define the scope of the invention and that methods and structures within the scope of these claims and their equivalents be covered thereby.

[0320] While preferred embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. It is not intended that the invention be limited by the specific examples provided within the specification. While the invention has been described with reference to the aforementioned specification, the descriptions and illustrations of the embodiments herein are not meant to be construed in a limiting sense. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the invention. Furthermore, it shall be understood that all aspects of the invention are not limited to the specific depictions, configurations or relative proportions set forth herein which depend upon a variety of conditions and variables. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention. It is therefore contemplated that the invention shall also cover any such alternatives, modifications, variations or equivalents. It is intended that the following claims define the scope of the invention and that methods and structures within the scope of these claims and their equivalents be covered thereby.

Claims

WHAT IS CLAIMED IS:

1. An artificial intelligence method for developing a model for classifying endometriosis status in a subject, comprising:(a) providing a data structure comprising, for a plurality of subjects, at least (1) responses to a plurality of items in a questionnaire; and(b) training a machine learning algorithm to generate a model that uses responses to one or more of the items to classify endometriosis status of a subject.

2. The method of claim 1, wherein the data structure further comprises (2) data indicating endometriosis status of the subject.

3. The method of claim 1 or claim 2, wherein the data structure further comprises (3) biomarker data.

4. The method of claim 3, wherein the biomarker data comprises miRNA data.

5. The method of claim 3 or claim 4, wherein the biomarker data comprises protein biomarker data (e.g., CA125).

6. The method of claim 1, wherein the data structure does not include biomarker data.

7. The method of any of claims 1-6, wherein the plurality of subjects comprises at least five subjects, at least 10 subjects, at least 50 subjects, at least 100 subjects, or at least 200 subjects in which endometriosis is present; and at least 10 subjects, at least 50 subjects, at least 100 subjects, or at least 200 subjects in which endometriosis is absent.

8. The method of claim 7, wherein the total number of subjects is at least 20 subjects, at least 50 subjects, at least 100 subjects, at least 200 subjects, or at least 400 subjects.

9. The method of any of claims 1-8, wherein the subjects comprise premenopausal females over the age of 18.

10. The method of any of claims 1-9, wherein the questionnaire comprises items involving information relating to one, two, three, four, five, six, seven, eight, nine, 10, 11, 12, 13, 14, 15, 16 or more of: demography, medical history, abdominal and pelvic surgical history, reproductive history, current symptoms, Endometriosis Health Profile (EHP-5) items, pain symptom score, vital signs, Endometriosis Phenome and Biospecimen Harmonisation Project items, surgical diagnosis, post surgical status, and medications, symptoms that the subject is currently experiencing, surgical data and surgical procedures performed, histopathology reports, concomitant, and fasting status.

11. The method of any of claims 1-10, wherein the number of variables (i.e., features) in the data structure is at least 100, at least 500 or at least 1000.

12. The method of any of claims 1-10, wherein the number of variables in the data structure is at least 2-times, at least 10 times or at least 100 times the number of subjects.

13. The method of any of claims 1-11, wherein the machine learning algorithm is selected from a support vector machine (SVM), a naive Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, a neural network, a gradient-boosted classifier.

14. The method of any of claims 1-13, wherein providing the data structure comprises receiving responses to a plurality of items in the questionnaire from the plurality of subjects; and entering the responses into the data structure.

15. The method of any of claims 1-14, wherein providing the data structure comprises receiving responses to a plurality of items in the questionnaire for a subject from health professional; and entering the responses into the data structure.

16. The method of any of claims 1-15, wherein responses are provided through a web portal.

17. The method of any of claims 1-16, wherein the responses are analyzed by a large language model.

18. The method of any of claims 1-17, wherein providing the data structure comprises obtaining a biological sample from a subject, measuring one or more biomarkers in the biological sample, and entering the measurement into the data structure.

19. The method of any of claims 1-18, wherein presence or absence of endometriosis in the subject is determined by a health professional.

20. The method of any of claims 1-19, wherein the endometriosis status of the subject classified by the model is selected from presence / absence, stage of endometriosis, risk of developing endometriosis, likelihood of having endometriosis.

21. An artificial intelligence method for classifying endometriosis status in a subject, comprising:(a) providing a data structure comprising, for a subject, at least, (I) responses to a plurality of items in a questionnaire; and(b) executing a model on the data structure to classify endometriosis status of the subject.

22. The method of claim 21, wherein the subject presents with one or more signs or symptoms selected from pelvic pain, infertility, dysmenorrhea, dyspareunia and abnormal menstruation, and classifying comprises providing a differential diagnosis of endometriosis.

23. The method of claim 21, wherein the data structure further comprises (II) biomarker data.

24. The method of claim 23, wherein the biomarker data comprises miRNA data.

25. The method of claim 23 or claim 24, wherein the biomarker data comprises protein biomarker data (e.g., CA125).

26. The method of any of claims 21-23, wherein the data structure does not include biomarker data.

27. The method of any of claims 21-26, wherein endometriosis status is selected from presence or absence, stage, risk level, or likelihood of endometriosis.

28. The method of any of claims 21-27, wherein the classifier comprises as decision tree.

29. The method of any of claims 21-28, wherein the model is a model of any of claims 1-20.

30. The method of any of claims 21-29, wherein the model uses at least any of 5, 10, 20, 50 or 100.

31. The method of any of claims 21-30, further comprising reporting the classification to the subject or to a healthcare provider.

32. A method comprising:(a) determining that a subject has or is at risk of having endometriosis by the method of any of claims 21-31; and(b) treating the subject for endometriosis.

33. The method of claim 32, wherein the treatment is selected from one or more of hormonal treatment, surgery, laparoscopic surgery, a statin, a non-steroidal anti-inflammatory drug (NSAID), an oral contraceptive, a progestin, a gonadotrophin releasing (GnRH) agonist, a GnRH antagonist, an androgen, an antiprogesterone, a selective estrogen receptor modulator (SERM), a selective progesterone receptor modulator (SPRM), atorvastatin, cerivastatin, fluvastatin, lovastatin, mevastatin, pitavastatin, pravastatin, rosuvastatin, simvastatin, paracetamol, a COX-2 inhibitor, and aspirin.

34. A computer system configured to:(a) obtain at least clinical data comprising a plurality of responses provided by a female subject to a questionnaire comprising a collection of items;(b) process the clinical data into a data structure; and(c) apply a trained machine learning model to at least the processed clinical data, to generate a classifier that classified endometriosis status of the female subject.

35. The computer system of claim 34, i) a computer comprising a monitor; ii) a remote server in communication with the computer; iii) wherein the server presents a webpage from a website to the monitor comprising items from a questionnaire and accepts responses from a user; iv) wherein the system comprises a classifier configured to classify endometriosis status of a subject based on responses from the user and, optionally, report the endometriosis status to a webpage on the website.

36. The computer system of claim 34 or 35, wherein the questionnaire comprises items involving information relating to one, two, three, four, five, six, seven, eight, nine, 10, 11, 12, 13, 14, 15, 16 or more of: demography, medical history, abdominal and pelvic surgical history, reproductive history, current symptoms, Endometriosis Health Profile (EHP-5) items, pain symptom score, vital signs, Endometriosis Phenome and Biospecimen Harmonisation Project items, surgical diagnosis, post surgical status, and medications, symptoms that the subject is currently experiencing, surgical data and surgical procedures performed, histopathology reports, concomitant, and fasting status.

37. A method comprising: a) displaying a website comprising a webpage that displays items on a questionnaire; b) receiving responses to the items through the website; c) using a classifier algorithm, predicting an endometriosis status based on the responses; and, optionally, d) reporting the predicted status through the website.

38. The method of claim 37, wherein the questionnaire comprises items involving information relating to one, two, three, four, five, six, seven, eight, nine, 10, 11, 12, 13, 14, 15, 16 or more of: demography, medical history, abdominal and pelvic surgical history, reproductive history, current symptoms, Endometriosis Health Profile (EHP-5) items, pain symptom score, vital signs, Endometriosis Phenome and Biospecimen Harmonisation Project items, surgicaldiagnosis, post surgical status, and medications, symptoms that the subject is currently experiencing, surgical data and surgical procedures performed, histopathology reports, concomitant, and fasting status.

39. A method comprising: a) obtaining from a subject a biological sample; b) obtaining from a subject responses to items in a questionnaire, wherein responses to the items serve as markers of endometriosis; c) measuring one or more biomarkers in the biological sample; d) generating a data structure comprising the responses and the measures; e) using a classifier, predicting a status for endometriosis for the subject.

40. The method of claim 39, wherein the items in the questionnaire include one or more of:Demographics (e.g., date of birth, race, ethnicity), Eligibility (e.g., informed consent)Reproductive History (e.g., age of first menarche, number of pregnancies, number of live births, number of C-sections, number of live births, experience of infertility, fertility treatment, family history of endometriosis),Medical History (e.g., prior medical conditions), Abdominal & Pelvic Surgical History,Current Symptoms (e.g., related to dysmenorrhea),Endometriosis Health Profile 5 (EHP-5) items (e.g., items related to pain, control and powerlessness, emotional well-being, social support, self-image)Pain Symptom Score (e.g., “Worst pelvic pain score during last month.”), Vital Signs (e.g., height, weight, BMI),Endometriosis Phenome and Biobanking Harmonisation Project (EPHect) Form (e.g., items related to pain, depression and anxiety, menstrual history, fertility, medical and surgical history, medication use, personal information)Diagnosis (e.g., diagnosis of endometriosis),Subject Status (e.g., completion of questionnaire),Concomitant medications (e.g., medication name, dose, frequency, route, indication, start date, end date).

41. A method of non-invasively or minimally invasively testing or screening for endometriosis, comprising:(a) obtaining clinical data comprising a plurality of responses provided by a subject to a questionnaire comprising a series of targeted questions or surveys;(b) processing the clinical data into a data structure or format that is conducive or compatible for processing with biomarker data; and(c) applying a trained machine learning model to at least the processed clinical data, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the subject, wherein the set of metrics is useable to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

42. A data analytics module comprising one or more processors that are configured to execute a set of instructions for performing the methods or steps in any one of claims 1 through 42.

43. A method comprising: aggregating a plurality of training datasets from a plurality of sources to generate a combined dataset, wherein the plurality of training datasets comprise (1) clinical data from one or more clinical sources and (2) biomarker data from biological samples processed by one or more laboratory facilities for a plurality of test subjects; using the combined dataset to train a machine learning model; applying the trained machine learning model to at least clinical data associated with a patient, to generate a set of metrics that indicates a presence, absence, stage, risk level or likelihood of endometriosis for the patient; and using in part the set of metrics to develop and effect a specific course of treatment or prophylaxis for endometriosis for the subject.

44. A computer-implemented method of detecting endometriosis in a subject, the computer-implemented method comprising:(a) obtaining clinical data comprising a plurality of responses provided by the subject to a questionnaire comprising a series of targeted questions, wherein the subject has endometriosis symptoms but has not been diagnosed with endometriosis;(b) processing the clinical data into a data structure to obtain processed clinical data compatible with a trained machine learning model; and(c) applying the trained machine learning model to the processed clinical data, to generate a set of metrics based on a statistical significance of the clinical data comprising5 the plurality of responses and wherein, based on the set of metrics, the trained machine learning model detects a presence, absence or likelihood of endometriosis in the subject.

Images