Method and system for predicting the progression of pregnancy-induced hypertension

Abstract

The present invention relates to a method for predicting a patient's health state, the method comprising receiving at least one patient-related kinetic characteristic data, receiving at least one patient-related covariate, processing the at least one patient-related kinetic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set, generating at least one health state hypothesis based on the patient-related processed data set, and predicting at least one health state based on the at least one health state hypothesis. The present invention also relates to a system for predicting a patient's health state, the system comprising at least one processing component configured to receive at least one patient-related kinetic characteristic data, receive at least one patient-related covariate, and process the at least one patient-related kinetic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set; and at least one analysis component configured to analyze the patient-related processed data set and generate at least one health state hypothesis based on the patient-related processed data set, wherein the system is configured to predict at least one health state based on the at least one health state hypothesis and to implement the method according to any of the preceding claims.

Description

Technical Field

[0001] The present invention is in the field of predicting a patient's medical condition, and more particularly in the field of predicting the progression and / or onset of preeclampsia in pregnant women, and its impact on their children. The object of the present invention is to provide a method and system for predicting potential medical outcomes for pregnant women and / or their children. More specifically, the present invention relates to a system, a method implemented in such a system, and the use of a corresponding system.

Background Art

[0002] Preeclampsia (PE) is a major cause of short- and long-term morbidity and mortality in pregnancy and the peripartum period worldwide (Tanner, 2022), affecting approximately 5% of all pregnancies (Mol, 2015). This has increased the need for better technologies and new approaches to reduce the burden on affected people. PE is characterized by new-onset pregnancy hypertension or prior hypertension in the presence of at least one new sign of end-organ dysfunction that cannot be explained by causes other than PE.

[0003] PE is one of the most severe pregnancy complications, posing a significant risk of morbidity and death for both the mother and her child. PE is a complex disease, and diagnosis is difficult because pregnant women often have pre-existing morbidities that overlap with PE, such as prior hypertension or fetal growth restriction. In black women, PE affects up to 8% of pregnancies. Despite extensive research, it remains almost impossible to adequately predict, treat, or prevent PE.

[0004] As demonstrated by the fetal programming of lifestyle diseases and the increased risk of developing maternal cardiovascular diseases, the effects of PE on infants and mothers persist for years after pregnancy (Wellmann, 2014). It is important to understand the three major ongoing pathological stages of PE, namely (i) placental hypoxia and oxidative stress, (ii) excessive release of anti-angiogenic factors and pro-inflammatory factors, and (iii) extensive systemic endothelial dysfunction and vasoconstriction (de Alwis, 2020).

[0005] As outlined by Scott et al. in recent years, various guidelines for PE have been devised over the past few decades, presenting conventional diagnostic methods and best practices for the early detection of medical conditions (Scott, 2022). In fact, the noble goal of all research and subsequent clinical care is to identify PE when the mother and baby are asymptomatic and prevent them from reaching the symptomatic stage.

[0006] MacDonald et al. have presented an up-to-date review of clinical tools and biomarkers for predicting PE (MacDonald, 2022). Peripheral blood biomarkers, namely soluble Fms-like tyrosine kinase-1 (sFlt1) and placental growth factor (PlGF), have shown promising performance when used as "rule-out" tests that mean excluding patients. However, the sensitivity for detecting affected patients is low. This review describes several placental and cardiovascular biomarkers that could improve future diagnostic performance.

[0007] Jhee et al. (Jhee, 2019) have presented a predictive model for late-onset PE using different predictive models (e.g., logistic regression, decision tree, naive Bayes classifier, support vector machine, etc.). The prediction of late-onset (i.e., after 34 weeks of pregnancy) was performed on a dataset using maternal characteristics and test parameters in the early second trimester. A detection rate of 77.1% was achieved, while the study endpoint was defined as the first hypertension accompanied by significant proteinuria.

[0008] Maric et al. (Maric, 2020) presented a method focusing on statistical analysis. A model is trained on all available clinical and examination data, enabling the inclusion of a large number of missing values. Doppler imaging is downplayed as a feature because it makes verification more difficult. This greatly increases its applicability in different medical organizations. Since it is trained with a flexible net, this method has not achieved performance metrics high enough to be used in a clinical environment. SUMMARY OF THE INVENTION

[0009] Accordingly, in view of the above, an object of the present invention is to overcome or at least mitigate the drawbacks and disadvantages of the prior art. More specifically, an object of the present invention is to provide a method for predicting a patient's health state, and a corresponding system for the method, with improved sensitivity and performance and less likely to produce an incorrect prediction of at least one health state of the patient.

[0010] These objects are achieved by the present invention.

[0011] In a first aspect, the present invention relates to a method for predicting a patient's health state, the method comprising receiving at least one patient-related kinetic characteristic data, receiving at least one patient-related covariate, processing the at least one patient-related kinetic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set, generating at least one health state hypothesis based on the patient-related processed data set, and predicting at least one health state based on the at least one health state hypothesis.

[0012] In one embodiment, the step of predicting the at least one health condition may be based on a computer-implemented dynamic model. The at least one health condition hypothesis may have a correlation with at least one medical condition of the patient. It should be understood that the patient may be a female patient, such as a pregnant woman and / or a non-pregnant woman. It should be understood that the term "female" as a patient is intended to refer to a female patient of reproductive age and / or childbearing age. Furthermore, the patient may be a fetus and / or a newborn. In one embodiment, the method may further comprise a step of predicting the dynamic behavior of the at least one health condition of the patient.

[0013] In a further embodiment, the method may comprise implementing at least one machine learning technique, wherein the method may comprise using the at least one machine learning technique to perform any of the aforementioned steps.

[0014] The at least one patient-related covariate may comprise at least one biomarker. The at least one biomarker may relate to at least one disease state, and the at least one biomarker may include at least one of the following: soluble Fms-like tyrosine kinase-1 (sFlt-1); placental growth factor (PlGF); neurofilament (NfL); C-terminal portion of arginine vasopressin (copeptin); albumin; hepatic transaminases; urea; hemoglobin; platelets; creatinine; albuminuria; proteinuria; estimated glomerular filtration rate (eGFR); creatinine clearance (CrCl); at least one additional renal function measure; placental biomarkers, such as placental RNA, placental proteins; endothelial / cardiovascular biomarkers, such as endothelial RNA, endothelial proteins; or any combination thereof.

[0015] At least one patient-related covariate may include at least one mother-related covariate including at least one of the following: age; weight; height; body mass index (BMI); pregnancy history; delivery history; number of fetuses in the current pregnancy; race; body temperature; heart rate; heart rate variability; respiratory rate; early rupture of membranes; white blood cell count; history of preeclampsia (family and maternal), gestational diabetes, obesity, cardiovascular / renal / liver / thyroid diseases, autoimmune diseases, anemia, antiphospholipid syndrome, sexually transmitted infections, coexisting conditions such as headache; smoking habit before and / or during pregnancy; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); uteroplacental perfusion parameters; Doppler measurements of pulsatility indices of umbilical artery, middle cerebral artery, cerebroplacental ratio, uterine artery, fetal descending aorta, ductus venosus, umbilical vein, inferior vena cava, uterine artery; soft tissue parameters such as upper arm partial volume and thigh partial volume; and measurements of at least one of said at least one biomarker.

[0016] At least one patient-related covariate may include at least one fetus-related covariate including at least one of the following: gender; fetal weight during pregnancy; biometric parameters of the fetus such as femur length, abdominal circumference, head circumference, mid-thigh circumference, and transverse diameter; proportion of small-for-gestational-age; gestational age; heart rate; heart rate variability; respiratory rate; uteroplacental perfusion parameters; and measurements of at least one of said at least one biomarker. Additionally or alternatively, at least one patient-related covariate may include at least one newborn-related covariate including at least one of the following: gender; birth weight; weight; height; gestational age at birth; postnatal age; body temperature; heart rate, heart rate variability; respiratory rate; duration of breastfeeding; duration of exclusive breastfeeding; pH value; respiratory support; oxygen requirement; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); Apgar score; at least one additional neonatal biometric parameter; and measurements of at least one of said at least one biomarker. Additionally or alternatively, at least one patient-related covariate may include at least one environmental covariate including at least one of the following: country of residence, country of birth, time of birth, humidity conditions at birth, and ambient temperature at birth.

[0017] In one embodiment, the method may comprise generating at least one threshold, where the at least one threshold represents an indicator of at least one potential medical condition. In a further embodiment, the method may comprise outputting at least one potential medical condition, where the at least one potential medical condition may include at least one of the following: seizure; respiratory; cardiovascular; blood dysfunction; endocrine; renal; hepatic; uteroplacental dysfunction; fetal growth retardation; unplanned preterm birth; placental abruption; hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome; and eclampsia. It should be understood that PE can cause multi-organ complications, i.e., central nervous system (CNS) complications, including seizure, respiratory, cardiovascular, blood dysfunction, endocrine, renal, hepatic, and uteroplacental dysfunction. Also, since PE affects the arteries that carry blood to the placenta, the complications of PE may include, but are not limited to, fetal growth retardation; preterm birth. Further, PE can lead to unplanned preterm birth, i.e., delivery before 37 weeks. Therefore, it should be noted that the present invention is applicable to gestational hypertension and gestational diabetes and is related to the maternal, fetal, and neonatal complications of these conditions in pregnant women.

[0018] In a further embodiment, the method may comprise determining a minimum threshold for the at least one threshold and determining a maximum threshold for the at least one threshold. Also, at least one patient-related data may be below the minimum threshold, and the method may comprise outputting a monitoring proposal. If at least one patient-related data can be above the maximum threshold, the method may comprise outputting a treatment proposal. Further, the method may comprise determining a baseline for the at least one patient-related data. Additionally or alternatively, the method may comprise determining at least one intermediate threshold, where the at least one intermediate threshold may include at least one value between the minimum threshold and the maximum threshold.

[0019] In one embodiment, the method may comprise correlating at least a respective range of each of the at least one intermediate threshold with the at least one medical condition, generating an interpreted data set based on the correlating, and outputting an automated report indicating the at least one potential medical condition. In a further embodiment, the method may comprise determining at least one medical condition change indicator, monitoring a change in the at least one medical condition change indicator, generating at least one medical condition change indicator trend, and predicting a progression of at least one of the at least one medical conditions, where the prediction may be based on the at least one medical condition change indicator trend. Also, the method may comprise monitoring a change in value of at least one of the at least one patient-related characteristic, and the method may comprise recording an initial value of the at least one patient-related characteristic, recording at least one subsequent value of the at least one patient-related characteristic, comparing the initial value with at least one of the at least one subsequent values, generating comparison value data, and outputting a hypothesis of the patient-related characteristic based on the comparison value data. The step of recording at least one subsequent value may comprise recording a current value of one of the at least one patient-related characteristics, where the current value may be different from the initial value.

[0020] In one embodiment, the method may be a non-diagnostic method. In another embodiment, the method may be a diagnostic method.

[0021] The method may also comprise performing the steps of the method described herein using data from at least one database. The at least one database may have at least one of the following: a public health database, a patient's personal database, a database of medical experts, a database of healthcare providers, and a private data bank. Additionally or alternatively, the method may comprise supplying data to the at least one server, training the computer-implemented dynamic model based on the data supplied to the at least one server, and generating an adjustment function based on the training data, where the adjustment function may be suitable for adjusting any step of the method described in any of the embodiments of the method described above.

[0022] In one embodiment, the method may comprise triggering at least one action proposal based on the at least one health state hypothesis. Additionally or alternatively, the method may comprise displaying the at least one action proposal to a user. In another embodiment, the method may comprise prompting the user to input at least one of acceptance of at least one of the at least one action proposals and rejection of at least one of the at least one action proposals. The user may reject at least one of the at least one action proposals, and the method may comprise prompting the user to provide at least one annotation. The computer-implemented dynamic model may be based on: Bayesian statistical techniques; artificial neural network (ANN) techniques; convolutional neural network (CNN) techniques; recurrent neural network (RNN) techniques; pharmacodynamics (PMX) modeling and / or simulation techniques; supervised learning techniques; deep learning (DL) techniques, multi-layer neural network techniques, and / or explainable AI (XAI) concepts.

[0023] At least one medical condition may have at least one of the following: a fetal growth-related condition; a PE-related condition; a gestational diabetes-related condition; a gestational hypertension-related condition; pregnancy medicine such as cyclooxygenase inhibitors, for example aspirin; at least one related drug treatment; and at least one potential medical condition.

[0024] The method may also comprise correlating at least one biomarker with at least one medical condition, wherein the at least one medical condition may have an underlying disease. The method may also comprise predicting the occurrence of at least one hypothesis over a given period, wherein the method further may comprise recognizing a plurality of different periods including at least one of a prepartum period; pregnancy; a delivery period; and a postpartum period. In one embodiment, the method may comprise outputting a likelihood of occurrence correlated with each of the periods. Further, the method may comprise executing at least one machine learning (ML) algorithm. The at least one ML algorithm may have a supervised algorithm architecture, an unsupervised algorithm architecture, or any combination thereof. Additionally or alternatively, the at least one ML algorithm may include at least one artificial deep learning (DL) architecture, wherein the at least one artificial DL architecture may include at least one of an ANN, a CNN, and an RNN. The unsupervised algorithm architecture may include implementing at least one clustering technique of at least one cluster. Further, the at least one analysis technique may include at least one of pattern recognition, probabilistic modeling, Bayesian schemes, reinforcement learning, statistical analysis, statistical models, principal component analysis (PCA), independent component analysis, dynamic time warping, maximum likelihood estimation (MLE), modeling, estimation, neural network (NN), CNN, RNN, deep convolutional network, DL, ultra-deep learning, genetic algorithms, Markov models, and / or hidden Markov models.

[0025] In a further embodiment, the method may comprise implementing at least one pharmaco - mathematical (PMX) model. The at least one PMX model may have a computer - implemented PMX model including at least one of a mathematical pharmacokinetic - pharmacodynamic (PK - PD) model; a physiology - based PK (PBPK) model; a physiology - based PK - PD (PBPKPD) model; a drug exposure - efficacy response model; and a drug exposure - safety response model.

[0026] At least one health state of the patient may include PE, gestational diabetes, a state related to fetal growth, and / or a state related to gestational hypertension. In one embodiment, the step of predicting at least one health state based on at least one health state hypothesis may include using at least one fetal - growth - related data. In another embodiment, the step of predicting at least one health state may include using at least one fetal - growth - related data, where the at least one health state hypothesis may be based on at least one fetal - growth - related data. The at least one fetal - growth - related data may be obtained from at least one of at least one database. In a further embodiment, the step of predicting at least one health state may include using at least one PE - related data, where the at least one health state hypothesis may be based on at least one PE - related data. The at least one PE - related data may be obtained from at least one of at least one database.

[0027] Furthermore, the method may comprise determining at least one medicament based on the at least one health state, wherein the at least one medicament may be suitable for preventing the occurrence and / or recurrence of the at least one medical condition and / or the at least one health state. The method may comprise determining at least one medicament based on the at least one health state, wherein the at least one medicament may be suitable for treating the at least one medical condition and / or the at least one health state. The at least one medicament may comprise at least one of the following: aspirin, ibuprofen; at least one corticosteroid drug; at least one antihypertensive drug; and at least one cardiovascular-related drug. Also, the method may comprise generating at least one medicament administration route, wherein the at least one medicament administration route may comprise at least one of the following: intravenous; intramuscular; subcutaneous; inhalation; transdermal; transcutaneous; oral; rectal; and sublingual. The method may comprise generating at least one administration regimen for the at least one medicament. Additionally or alternatively, the method may further comprise optimizing the at least one administration regimen, wherein the at least one administration regimen may have at least one of the following: the at least one medicament; the administration route of the at least one medicament; at least one dosing scheme; at least one medicament administration period; and at least one medicament administration frequency. The step of optimizing the at least one administration regimen may be based on at least one health state hypothesis. The method may also comprise implementing at least one optimal control theory, wherein the at least one optimal control theory is computer-implemented. It should be understood that the method described herein is a computer-implemented method. The method may comprise optimizing at least one ongoing treatment of at least one medical condition. The method may comprise optimizing at least one ongoing treatment of at least one potential medical condition. The optimizing step may be based on at least one health state hypothesis and / or at least one health state.The method may comprise a step of generating at least one treatment proposal, where the at least one treatment proposal may be based on at least one health state hypothesis and / or at least one health state. The method may comprise a step of optimizing at least one treatment proposal, where the method may comprise a step of performing an optimization step after executing the at least one treatment proposal.

[0028] In one embodiment, the method may comprise a step of adapting any of the embodiments of the aforementioned method to a patient and a step of generating at least one individual treatment protocol, where the at least one individual treatment protocol may be based on at least one health state of the patient. The method may comprise a step of optimizing at least one individual treatment protocol, where the method may comprise a step of performing an optimization step after executing the at least one individual treatment protocol. Further, the method may comprise a step of implanting any of the aforementioned optimization steps assisted by a computer-implemented pharmacodynamic approach. The method may comprise a step of performing the method described herein in the absence of a patient. Further, the method may comprise a step of performing the method described herein using at least one historical data. The at least one historical data may be the patient's historical data. The at least one historical data may have data from at least one of the following: a public health database, the patient's personal database, a medical expert's database, a healthcare provider's database, and a private data bank.

[0029] In another embodiment, the method may comprise at least one of the following: a step of capturing at least one image data of a patient; and a step of receiving at least one image data of the patient, where the at least one image data may include data related to at least one medical condition and / or at least one potential medical condition of the patient. Also, the method is suitable for implementation in at least one medical device such as an ultrasonic device.

[0030] In a second aspect, the present invention is a system for predicting a patient's health state, the system receiving at least one patient-related kinetic characteristic data, receiving at least one patient-related covariate, and processing the at least one patient-related kinetic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set; at least one analysis component configured to analyze the patient-related processed data set and generate at least one health state hypothesis based on the patient-related processed data set, wherein the system is configured to predict at least one health state based on the at least one health state hypothesis, regarding the system.

[0031] Furthermore, the system may comprise at least one storage component configured to store data related to the at least one health state of the patient. The system may also comprise at least one computing component configured to implement a kinetic model for predicting the at least one health state. The at least one health state hypothesis may have a correlation with at least one medical condition of the patient.

[0032] The patient may be a neonate, a fetus, and / or a female, where the female may be at least one of a pregnant female and a non-pregnant female.

[0033] The system may be configured to predict the kinetic behavior of at least one health state of the patient. The system may be configured to implement any of the steps described in any of the embodiments of the foregoing method by at least one machine learning technique.

[0034] At least one patient-related covariate may have at least one biomarker. The at least one biomarker may relate to at least one medical condition, and the at least one biomarker may include at least one of the following: soluble Fms-like tyrosine kinase-1 (sFlt-1); placental growth factor (PlGF); neurofilament (NfL); C-terminal portion of arginine vasopressin (copeptin); albumin; liver transaminase; urea; hemoglobin; platelets; creatinine; albuminuria; proteinuria; estimated glomerular filtration rate (eGFR); creatinine clearance (CrCl); at least one additional renal function measurement; placental biomarkers such as placental RNA, placental protein; endothelial / cardiovascular biomarkers such as endothelial RNA, endothelial protein; or any combination thereof.

[0035] At least one patient-related covariate may have at least one neonate-related covariate including at least one of the following: gender; race; birth weight; gestational age; mode of birth such as vaginal, vacuum extraction, cesarean section; body temperature; heart rate; respiratory rate; pH value; umbilical cord pH value; respiratory assistance; oxygen requirement; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); Apgar score; and measurement values of at least one of at least one biomarker. Also, at least one patient-related covariate may have at least one mother-related covariate including at least one of the following: age; race; premature rupture of membranes; body temperature; risk factors such as diabetes, obesity, pregnancy history, delivery history, white blood cells; and measurement values of at least one of at least one biomarker. Additionally or alternatively, at least one patient-related covariate may include at least one fetus-related covariate including at least one of the following: gender; fetal weight during pregnancy; fetal biometric parameters such as femur length, abdominal circumference, head circumference, mid-thigh circumference, and transverse diameter; proportion of small-for-gestational-age; gestational age; heart rate; heart rate variability; respiratory rate; uteroplacental perfusion parameters; and measurement values of at least one of the at least one biomarker. Further, at least one patient-related covariate may include at least one environmental covariate including at least one of the following: country of residence; country of birth; date and time of birth; humidity conditions at birth; and ambient temperature at birth.

[0036] The system may be configured to generate at least one threshold, where the at least one threshold represents an indicator of at least one potential medical condition. The system may be configured to output at least one potential medical condition, where the at least one potential medical condition may include at least one of the following: seizure; respiratory; cardiovascular; blood dysfunction; endocrine; renal; hepatic; uteroplacental dysfunction; fetal growth retardation; unplanned preterm birth; placental abruption; hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome; and eclampsia. At least one analysis component may be configured to determine a minimum threshold for the at least one threshold and to determine a maximum threshold for the at least one threshold. If at least one patient-related data may be below the minimum threshold, the at least one analysis component outputs a monitoring proposal. If at least one patient-related data may be above the maximum threshold, the at least one analysis component outputs a treatment proposal. At least one analysis component may be configured to determine a baseline for at least one patient-related data. At least one analysis component may be configured to determine at least one intermediate threshold, where the at least one intermediate threshold may include at least one value between the minimum threshold and the maximum threshold. At least one analysis component may correlate each of the at least one intermediate threshold ranges with the at least one medical condition, generate an interpreted data set based on the correlating step, and be configured to output an automated report indicating at least one potential medical condition. At least one analysis component may be configured to determine at least one medical condition change indicator, monitor a change in the at least one medical condition change indicator, generate at least one medical condition change indicator trend, and predict a progression of at least one of the at least one medical conditions based on the at least one medical condition change indicator trend.

[0037] In a further embodiment, the system may comprise at least one monitoring component configured to monitor a change in value of at least one of the at least one patient-related characteristic, the at least one monitoring component further recording an initial value of the at least one patient-related characteristic, recording at least one subsequent value of the at least one patient-related characteristic, comparing the initial value with at least one of the at least one subsequent values to generate comparison value data, and being configured to output a hypothesis of the patient-related characteristic based on the comparison value data. The at least one monitoring component may be configured to record a current value of one of the at least one patient-related characteristics, the current value being different from the initial value. The system may be a non-diagnostic system. In another embodiment, the system may be a diagnostic system.

[0038] The system may be configured to perform the steps of the method according to any of the embodiments of the method described above using data from at least one database. The at least one database may comprise at least one of a public health database, a patient's personal database, a medical expert database, a healthcare provider database, and a private data bank. Further, the system may be configured to supply data to the at least one server, train the computer-implemented dynamic model based on the data supplied to the at least one server, and generate an adjustment function based on the training data, where the adjustment function may be suitable for adjusting any configuration of the system according to any of the embodiments of the system described above. The system may be configured to trigger at least one action recommendation based on the at least one health state hypothesis and / or at least one health state. The system may be configured to display the at least one action recommendation to a user.

[0039] In one embodiment, the system may be configured to prompt the user to input at least one of acceptance of at least one of the at least one action proposal and rejection of at least one of the at least one action proposal. If the user rejects at least one of the at least one action proposal, the system may be configured to prompt the user to provide at least one annotation. The computer-implemented dynamic model may be based on: Bayesian statistical methods, artificial neural network (ANN) methods, convolutional neural network (CNN) methods, recurrent neural network (RNN) methods, pharmacometrics (PMX) methods, supervised learning methods, deep learning (DL) and / or multi-layer neural network methods, and / or explainable AI (XAI) concepts.

[0040] At least one medical condition may include at least one of the following: a fetal growth-related condition; neonatal hypothyroidism; a PE-related condition; a gestational diabetes-related condition; a gestational hypertension-related condition; and gestational hypothyroidism. The system may be configured to correlate at least one biomarker to at least one medical condition, where the at least one medical condition may have an underlying disease. The system may also be configured to predict the occurrence of at least one hypothesis over a given period, where the system may further be configured to recognize a plurality of different periods including at least one of a prenatal period, pregnancy, a delivery period, and a postnatal period. The system may be configured to output the likelihood of occurrence correlated to each of the periods. Further, the system may be configured to execute at least one machine learning (ML) algorithm, where the at least one ML algorithm may have a supervised algorithm architecture, an unsupervised algorithm architecture, or any combination thereof. The at least one ML algorithm may include at least one artificial deep learning (DL) architecture, where the at least one artificial DL architecture may include at least one of an ANN, a CNN, and an RNN. The unsupervised algorithm architecture may be configured to implement at least one clustering technique of at least one cluster.

[0041] The system may also be configured to execute at least one analysis method, where the at least one analysis method may include at least one of pattern recognition, probabilistic modeling, Bayesian schemes, reinforcement learning, statistical analysis, statistical models, principal component analysis (PCA), independent component analysis, dynamic time warping, maximum likelihood estimation (MLE), modeling, estimation, neural networks (NN), convolutional neural networks (CNN), recurrent neural networks (RNN), deep convolutional networks, deep learning (DL), ultra-deep learning, genetic algorithms, Markov models, and / or hidden Markov models. The system may be configured to implement at least one pharmacodynamics (PMX) model, and the at least one PMX model may include at least one of the following: mathematical-statistical pharmacokinetic-pharmacodynamic (PK-PD) models; physiologically based PK (PBPK) models; physiologically based PK-PD (PBPKPD) models; drug exposure-efficacy response models; and drug exposure-safety response models.

[0042] The system may be configured to predict at least one health state based on at least one health state hypothesis and use at least one fetal growth-related data. The system may be configured to predict at least one health state using at least one fetal growth-related data, where the at least one health state hypothesis may be based on the at least one fetal growth-related data. The at least one fetal growth-related data may be obtained from at least one of at least one database. The system may be configured to predict at least one health state and may include using at least one PE-related data, where the at least one health state hypothesis may be based on the at least one PE-related data. The at least one PE-related data may be obtained from at least one of at least one database.

[0043] The system may comprise at least one imaging component configured to perform at least one of capturing at least one image data of the patient and receiving at least one image data of the patient, wherein the at least one image data may include data related to at least one medical condition and / or at least one potential medical condition of the patient.

[0044] The system is further configured to perform any of the steps of the methods described herein.

[0045] The system may comprise at least one implementation component configured to connect the system to at least one medical device, such as an ultrasonic device, wherein the system, once connected to the at least one medical device, is configured to perform any of the steps described in the methods described herein. The system may be configured to operate in the absence of a patient.

[0046] Furthermore, the method comprises the step of using the system described herein to perform any of the steps of the methods described herein.

[0047] In a third aspect, the invention relates to a method of treatment for treating a medical condition of a patient, wherein the treatment has the step of generating a treatment protocol comprising at least one therapeutic agent and a treatment regimen, and the treatment regimen is based on at least one health state hypothesis. The at least one health state hypothesis may be provided by the methods described herein. The at least one agent may be provided by the methods described herein. The at least one health state of the patient may include PE, gestational diabetes, and / or fetal growth problems.

[0048] The treatment may further include the step of treating the patient for at least one potential medical condition prior to the onset of at least one medical condition. The patient may be at least one of a pregnant woman and a non-pregnant woman, a fetus and / or a neonate.

[0049] In a fourth aspect, the present invention relates to a diagnostic method for diagnosing a patient's medical condition, where the diagnosis comprises generating at least one diagnostic finding comprising at least one medical condition of the patient, and where the at least one diagnostic finding is based on at least one health state hypothesis. The diagnostic method may comprise generating at least one treatment method, and the at least one treatment method may be for treating at least one medical condition of the patient. The diagnostic method may comprise generating at least one diagnostic finding, and the at least one diagnostic finding may comprise at least one medical condition of the patient prior to the onset of at least one medical condition. The diagnostic method may comprise at least one preventive method of treatment, and the at least one preventive method of treatment may be for treating at least one medical condition of the patient prior to the onset of at least one medical condition. The at least one health state hypothesis may be provided by the method according to any of the embodiments of the methods described above.

[0050] The diagnostic method may comprise providing at least one medicament, and the at least one medicament may be provided by the method according to any of the embodiments of the methods described above. The at least one health state of the patient may comprise PE, pregnancy-induced hypertension, gestational diabetes, and / or fetal growth problems. The patient may be at least one of a pregnant woman and a non-pregnant woman, a fetus and / or a neonate. The diagnostic method may comprise proposing a treatment method according to any of the treatment methods described in the embodiments of the treatment methods described above.

[0051] In a fifth aspect, the present invention relates to the use of the system described herein for carrying out the methods described herein. The method may comprise prompting the system described herein to perform the steps of the method described herein. Use of the method described herein for implementing the methods of treatment described herein. Use of the method described herein for implementing the diagnostic methods described herein. Use of the method described herein for implementing the diagnostic methods and the methods of treatment described herein, where implementing the diagnostic method precedes implementing the method of treatment.

[0052] Briefly stated, the present invention relates to predicting diseases in the field of perinatal pharmaceuticals. More specifically, the method of the present invention enables the combination of different components such as machine learning, data augmentation, artificial intelligence, dynamic pharmacology, and pharmaceutics, which are suitable for neonatology and obstetrics. The present invention also enables the use of a plurality of PE-related biomarkers to detect and monitor stress factors in the mother, fetus, and neonate over time in clinical studies, for example, over the past 15 years. Since PE is a progressive multi-system disease, the present invention enables the examination of a plurality of PE-related biomarkers, such as (i) cardiovascular markers in triage (Wellmann, 2014), (ii) biomarkers for detecting and monitoring preclinical maternal end-organ dysfunction such as copeptin in the renal system (Wellmann, 2014) and NfL for the central nervous system (Evers, 2018), and (iii) biomarkers for the diagnosis and monitoring of fetal stress response (Burkhardt, 2012) and fetal adverse outcomes (Letzner, 2011), (Depoorter, 2018). Such a combined analysis of biomarkers enables the present invention to predict the health status of a patient, which is particularly advantageous. It should be understood that the prediction of the health status may include the current, future, and / or past health status of the patient. That is, the present invention may be able to predict the future health status of a patient before the onset of a medical condition, and may also be able to predict the current health status of a patient before the onset of a medical condition. Together, the present invention provides an integrated approach that combines and utilizes multi-dimensional time-series data, computer-implemented data processing, and the use of AI and PMX-based computer-implemented models to personalize and optimize the prevention, diagnosis, management, and treatment of PE. The present invention is also advantageous in that it can avoid PE-related complications in a patient or group of patients, such as complications in the mother and her unborn and born children.Thus, the present invention combines available multi-source inputs to improve the perinatal prevention, diagnosis, and management of PE and its complications, where the methods of the present invention can be implemented without the need for human intervention. The reason is that computer-implemented methods enable the utilization of sequential multi-source data to optimize the prevention, diagnosis, and management of diseases, provide solutions that can reduce morbidity not only before birth (i.e., mother and fetus) but also after birth (i.e., mother and neonate), and through the intellectual integration of concepts of multiple components including clinical data, biomarkers, uterine-placental perfusion, and fetal growth data, signal processing data along with measurements over time, data integration at all levels is achieved, and ML and other AI methods are combined with pharmacological principles and computer models of innovative pharmacokinetics and pharmacodynamics, and by utilizing computer-implemented simulation techniques of pharmacodynamics, dosing is optimized and personalized to maximize the balance of efficacy / safety for not only the mother but also her unborn and born children. This approach is particularly advantageous as it provides a more accurate, effective, and efficient method, corresponding system or method for predicting a patient's health state, with improved sensitivity and performance and less likely to result in an incorrect prediction of at least one of the patient's health states.

[0053] The present technology is also described by the following numbered embodiments.

[0054] The following method embodiments are considered. These embodiments are abbreviated by the letter "M" followed by a number. When referring to method embodiments in this specification, these embodiments are meant.

[0055] M1. A method for predicting a patient's health state, the method comprising: receiving at least one patient-related kinetic characteristic data; receiving at least one patient-related covariate; Processing the at least one patient-related kinetic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set. Generating at least one health state hypothesis based on the patient-related processed data set, and Predicting at least one health state based on the at least one health state hypothesis A method comprising:

[0056] M2. The step of predicting the at least one health state is based on a computer-implemented kinetic model, the method according to the foregoing embodiments.

[0057] M3. The at least one health state hypothesis has a correlation with at least one medical condition of the patient, the method according to any of the foregoing embodiments of the method.

[0058] M4. The method according to any of the foregoing embodiments, wherein the patient is a female patient.

[0059] M5. The method according to the foregoing embodiment, wherein the female patient is a pregnant woman.

[0060] M6. The method according to any of the foregoing two embodiments, wherein the female patient includes non-pregnant women.

[0061] M7. The method according to any of the foregoing embodiments of the method, wherein the patient is a fetus.

[0062] M8. The method according to any of the foregoing embodiments of the method, wherein the patient is a neonate.

[0063] M9. The method according to any of the foregoing embodiments of the method, further comprising predicting the kinetic behavior of at least one health state of the patient.

[0064] The method according to any of the embodiments of the foregoing method comprises a step of implementing at least one machine learning technique, and the method comprises a step of using the at least one machine learning technique to implement any of the foregoing steps.

[0065] The method according to any of the embodiments of the foregoing method, wherein at least one patient-related covariate comprises at least one biomarker.

[0066] The method according to the foregoing embodiment, wherein at least one biomarker is related to at least one medical condition.

[0067] The method according to any of the foregoing three embodiments, wherein at least one biomarker comprises at least one of soluble Fms-like tyrosine kinase-1 (sFlt-1); placental growth factor (PlGF); neurofilament (NfL); the C-terminal portion of arginine vasopressin (copeptin); albumin; liver transaminase; urea; hemoglobin; platelets; creatinine; albuminuria; proteinuria; estimated glomerular filtration rate (eGFR); creatinine clearance (CrCl); at least one additional renal function measurement; placental biomarkers such as placental RNA, placental protein; endothelial / cardiovascular biomarkers such as endothelial RNA, endothelial protein; or any combination thereof.

[0068] M14. At least one patient-related covariate includes age; weight; height; body mass index (BMI); pregnancy history; delivery history; number of fetuses in the current pregnancy; race; body temperature; heart rate; heart rate variability; respiratory rate; early rupture of membranes; white blood cell count; history of PE (family and maternal), gestational diabetes, obesity, cardiovascular / renal / liver / thyroid diseases, autoimmune diseases, anemia, antiphospholipid syndrome, sexually transmitted infections, coexisting conditions such as headache; smoking habit before and / or during pregnancy; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); uteroplacental perfusion parameters; Doppler measurements of pulsatility indices of umbilical artery, middle cerebral artery, cerebroplacental ratio, uterine artery, fetal descending aorta, ductus venosus, umbilical vein, inferior vena cava, uterine artery; soft tissue parameters such as upper arm partial volume and thigh partial volume; and at least one maternal-related covariate including at least one of at least one of the measurements of said at least one biomarker, a method having the features of embodiments M4 to M6 as described in any of the embodiments of the foregoing method.

[0069] M15. At least one patient-related covariate includes sex; fetal weight during pregnancy; fetal biometric parameters such as femur length, abdominal circumference, head circumference, mid-thigh circumference, and transverse diameter; proportion of small for gestational age; gestational age at birth; heart rate; heart rate variability; respiratory rate; uteroplacental perfusion parameters; and at least one fetal-related covariate including at least one of at least one of the measurements of said at least one biomarker, a method having the features of embodiment M7 as described in any of the embodiments of the foregoing method.

[0070] M16. At least one patient-related covariate includes sex; birth weight; weight; height; gestational age at birth; age after delivery; body temperature; heart rate, heart rate variability; respiratory rate; breast milk; duration of exclusive breast milk; pH value; respiratory support; oxygen requirement; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); Apgar score; at least additional neonatal biometric parameters; and at least one neonatal-related covariate including at least one of at least one of the measurements of said at least one biomarker, a method having the features of embodiment M8 as described in any of the embodiments of the foregoing method.

[0071] M17. In an embodiment of the method described above, at least one patient-related covariate includes at least one environmental covariate including at least one of country of residence, country of birth, date and time of birth, humidity conditions at birth, and ambient temperature at birth.

[0072] M18. The method comprises the step of generating at least one threshold, wherein the at least one threshold represents an indicator of at least one potential medical condition, in any of the embodiments of the method described above.

[0073] M19. The method comprises the step of outputting at least one potential medical condition, wherein the at least one potential medical condition includes at least one of seizure; respiratory; cardiovascular; blood dysfunction; endocrine; kidney; liver; uteroplacental dysfunction; fetal growth retardation; unplanned preterm birth; placental abruption; hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome; and eclampsia, in any of the embodiments of the method described above.

[0074] M20. The method comprises the step of determining a minimum threshold for the at least one threshold, and the step of determining a maximum threshold for the at least one threshold in any of the embodiments of the method described above.

[0075] M21. If at least one patient-related data is below the minimum threshold, the method comprises the step of outputting a monitoring proposal, in the method described in the above embodiment.

[0076] M22. If at least one patient-related data is above the maximum threshold, the method comprises the step of outputting a treatment proposal, in any of the two embodiments described above.

[0077] M23. The method comprises the step of determining a baseline for the at least one patient-related data, in any of the three embodiments described above.

[0078] The method of M24 comprises the step of determining at least one intermediate threshold, where the at least one intermediate threshold includes at least one value between the minimum threshold and the maximum threshold, and is the method according to any of the four foregoing embodiments.

[0079] M25. The method correlating at least a respective range of each of the at least one intermediate threshold with the at least one medical condition, generating an interpreted data set based on the correlating step, and outputting an automated report indicating the at least one potential medical condition and is the method according to the foregoing embodiments.

[0080] M26. The method comprises determining at least one medical condition change indicator, monitoring a change in the at least one medical condition change indicator, generating at least one medical condition change indicator trend, and predicting the progression of at least one of the at least one medical condition wherein the prediction is based on the at least one medical condition change indicator trend, and is the method according to any of the embodiments of the foregoing method.

[0081] M27. The method comprises monitoring a change in at least one value of at least one of the at least one patient-related characteristic, and the method records an initial value of the at least one patient-related characteristic, records at least one subsequent value of the at least one patient-related characteristic, compares the initial value with at least one of the at least one subsequent value, generates comparison value data, and outputs a hypothesis of the patient-related characteristic based on the comparison value data The method according to any of the embodiments of the aforementioned method, comprising

[0082] M28. The step of recording at least subsequent values has a step of recording a current value of one of the at least one patient-related characteristic, the current value being different from the initial value, the method according to the embodiments described above.

[0083] M29. The method according to any of the embodiments of the aforementioned method, wherein the method is a non-diagnostic method.

[0084] M30. The method according to any of the embodiments of the aforementioned method, wherein the method is a diagnostic method.

[0085] M31. The method according to any of the embodiments of the aforementioned method, comprising the step of performing the steps of the method according to any of the embodiments described above using data from at least one database.

[0086] M32. The method according to the embodiments described above, wherein the at least one database has at least one of a public health database, a patient's personal database, a database of medical experts, a database of medical personnel, and a private data bank.

[0087] M33. The method comprising the step of supplying data to the at least one server, training the computer-implemented dynamic model based on the data supplied to the at least one server, and generating an adjustment function based on the training data wherein the adjustment function is suitable for adjusting any step of the method according to any of the embodiments of the aforementioned method, the method according to any of the embodiments of the aforementioned method.

[0088] M34. The method according to any of the embodiments of the aforementioned method, comprising the step of triggering at least one action proposal based on the at least one health state hypothesis.

[0089] M35. The method is the method according to the foregoing embodiment, comprising the step of presenting the at least one action proposal to the user.

[0090] M36. The method is at least one acceptance of at least one of the at least one action proposals, and at least one rejection of at least one of the at least one action proposals The method according to any of the foregoing two embodiments, comprising the step of prompting the user to input at least one of them.

[0091] M37. If the user rejects at least one of the at least one action proposals, the method is the method according to the foregoing embodiment, comprising the step of prompting the user to provide at least one annotation.

[0092] M38. The computer-implemented dynamic model is the method according to any of the foregoing method embodiments, based on Bayesian statistical techniques.

[0093] M39. The computer-implemented dynamic model is the method according to any of the foregoing method embodiments, based on ANN, CNN, or RNN techniques.

[0094] M40. The computer-implemented dynamic model is the method according to any of the foregoing method embodiments, based on pharmacodynamic modeling and / or simulation techniques.

[0095] M41. The computer-implemented dynamic model is the method according to any of the foregoing method embodiments, based on supervised learning techniques.

[0096] M42. The computer-implemented dynamic model is the method according to any of the foregoing method embodiments, based on deep learning and / or multi-layer neural network techniques.

[0097] The computer-implemented dynamic model is a method according to any of the embodiments of the foregoing method based on the explainable AI concept (XAI).

[0098] At least one medical condition is a method according to any of the embodiments of the foregoing method having at least one of a fetal growth-related condition; a PE-related condition; a gestational diabetes-related condition; a gestational hypertension-related condition; pregnancy medicine such as a cyclooxygenase inhibitor, for example aspirin; at least one related drug treatment; and at least one potential medical condition.

[0099] The method comprises correlating at least one biomarker with at least one medical condition, wherein said at least one medical condition has the features of embodiment M13 of a method according to any of the embodiments of the foregoing method having a potential disease.

[0100] The method comprises predicting the occurrence of at least one hypothesis over a given period, wherein the method further comprises recognizing a plurality of different periods including at least one of a prepartum period, pregnancy, a delivery period, and a postpartum period, a method according to any of the embodiments of the foregoing method.

[0101] The method comprises outputting the likelihood of occurrence correlated with each of the periods, a method according to the foregoing embodiment.

[0102] The method comprises executing at least one machine learning algorithm, a method according to any of the embodiments of the foregoing method.

[0103] The at least one machine learning algorithm has a supervised algorithm architecture, an unsupervised algorithm architecture, or any combination thereof, a method according to the foregoing embodiment.

[0104] M50. The method according to any of the embodiments of the foregoing method, wherein at least one machine learning algorithm includes at least one artificial deep learning (DL) architecture.

[0105] M51. The method according to the foregoing embodiments, wherein at least one artificial deep learning architecture includes at least one of ANN, CNN, and RNN.

[0106] M52. The method having the features of embodiment M49 according to any of the embodiments of the foregoing method, wherein the unsupervised algorithm architecture includes implementing at least one clustering method for at least one cluster.

[0107] M53. The method according to any of the embodiments of the foregoing method, comprising the step of performing at least one analysis method, wherein the at least one analysis method has at least one of pattern recognition, probability modeling, Bayesian scheme, reinforcement learning, statistical analysis, statistical model, principal component analysis, independent component analysis, dynamic time warping, maximum likelihood estimation, modeling, estimation, neural network, convolutional network, recurrent network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithm, Markov model, and / or hidden Markov model.

[0108] M54. The method according to any of the embodiments of the foregoing method, comprising the step of implementing at least one pharmacodynamic model.

[0109] M55. The method according to the foregoing embodiments, wherein at least one pharmacodynamic model has a computer-implemented pharmacodynamic model including at least one of a mathematical-statistical PKPD model; a physiology-based PK (PBPK) model; a physiology-based PKPD (PBPKPD) model; a drug exposure-efficacy response model; and a drug exposure-safety response model.

[0110] M56. The method according to any of the embodiments of the foregoing method, wherein at least one health state of the patient includes PE.

[0111] M57. At least one health state of the patient is the method according to any of the embodiments of the foregoing method, including gestational diabetes.

[0112] M58. The method according to any of the embodiments of the foregoing method, wherein at least one health state of the patient includes a state related to fetal growth.

[0113] M59. At least one health state of the patient is the method according to any of the embodiments of the foregoing method, including a state related to gestational hypertension.

[0114] M60. The step of predicting at least one health state based on at least one health state hypothesis includes the step of using at least one fetal growth-related data, and is the method according to any of the embodiments of the foregoing method.

[0115] M61. The step of predicting at least one health state includes the step of using at least one fetal growth-related data, and at least one health state hypothesis is based on at least one fetal growth-related data, and is the method according to any of the embodiments of the foregoing method.

[0116] M62. At least one fetal growth-related data is obtained from at least one of at least one database, and the method has the characteristics of Embodiment M32 described in the foregoing embodiment.

[0117] M63. The step of predicting at least one health state includes the step of using at least one PE-related data, and at least one health state hypothesis is based on at least one PE-related data, and is the method according to any of the embodiments of the foregoing method.

[0118] M64. At least one PE-related data is obtained from at least one of at least one database, and the method has the characteristics of Embodiment M32 described in the foregoing embodiment.

[0119] The method according to any of the preceding embodiments comprises the step of determining at least one medicament based on the at least one health state, wherein the at least one medicament is suitable for preventing the occurrence of the at least one medical condition and / or the at least one health state.

[0120] The method according to any of the preceding embodiments comprises the step of determining at least one medicament based on the at least one health state, wherein the at least one medicament is suitable for treating the at least one medical condition and / or the at least one health state.

[0121] The method according to any of the two preceding embodiments, wherein the at least one medicament comprises at least one of aspirin; ibuprofen; at least one corticosteroid drug; at least one antihypertensive drug; and at least one cardiovascular-related drug.

[0122] The method according to any of the three preceding embodiments comprises the step of generating at least one medicament administration route, wherein the at least one medicament administration route comprises at least one of intravenous; intramuscular; subcutaneous; inhalation; transdermal; transcutaneous; oral; rectal; and sublingual.

[0123] The method according to any of the preceding embodiments, having the features of embodiments 45M and 65M to 68M, comprises the step of generating at least one administration regimen for at least one medicament.

[0124] The method further comprises the step of optimizing the at least one administration regimen, wherein the at least one administration regimen has at least one of the at least one medicament, the administration route of the at least one medicament, at least one administration scheme, at least one medicament administration period, and at least one medicament administration frequency, according to the method described in the preceding embodiments.

[0125] The method according to any of the preceding embodiments, comprising the step of optimizing at least one dosing regimen based on said at least one health state hypothesis.

[0126] The method according to any of the two preceding embodiments, comprising the step of implementing at least one optimal control theory, wherein at least the optimal control theory is computer-implemented.

[0127] The method according to any of the embodiments of the method described above, which is a computer-implemented method.

[0128] The method according to any of the embodiments of the method described above, comprising the step of optimizing at least one ongoing treatment of at least one medical condition.

[0129] The method according to any of the embodiments of the method described above, comprising the step of optimizing at least one ongoing treatment of at least one potential medical condition.

[0130] The optimization step is according to any of the two preceding embodiments, based on at least one health state hypothesis and / or at least one health state.

[0131] The method according to any of the embodiments of the method described above, comprising the step of generating at least one treatment proposal, wherein at least one treatment proposal is based on at least one health state hypothesis and / or at least one health state.

[0132] The method according to the preceding embodiment, comprising the step of optimizing at least a treatment proposal, wherein the method comprises the step of performing an optimization step after performing at least one treatment proposal.

[0133] The method is adapting any of the embodiments of the method described above to a patient, and comprising generating at least one individualized treatment protocol, wherein the at least one individualized treatment protocol is a method according to any of the embodiments of the foregoing method, based on at least one health condition of a patient.

[0134] M80. The method comprises at least optimizing the at least one individualized treatment protocol, and wherein the step of optimizing is performed after the step of executing the at least one individualized treatment protocol, according to the method described in any of the foregoing embodiments.

[0135] M81. The method comprises implanting any of the foregoing optimization steps assisted by a computer-implemented pharmacodynamic approach, according to the method described in any of the embodiments of the foregoing method.

[0136] M82. The method comprises performing any of the embodiments of the foregoing method in the absence of a patient, according to any of the foregoing embodiments.

[0137] M83. The method comprises performing any of the embodiments of the foregoing method using at least one historical data, according to any of the foregoing embodiments.

[0138] M84. The at least one historical data is patient historical data, according to the method described in the foregoing embodiments.

[0139] M85. The at least one historical data is public health database, patient personal database, medical expert database, medical staff database, and private data bank and having data from at least one of them, according to any of the two foregoing embodiments.

[0140] M86. The method is comprising capturing at least one image data of a patient; and Receiving at least one image data of a patient The method according to any of the embodiments of the foregoing method, comprising at least one of the following, wherein the at least one image data has data related to at least the medical condition and / or at least one potential medical condition of the patient.

[0141] The method according to any of the foregoing embodiments, which is suitable for implementation in at least one medical device such as an ultrasonic device.

[0142] The method according to any of the embodiments of the foregoing method, comprising using a system according to any of the embodiments of the system for performing any of the steps according to any of the embodiments of the method.

[0143] Hereinafter, embodiments of the system will be considered. These embodiments are abbreviated by the letter "S" followed by a number. Whenever reference is made to embodiments of the system in this specification, these embodiments are meant.

[0144] S1. A system for predicting a patient's health state, the system comprising Receiving at least one patient-related kinetic characteristic data Receiving at least one patient-related covariate Processing the at least one patient-related kinetic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set At least one processing component configured to Analyzing the patient-related processed data set Generating at least one health state hypothesis based on the patient-related processed data set At least one analysis component configured to Comprising Here, the system is configured to predict at least one health state based on the at least one health state hypothesis.

[0145] S2. The system according to the foregoing embodiment includes at least one memory component configured to store data related to the at least one health state of the patient.

[0146] S3. The system according to the foregoing embodiment includes at least one computing component configured to implement a dynamic model for predicting the at least one health state.

[0147] S3. The at least one health state hypothesis has a correlation with at least one medical condition of the patient, for the system according to any of the embodiments of the foregoing system.

[0148] S4. The patient is a female patient, for the system according to any of the embodiments of the foregoing system.

[0149] S5. The female patient is a pregnant woman, for the system according to the foregoing embodiment. S6. The female patient includes non-pregnant women, for the system according to any of the two foregoing embodiments. S7. The patient is a fetus, for the system according to any of the embodiments of the foregoing system.

[0150] S8. The patient is a newborn, for the method according to any of the embodiments of the foregoing method.

[0151] S9. The system is configured to predict the dynamic behavior of at least one health state of the patient, for the system according to any of the embodiments of the foregoing system.

[0152] S10. The system is configured to implement at least one machine learning technique, where the system is configured by the at least one machine learning technique to implement any of the steps described in any of the embodiments of the foregoing method, and is the system described in any of the embodiments of the foregoing system.

[0153] S11. The system described in any of the embodiments of the foregoing system includes at least one patient-related covariate that includes at least one biomarker.

[0154] S12. The system described in the foregoing embodiment, where at least one biomarker is related to at least one medical condition.

[0155] S13. The system described in any of the foregoing three embodiments, where at least one biomarker includes at least one of soluble Fms-like tyrosine kinase-1 (sFlt-1); placental growth factor (PlGF); neurofilament light chain (NfL); copeptin; placental biomarkers such as placental RNA, placental protein; endothelial / cardiovascular biomarkers such as endothelial RNA, endothelial protein; or any combination thereof.

[0156] S14. The system described in any of the embodiments of the foregoing system, having the characteristics of embodiment S4, where at least one patient-related covariate includes at least one neonatal-related covariate that includes at least one of gender; race, birth weight, gestational age, mode of birth such as vaginal, vacuum extraction, cesarean section, body temperature, heart rate, respiratory rate, pH value; umbilical cord pH value; respiratory support; oxygen requirement; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); Apgar score; and at least one measurement of at least one of the at least one biomarker.

[0157] S15. At least one patient-related covariate includes at least one mother-related covariate having at least one of age, race, premature rupture of membranes, body temperature, diabetes, obesity, parity, gravidity, risk factors such as white blood cells, and at least one measurement of at least one of the at least one biomarker, and has the features of Embodiment S5, according to any of the embodiments of the system described above.

[0158] S16. At least one patient-related covariate includes at least one fetal-related covariate having gender; fetal weight during pregnancy; fetal biometric parameters such as femur length, abdominal circumference, head circumference, mid-thigh circumference, and transverse diameter; percentage of intrauterine growth restriction; gestational age; heart rate; heart rate variability; respiratory rate; uteroplacental perfusion parameters; and at least one measurement of at least one of the at least one biomarker, and has the features of Embodiment S6, according to any of the embodiments of the system described above.

[0159] S17. At least one patient-related covariate includes at least one environmental covariate having at least one of country of residence, country of birth, time of birth, humidity conditions at birth, and ambient temperature at birth, according to any of the embodiments of the system described above.

[0160] S18. The system is configured to generate at least one threshold, where the at least one threshold represents an indicator of at least one potential medical condition, according to any of the embodiments of the system described above.

[0161] S19. The system is configured to output at least one potential medical condition, where the at least one potential medical condition includes at least one of seizure; respiratory; cardiovascular; hematologic dysfunction; endocrine; renal; hepatic; uteroplacental dysfunction; fetal growth retardation; preterm birth; placental abruption; hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome; and eclampsia, according to any of the embodiments of the system described above.

[0162] S20. At least one analysis component is configured to determine a minimum threshold for the at least one threshold, and determine a maximum threshold for the at least one threshold for any of the embodiments of the system described above.

[0163] S21. If at least one patient-related data is below the minimum threshold, at least one analysis component outputs a monitoring proposal, for the system described in the above embodiments.

[0164] S22. If at least one patient-related data is above the maximum threshold, at least one analysis component outputs a treatment proposal, for any of the two embodiments described above.

[0165] S23. At least one analysis component is configured to determine a baseline for the at least one patient-related data, for any of the three embodiments described above.

[0166] S24. At least one analysis component is configured to determine at least one intermediate threshold, where the at least one intermediate threshold includes at least one value between the minimum threshold and the maximum threshold, for any of the four embodiments described above.

[0167] S25. At least one analysis component is configured to correlate each of the at least one intermediate thresholds with the at least one medical condition, generate an interpreted dataset based on the correlating step, and output an automated report indicating at least one potential medical condition for the system described in the above embodiments.

[0168] S26. At least one analysis component is configured to determine at least one medical condition change indicator, monitor the changes of the at least one disease state change indicator, generate at least one disease state change indicator trend, predict the progression of at least one of the at least one disease states based on the at least one disease state change indicator trend The system according to any of the embodiments of the foregoing system, configured as described above.

[0169] S27. The system includes at least one monitoring component configured to monitor a change in value of at least one of the at least one patient-related characteristic, and the at least one monitoring component further records an initial value of the at least one patient-related characteristic, records at least one subsequent value of the at least one patient-related characteristic, compares the initial value with at least one of the at least one subsequent value, generates comparison value data, outputs a hypothesis of the patient-related characteristic based on the comparison value data The system according to any of the embodiments of the foregoing system, configured as described above.

[0170] S28. The at least one monitoring component is configured to record a current value of one of the at least one patient-related characteristic, and the current value is different from the initial value, the system according to the foregoing embodiment.

[0171] S29. The system is a non-diagnostic system, the system according to any of the embodiments of the foregoing system.

[0172] S30. The system is a diagnostic system, the system according to any of the embodiments of the foregoing system.

[0173] S31. The system is a system according to any of the foregoing embodiments of the system, configured to execute the steps of the method according to any of the embodiments of the foregoing method using data from at least one database.

[0174] S32. The system according to the foregoing embodiment, wherein the at least one database has at least one of a public health database, a patient personal database, a medical expert database, a healthcare professional database, and a private data bank.

[0175] S33. The system is configured to supply data to the at least one server, train the computer-implemented dynamic model based on the data supplied to the at least one server, generate an adjustment function based on the training data wherein the adjustment function is suitable for adjusting any configuration of the system according to any of the embodiments of the foregoing system.

[0176] S34. The system is a system according to any of the foregoing embodiments of the system, configured to trigger at least one action proposal based on the at least one health state hypothesis.

[0177] S35. The system is a system according to the foregoing embodiment, configured to display the at least one action proposal to a user.

[0178] S36. The system is configured to prompt the user to input at least one of acceptance of at least one of the at least one action proposals and rejection of at least one of the at least one action proposals wherein the system is according to any of the two foregoing embodiments.

[0179] S37. If the user rejects at least one of the at least one action proposal, the system is the system described in the foregoing embodiments configured to prompt the user to provide at least one annotation.

[0180] S38. The computer-implemented dynamic model is the system described in any of the foregoing embodiments of the system based on Bayesian statistical methods.

[0181] S39. The computer-implemented dynamic model is the system described in any of the foregoing embodiments of the system based on ANN, CNN, or RNN methods.

[0182] S40. The computer-implemented dynamic model is the system described in any of the foregoing embodiments of the system based on pharmacodynamic methods.

[0183] S41. The computer-implemented dynamic model is the system described in any of the foregoing embodiments of the system based on supervised learning methods.

[0184] S42. The computer-implemented dynamic model is the system described in any of the foregoing embodiments of the system based on deep learning and / or multi-layer neural network methods.

[0185] S43. The computer-implemented dynamic model is the system described in any of the foregoing embodiments of the system based on explainable AI concepts (XAI).

[0186] S44. The at least one medical condition includes at least one of a fetal growth-related condition; neonatal hypothyroidism; a PE-related condition; a gestational diabetes-related condition; a gestational hypertension-related condition; and gestational hypothyroidism, in any of the foregoing embodiments of the system.

[0187] S45. The system is configured to correlate at least one biomarker with at least one medical condition, where the at least one medical condition has a potential disease and is a system having the features of Embodiment S13 described in any of the embodiments of the aforementioned system.

[0188] S46. The system is configured to predict the occurrence of at least one hypothesis during a given period, where the system is configured to recognize a plurality of different periods including at least one of the prenatal period, pregnancy, the delivery period, and the postnatal period, and is a system described in any of the embodiments of the aforementioned system.

[0189] S47. The system is configured to output the likelihood of occurrence correlated with each of the periods, and is a system described in the aforementioned embodiment.

[0190] S48. The system is configured to execute at least one machine learning algorithm, and is a system described in any of the embodiments of the aforementioned system.

[0191] S49. The at least one machine learning algorithm has a supervised algorithm architecture, an unsupervised algorithm architecture, or any combination thereof, and is a system described in the aforementioned embodiment.

[0192] S50. The at least one machine learning algorithm includes at least one artificial deep learning (DL) architecture, and is a system described in any of the embodiments of the aforementioned system.

[0193] S51. The at least one artificial deep learning architecture includes at least one of ANN, CNN, and RNN, and is a system described in the aforementioned embodiment.

[0194] The teacherless algorithm architecture is a system having the features of Embodiment S47, described in any of the embodiments of the aforementioned system, configured to implement at least one clustering method of at least one cluster.

[0195] The system is configured to execute at least one analysis method, and the at least one analysis method has at least one of pattern recognition, probability modeling, Bayesian scheme, reinforcement learning, statistical analysis, statistical model, principal component analysis, independent component analysis, dynamic time warping, maximum likelihood estimation, modeling, estimation, neural network, convolutional network, recurrent network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithm, Markov model, and / or hidden Markov model, and is a system described in any of the embodiments of the aforementioned system.

[0196] The system is a system described in any of the embodiments of the aforementioned system, configured to implement at least one pharmacodynamic model.

[0197] The system having at least one pharmacodynamic model includes a PKPD model, and is a system described in the aforementioned embodiments.

[0198] The system where at least one health state of a patient includes PE is a system described in any of the embodiments of the aforementioned system.

[0199] The system where at least one health state of a patient includes gestational diabetes is a system described in any of the embodiments of the aforementioned system.

[0200] The system where at least one health state of a patient includes fetal growth-related problems is a system described in any of the embodiments of the aforementioned system.

[0201] The system where at least one health state of a patient includes pregnancy-related hypertension conditions is a system described in any of the embodiments of the aforementioned system. S60. The system is the system according to any of the embodiments of the aforementioned system, configured to predict at least one health state based on at least one health state hypothesis and use at least one fetal growth-related data.

[0202] S61. The system is configured to predict at least one health state using at least one fetal growth-related data, wherein at least one health state hypothesis is based on at least one fetal growth-related data, and it is the system according to any of the embodiments of the aforementioned system.

[0203] S62. The at least one fetal growth-related data is obtained from at least one of at least one database, and it is the system having the features of Embodiment S30 according to the aforementioned embodiments.

[0204] S63. The system is configured to predict at least one health state using at least one PE-related data, wherein at least one health state hypothesis is based on at least one PE-related data, and it is the system according to any of the embodiments of the aforementioned system.

[0205] S64. The at least one PE-related data is obtained from at least one of at least one database, and it is the system having the features of Embodiment S30 according to the aforementioned embodiments.

[0206] S65. The said system is configured to perform at least one of capturing at least one image data of the said patient and receiving at least one image data of the said patient and includes at least one imaging component, wherein the at least one image data includes data related to at least the medical condition and / or at least one potential medical condition of the said patient, and it is the system according to any of the embodiments of the aforementioned system.

[0207] S66. The system is a system according to any of the embodiments of the system described above, configured to perform any of the steps described in any of the method embodiments.

[0208] S67. The system comprises at least one implementation component configured to connect the system to at least one medical device, such as an ultrasonic device, wherein the system, once connected to the at least one medical device, is configured to perform any of the steps described in any of the method embodiments, and is a system according to any of the embodiments of the system described above.

[0209] S68. The system is a system according to any of the embodiments of the system described above, configured to operate in the absence of a patient.

[0210] Hereinafter, embodiments of the treatment method will be considered. These embodiments are abbreviated by the letter "T" followed by a number. When referring to the embodiments of the treatment method in this specification, these embodiments are meant.

[0211] T1. A treatment method for treating a patient's medical condition, wherein the treatment comprises generating a treatment protocol comprising at least one therapeutic agent and a treatment regimen, and the treatment regimen is based on at least one health state hypothesis.

[0212] T2. The at least one health state hypothesis is the treatment according to the foregoing embodiment provided by the method according to any of the embodiments of the method described above.

[0213] T3. The at least one agent is the treatment according to any of the foregoing two embodiments provided by the method according to any of the embodiments of the method described above.

[0214] T4. The at least one health state of the patient is the treatment according to any of the embodiments of the treatment described above, including PE.

[0215] T5. At least one health condition of the patient is a treatment as described in any of the embodiments of the aforementioned treatment, including gestational diabetes and / or gestational hypertension.

[0216] T6. At least one health condition of the patient is a treatment as described in any of the embodiments of the aforementioned treatment, including fetal growth problems.

[0217] T7. The treatment further comprises treating the patient for at least one potential condition of the patient prior to the onset of at least one medical condition, a treatment as described in any of the embodiments of the aforementioned treatment.

[0218] T8. The patient is at least one of a pregnant woman and a non-pregnant woman, a treatment as described in any of the embodiments of the aforementioned treatment.

[0219] T9. The patient is a fetus, a treatment as described in any of the embodiments of the aforementioned treatment.

[0220] T10. The patient is a newborn, a treatment as described in any of the embodiments of the aforementioned treatment.

[0221] Hereinafter, embodiments of a diagnostic method will be considered. These embodiments are abbreviated by the letter "D" followed by a number. In this specification, when referring to embodiments of a diagnosis, these embodiments are meant.

[0222] D1. A diagnostic method for diagnosing a medical condition of a patient, where the diagnosis comprises generating at least one diagnostic finding comprising at least one medical condition of the patient, and where the at least one diagnostic finding is based on at least one health condition hypothesis, a diagnostic method.

[0223] D2. The diagnosis comprises generating at least one method of treatment, where the at least one method of treatment is for treating at least one medical condition of the patient, a diagnosis as described in the aforementioned embodiment.

[0224] Diagnosis comprises a step of generating at least one diagnostic finding, and at least the diagnostic finding comprises at least one medical condition of the patient prior to the onset of at least one medical condition, according to any of the two embodiments described above.

[0225] Diagnosis comprises at least one preventive method of treatment, and at least one preventive method of treatment is for treating at least one medical condition of the patient prior to the onset of at least one medical condition, according to the diagnosis described in the above embodiment.

[0226] At least one health state hypothesis is a diagnosis according to any of the embodiments of the diagnosis described in any of the embodiments of the method described above.

[0227] Diagnosis comprises providing at least one medicament, and at least one medicament is a diagnosis according to any of the embodiments of the diagnosis provided by the method according to any of the embodiments of the method described above.

[0228] At least one health state of the patient is a diagnosis according to any of the embodiments of the diagnosis described above, including PE.

[0229] At least one health state of the patient is a diagnosis according to any of the embodiments of the diagnosis described above, including gestational diabetes and / or gestational hypertension.

[0230] At least one health state of the patient is a diagnosis according to any of the embodiments of the diagnosis described above, including fetal growth problems.

[0231] The patient is at least one of a pregnant woman and a non-pregnant woman, according to any of the embodiments of the diagnosis described above.

[0232] The patient is a fetus, according to any of the embodiments of the diagnosis described above.

[0233] D12. The patient is a neonate, and the diagnosis described in any of the foregoing embodiments of the diagnosis.

[0234] D13. The diagnosis includes proposing a treatment method described in any of the foregoing embodiments of the treatment method, and the diagnosis described in any of the foregoing embodiments of the diagnosis.

[0235] Hereinafter, embodiments of use will be considered. These embodiments are abbreviated by the letter "U" followed by a number. When referring to embodiments of the system in this specification, these embodiments are meant.

[0236] U1. Use of a system described in any of the foregoing embodiments of the system for executing a method described in any of the foregoing embodiments of the method.

[0237] U2. The method includes prompting a system described in any of the foregoing embodiments to perform steps of a method described in any of the foregoing embodiments of the method. Use of a method described in any of the foregoing embodiments of the method.

[0238] U3. Use of a method described in any of the foregoing embodiments of the method for implementing a treatment method described in any of the foregoing embodiments of the treatment method.

[0239] U4. Use of a method described in any of the foregoing embodiments of the method for implementing a diagnostic method described in any of the foregoing embodiments of the diagnostic method.

[0240] U5. Use of a method described in any of the foregoing embodiments of the method for implementing a diagnostic method described in any of the foregoing embodiments of the diagnostic method and a treatment method described in any of the foregoing embodiments of the treatment method, where implementing the diagnostic method precedes implementing the treatment method.

Brief Description of the Drawings

[0241] Here, the present invention will be described with reference to the accompanying drawings showing embodiments of the present invention. These embodiments are merely illustrative of the present invention and should not limit the present invention.

[0242]

Figure 1

Figure 2

Figure 3

Figure 4

[0243] Note that not all reference numerals are shown in all the drawings. Instead, in some drawings, some reference numerals are omitted in order to simplify the illustration for brevity. Here, embodiments of the present invention will be described with reference to the accompanying drawings.

Mode for Carrying Out the Invention

[0244] FIG. 1 schematically shows a system 1000 for predicting a patient's health state. Briefly, the system 1000 includes a processing component 1100, an analysis component 1200, a computing component 1300, a memory component 1400, and a monitoring component 1500. It should be understood that in some embodiments, the system 1000 may include one or more of these components.

[0245] In one embodiment, the memory component 1400 may be an external component such as a remote component. In FIG. 1, this is shown by a dashed line. However, it should be understood that any other component of the system 1000 may also be external. For example, the monitoring component 1500 may be a remote component. When a component of the system 1000 is an external component, it should be understood that this may be assigned to a server (remote or local), or may even be in the cloud.

[0246] The processing component 1100 is configured to receive at least one patient-related kinetic characteristic data, receive at least one patient-related covariate, and process at least one patient-related kinetic characteristic data and at least one patient-related covariate data to generate a patient-related processed data set. That is, the processing component 1100 is tasked with receiving data such as raw data or unprocessed data from different systems such as a database, manual input by a user, or automatic input performed by another device or system. When the processing component 1100 receives data, it can process the data autonomously or at least partially autonomously to generate a patient-related processed data set.

[0247] The analysis component 1200 is configured to analyze a patient-related processed data set and generate at least one health state hypothesis based on the patient-related processed data set.

[0248] In one embodiment of the system 1000, the processing component 1100 and the analysis component 1200 may represent an integrated component.

[0249] System 1000 is configured to utilize a plurality of different data as inputs. Without limitation, and among other things, System 1000 may receive, process, and / or analyze a plurality of biomarkers such as parental, fetal, and / or neonatal biomarkers; a plurality of clinical parameters such as parental, fetal, and / or neonatal clinical parameters; demographics, lifestyle, and psychometric scores regarding a patient and / or group of patients; a plurality of environmental parameters; drug therapies such as current drug therapies administered to a patient and / or drug therapies recommended in valid or ongoing guidelines; dosing regimens, medication histories regarding a patient or group of patients; cardiotocography data (CTG); electroencephalogram data (EEG); electrocardiogram data (ECG), pulse, and / or oxygen measurements; deformation modalities such as sound, and Doppler, duplex, etc.; magnetic resonance imaging (MRI); data provided by X-rays.

[0250] In one embodiment, System 1000 may also include one or more imaging components (not shown) configured to capture an image of a patient that may be related to at least one health and / or medical condition.

[0251] Monitoring component 1500 is configured to monitor System 1000, i.e., the components of System 1000. Further, monitoring component 1500 is configured to monitor at least one change in value of at least one of at least one patient-related characteristic; record an initial value of at least one patient-related characteristic; record at least one subsequent value of at least one patient-related characteristic; compare at least one of the initial value and at least one of the at least one subsequent value; generate comparison value data; and output a hypothesis regarding the patient-related characteristic based on the comparison value data.

[0252] Additionally or alternatively, the monitoring component 1500 may also be configured to record a current value of one of the at least one patient-related characteristic, where the current value is different from the initial value. That is, the monitoring component 1500 is configured to monitor changes in the value of at least one patient-related characteristic over time. For this purpose, it should be understood that the monitoring component 1500 or the system 1000 or a component of the system 1000 may record and / or determine the initial value. However, it should also be understood that this initial value may already be included in the received data. In some embodiments, the initial value may be referred to as a baseline.

[0253] Also, the system 1000 is configured to predict at least one health state of a patient based on at least one state hypothesis.

[0254] The computing component 1300 is configured to implement a dynamic model for predicting at least one health state. In one embodiment, the computing component 1300 is also configured to implement a plurality of models for proposing, generating, and / or improving agents for treating at least one health state, for improving insights, and for predicting at least one health state.

[0255] Also, the memory component 1400 is configured to store data related to at least one health state of a patient. In one embodiment, the memory component 1400 may also include a server having a plurality of computer-implemented modules. In a further embodiment, the memory component 1400 may also at least partially include the processing component 1100, the analysis component 1200, the computing component 1400, and / or the monitoring component 1500.

[0256] In one embodiment, the computing component 1300 may also include a computing device further described in FIG. 3.

[0257] The system is further configured to output a plurality of data including patient-related information such as PE. This information may include, inter alia, onset data, severity data scoring and prediction, onset dynamics analysis and interpretation, and risk assessment of patients such as mothers, fetuses, or neonates. The risk assessment may further include prediction and / or estimation of maternal, fetal, and / or neonatal complications, the type of complications, and / or the level of complications. Also, system 1000 is configured to output at least one treatment proposal, and / or treatment protocol, and / or optimization of an ongoing treatment protocol.

[0258] In one embodiment, system 1000 may also include a signal processing component (not shown) configured to process a plurality of signals supplied from one or more devices external to and / or independent of system 1000. The signal processing component may also be included in processing component 1100 and may be configured to process data received as signal data.

[0259] Figure 2 schematically shows a layered representation of an implementation form of a method according to an embodiment of the present invention. The method is a computer-implemented method. The method is implemented by system 1000. Briefly, the layered representation has three layers L1, L2, and L3. Layer L1 may also be referred to as the input layer, and L2 may be referred to as the model layer, modeling layer, processing layer, and / or analysis layer. L3 may be referred to as the output layer and / or result layer.

[0260] The input layer L1 may receive a plurality of inputs 210, 220, 230, which may include, but are not limited to, biomarkers such as parental, fetal, and / or neonatal biomarkers; a plurality of clinical parameters such as parental, fetal, and / or neonatal clinical parameters; demographics, lifestyle, and psychometric scores related to the patient and / or patient group; a plurality of environmental parameters; drug therapies such as current drug therapies administered to the patient and / or drug therapies recommended in valid or ongoing guidelines; dosing regimens, medication histories related to the patient or patient group; cardiotocography data (CTG); electroencephalogram data (EEG); electrocardiogram data (ECG), pulse, and / or oxygen measurements; deformation modalities such as sounds, and Doppler, duplex, etc.; magnetic resonance imaging (MRI); data provided by X-rays.

[0261] These inputs may be processed within the modeling L2 layer, where a plurality of computer-implemented dynamic models 310, 320, 330 may be applied to the input data to generate processed data, which may be further analyzed and interpreted to generate at least one finding, which can be represented as at least one hypothesis regarding at least one health state of the patient or patient group by a computer-implemented prediction stage. The multi-layer computer-implemented method may further utilize an output layer L3, where the interpreted data may be provided to a user such as a physician. Such output may include, among other things, PE-related predictions, evaluations 410, risk assessments 420 of PE or any other medical condition of the patient or patient group, which may include, for example, mother-related risk assessment S1, fetus-related risk assessment S2, and / or neonate-related risk assessment S3. The output layer L3 may also include one or more treatment methods 430, such as proposals for treatment protocols and / or treatment techniques, and optimization of current and / or future treatment methods.

[0262] This is particularly advantageous since the multi-layer computer implementation method provides at least one hypothesis regarding at least one health state of a patient or group of patients by system 1000, where the hypothesis is based on individual available data which may include current data and / or historical data. That is, the computer implementation method can process, analyze, and interpret data by system 1000, for example, regarding PE during pregnancy and its impact on the in-utero development of the fetus shown in FIG. 4, which shows the progression for a healthy pregnant woman 100A and a pregnant woman 100B suffering from PE.

[0263] FIG. 3 provides a schematic view of computing device 100. Computing device 100 may include a computing unit 35, a first data storage unit 30A, a second data storage unit 30B, and a third data storage unit 30C.

[0264] Computing device 100 can be a single computing device or an assembly consisting of a plurality of computing devices. Computing device 100 can be placed locally or remotely, such as in a cloud solution.

[0265] Different data can be stored on different data storage units 30. Additional data storage can be provided and / or the previously mentioned ones can be at least partially combined.

[0266] The computing unit 35 can access the first data storage unit 30A, the second data storage unit 30B, and the third data storage unit 30C through an internal communication path 160 which may include a bus connection 160.

[0267] The computing unit 30 may be a single processor or multiple processors, including but not limited to a CPU (Central Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), APU (Accelerator Processing Unit), ASIC (Application-Specific Integrated Circuit), ASIP (Application-Specific Instruction Set Processor), or FPGA (Field Programmable Gate Array). The first data storage unit 30A may be single or multiple, including but not limited to volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or parameter RAM (P-RAM).

[0268] The second data storage unit 30B may be single or multiple, including but not limited to volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or parameter RAM (P-RAM).

[0269] The third data storage unit 30C may be single or multiple, including but not limited to volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or parameter RAM (P-RAM).

[0270] The first data storage unit 30A (also referred to as the encryption key storage unit 30A), the second data storage unit 30B (also referred to as the data share storage unit 30B), and the third data storage unit 30C (also referred to as the decryption key storage unit 30C) should be understood to generally be part of the same memory. That is, only one general data storage unit 30 may be provided for each device, and storage of each encryption key (such that the portion of the data storage unit 30 that stores the encryption key becomes the encryption key storage unit 30A), storage of each data element share (such that the portion of the data storage unit 30 that stores the data element share becomes the data share storage unit 30B), and storage of each decryption key (such that the portion of the data storage unit 30 that stores the decryption key becomes the decryption key storage unit 30A) may be configured to be performed.

[0271] In some embodiments, the third data storage unit 30C can be a secure memory device 30C such as a self-encrypting memory capable of automatically encrypting all stored data, a hardware-based full disk encryption memory, and the like. The data can be decrypted from the memory component only when authentication of the party requiring access to the third data storage unit 30C is successful, where the party can be a user, a computing device, a processing unit, and the like. In some embodiments, the third data storage unit 30C can be connected only to the computing unit 35, and the computing unit 35 can be configured to never output the data received from the third data storage unit 30C. Thereby, secure storage and handling of the encryption key (i.e., the private key) stored in the third data storage unit 30C can be ensured.

[0272] In some embodiments, it may not be necessary to provide the second data storage unit 30B. Instead, it is also possible to configure the computing device 100 to receive the corresponding encrypted shares from the database 60. In some embodiments, the computing device 100 may include the second data storage unit 30B and can be configured to receive the corresponding encrypted shares from the database 60.

[0273] The computing device 100 may further include additional memory components 140, which may be single or plural and are not limited, but may be volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), flash memory, magnetoresistive RAM (MRAM), ferroelectric RAM (F-RAM), or parameter RAM (P-RAM). Also, the memory components 140 may be connected to other components of the computing device 100 (such as the computing component 35) via an internal communication path 160.

[0274] Furthermore, the computing device 100 may include an external communication component 130. The external communication component 130 can be configured to facilitate the transmission and reception of data with external devices (e.g., backup devices, recovery devices, databases). The external communication component 130 may include an antenna (e.g., Wi-Fi (registered trademark) antenna, NFC antenna, 2G / 3G / 4G / 5G antenna and the like), a USB port / plug, a LAN port / plug, a contact pad providing an electrical connection and the like. The external communication component 130 can transmit and / or receive data based on a communication protocol that can include instructions for transmitting and / or receiving data. The instructions can be stored in the memory component 140 and executed by the computing unit 35 and / or the external communication component 130. The external communication component 130 can be connected to the internal communication path 160. Thus, the data received by the external communication component 130 can be provided to the memory component 140, the computing unit 35, the first data storage unit 30A and / or the second data storage unit 30B and / or the third data storage unit 30C. Similarly, the data stored in the memory component 140, the first data storage unit 30A and / or the second data storage unit 30B and / or the third data storage unit 30C, and / or the data generated by the computing unit 35 can be provided to the external communication component 130 for transmission to an external device.

[0275] Furthermore, the computing device 100 may include an input user interface 110 that enables a user of the computing device 100 to provide at least one input (e.g., an instruction) to the computing device 100. For example, the input user interface 110 may include buttons, a keyboard, a trackpad, a mouse, a touch screen, a joystick and the like.

[0276] Furthermore, the computing device 100 may include an output user interface 120 that enables the computing device 100 to provide an indicator to the user. For example, the output user interface 110 may be an LED, a display, a speaker, and the like.

[0277] The output and input user interface 100 may also be connected to the internal components of the device 100 through the internal communication component 160.

[0278] The processor may be single or multiple, and may be, but is not limited to, a CPU, a GPU, a DSP, an APU, or an FPGA. The memory may be single or multiple, and may be, but is not limited to, volatile or non-volatile memory, such as SDRAM, DRAM, SRAM, flash memory, MRAM, F-RAM, or P-RAM. The data processing device may be capable of including data processing means such as a processor unit, a hardware accelerator, and / or a microcontroller. The data processing device 20 may have memory components, such as main memory (e.g., RAM), cache memory (e.g., SRAM), and / or secondary memory (e.g., HDD, SDD). The data processing device may be provided with a bus configured to facilitate data exchange between components of the data processing device, such as communication between the memory component and the processing component. The data processing device may be provided with a network interface card configured to connect the data processing device to a network such as the Internet. The data processing device may be capable of including, for example, the following user interfaces. (1) Output user interface, for example: A screen or monitor configured to display visual data (e.g., display the graphical user interface of the questionnaire to the user), A speaker configured to communicate audio data (e.g., play audio data to a user) (2) An input user interface, for example: A camera configured to capture visual data (e.g., capture an image and / or video of a user), A microphone configured to capture audio data (e.g., record sound from a user), A keyboard configured to allow insertion of text and / or other keyboard commands (e.g., enable a user to input text data and / or other keyboard commands by typing on the keyboard) and / or a trackpad, mouse, touch screen, joystick configured to facilitate navigation of different graphical user interfaces of a questionnaire.

[0279] The data processing device can be a processing unit configured to execute program instructions. The data processing device can be a system-on-chip comprising a processing unit, a memory component, and a bus. The data processing device can be a personal computer, laptop, pocket computer, smartphone, tablet computer. The data processing device can be either a local and / or a remote server. The data processing device can be a processing unit or a system-on-chip that can interface with a personal computer, laptop, pocket computer, smartphone, tablet computer, and / or a user interface (such as the user interfaces described above).

[0280] Note that not all reference signs are shown in all drawings. Instead, in some drawings, some reference signs are omitted for the sake of brevity and to simplify the illustration. Here, embodiments of the present invention will be described with reference to the accompanying drawings.

[0281] Reference signs and characters that appear between parentheses in the claims identify features described in the embodiments and shown in the accompanying drawings, and are provided as an aid to the reader as examples of the claimed subject matter. Such reference signs and characters shall not be construed as imposing any limitation on the claims.

[0282] The term "at least one of the first option and the second option" is intended to mean the first option, or the second option, or both the first option and the second option.

[0283] In the foregoing, the preferred embodiments have been described with reference to the accompanying drawings. However, those skilled in the art will understand that this embodiment is provided for illustrative purposes only and should never be construed as limiting the scope of application of the present invention defined by the claims.

[0284] Whenever relative terms such as "about", "substantially", or "nearly" are used in this specification, such terms shall be construed to include the exact term as well. That is, for example, "substantially straight" shall be construed to include "(completely) straight".

[0285] Whenever steps are described in this document or in the appended claims, it should be noted that the order of the steps described herein may be accidental. That is, unless otherwise specified or unclear to a person skilled in the art, the order of the steps described may be accidental. That is, for example, if a method comprises steps (A) and (B), as presented in this document, this does not necessarily mean that step (A) precedes step (B), it is also possible that step (A) is executed (at least partially) simultaneously with step (B), or that step (B) precedes step (A). Furthermore, if step (X) is said to precede another step (Z), this does not imply that there are no steps between step (X) and (Z). That is, step (X) preceding step (Z) encompasses the situation where step (X) is performed immediately before step (Z), but also encompasses the situation where one or more steps (Y1), (Y2) are performed after (X) and then step (z) follows. When terms such as "after" or "before" are used, the corresponding considerations apply.

Claims

1. A method for predicting a patient's health status, the method comprising: receiving at least one patient-related kinetic characteristic data; receiving at least one patient-related covariate; processing the at least one patient-related kinetic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set; generating at least one health status hypothesis based on the patient-related processed data set; and predicting at least one health status based on the at least one health status hypothesis A method comprising the steps of:

2. The step of predicting the at least one health status is based on a computer-implemented kinetic model, the at least one health status hypothesis has a correlation with at least one medical condition of the patient, and the method further comprises predicting the kinetic behavior of the at least one health status of the patient. The method according to claim 1.

3. The at least one patient-related covariate is: at least one biomarker, wherein the at least one biomarker is related to at least one medical condition, and wherein the at least one biomarker comprises at least one of the following: soluble Fms-like tyrosine kinase (sFlt-1); placental growth factor (PlGF); neurofilament (NfL); C-terminal portion of arginine vasopressin (copeptin); albumin; liver transaminase; urea; hemoglobin; platelets; creatinine; albuminuria; proteinuria; estimated glomerular filtration rate (eGFR); creatinine clearance (CrCl); at least one additional renal function measurement; placental biomarkers such as placental RNA, placental protein; endothelial / cardiovascular biomarkers such as endothelial RNA, endothelial protein; or any combination thereof; At least one maternal-related covariate comprising at least one of the following: age; weight; height; body mass index (BMI); pregnancy history; delivery history; number of fetuses in the current pregnancy; race; body temperature; heart rate; heart rate variability; respiratory rate; premature rupture of membranes; white blood cell count; history of preeclampsia (family and maternal), gestational diabetes, obesity, cardiovascular / renal / liver / thyroid diseases, autoimmune diseases, anemia, antiphospholipid syndrome, sexually transmitted infections, coexisting conditions such as headache; smoking habits before and / or during pregnancy; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); uteroplacental perfusion parameters; Doppler measurements of pulsatility indices of the umbilical artery, middle cerebral artery, cerebroplacental ratio, uterine artery, fetal descending aorta, ductus venosus, umbilical vein, inferior vena cava, uterine artery; soft tissue parameters such as upper arm partial volume and thigh partial volume; and measurements of at least one of the at least one biomarker; At least one fetal-related covariate comprising at least one of the following: gender; fetal weight during pregnancy; biometric parameters of the fetus such as femur length, abdominal circumference, head circumference, mid-thigh circumference, and transverse diameter; proportion of small-for-gestational-age; gestational age; heart rate; heart rate variability; respiratory rate; uteroplacental perfusion parameters; and measurements of at least one of the at least one biomarker; At least one neonatal-related covariate comprising at least one of the following: gender; birth weight; weight; height; gestational age at birth; postnatal age; body temperature; heart rate, heart rate variability; respiratory rate; breast milk; duration of exclusive breast milk; pH value; respiratory support; oxygen requirement; blood oxygen saturation (SpO2); blood pressure (systolic / diastolic); Apgar score; at least additional neonatal biometric parameters; and measurements of at least one of the at least one biomarker; and At least one environmental covariate comprising at least one of the following: country of residence, country of birth, time of birth, humidity conditions at birth, and ambient temperature at birth The method according to claim 1, having at least one of the above.

4. The method comprises generating at least one threshold, wherein the at least one threshold represents an indicator of at least one potential medical condition; Outputting at least one potential medical condition, where the at least one potential medical condition includes at least one of the following: seizure; respiratory; cardiovascular; blood dysfunction; endocrine; renal; hepatic; uteroplacental dysfunction; fetal growth retardation; preterm birth; placental abruption; hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome; and eclampsia; Determining a minimum threshold for the at least one threshold, and Determining a maximum threshold for the at least one threshold, Determining a baseline for the at least one patient-related data, and Determining at least one intermediate threshold, where the at least one intermediate threshold includes at least one value between the minimum threshold and the maximum threshold, The method according to claim 1, comprising.

5. The method is Correlating at least a respective range of each of the at least one intermediate threshold with the at least one medical condition, Generating an interpreted data set based on the correlating step, and Outputting an automated report indicating the at least one potential medical condition The method according to claim 4, comprising.

6. The method is Determining at least one medical condition change indicator, Monitoring a change in the at least one medical condition change indicator, Generating at least one medical condition change indicator trend, and Predicting the progression of at least one of the at least one medical condition, where the prediction is based on the at least one medical condition change indicator trend, Comprising, where the method comprises monitoring a change in at least one value of at least one of the at least one patient-related characteristic, and the method is Recording an initial value of the at least one patient-related characteristic, Recording at least one subsequent value of the at least one patient-related characteristic, Comparing the initial value with at least one of the at least one subsequent value, Generating comparison value data, and Outputting a patient-related characteristic hypothesis based on the comparison value data Comprising, where the step of recording at least one subsequent value has a step of recording a current value of one of the at least one patient-related characteristic, and the current value is different from the initial value, the method according to claim 1.

7. The method is Supplying data to the at least one server, Training a computer-implemented dynamic model based on data supplied to the at least one server, and Generating an adjustment function based on the data for training, wherein the adjustment function is suitable for adjusting any stage of the method according to claim 1; Triggering at least one action proposal based on the at least one health state hypothesis; Displaying the at least one action proposal to the user, and Prompting the user to input at least one of acceptance of at least one of the at least one action proposal and rejection of at least one of the at least one action proposal; The method according to claim 1, comprising.

8. The method is Determining at least one drug based on the at least one health state, wherein the at least one drug is Preventing the occurrence of the at least one medical condition and / or the at least one health state, and Treating the at least one medical condition and / or the at least one health state Suitable for at least one of; Generating at least one dosing regimen for the at least one drug; Optimizing the at least one dosing regimen, wherein the at least one dosing regimen has at least one of the at least one drug, the route of administration of the at least one drug, at least one dosing regimen, at least one drug administration period, and at least one drug administration frequency, and wherein the step of optimizing the at least one dosing regimen is based on the at least one health state hypothesis; The method according to claim 1, comprising.

9. A system for predicting a patient's health state, the system comprising Receiving at least one patient-related dynamic characteristic data, Receiving at least one patient-related covariate, At least one processing component configured to process the at least one patient-related dynamic characteristic data and the at least one patient-related covariate data to generate a patient-related processed data set; Analyzing the patient-related processed data set, At least one analysis component configured to generate at least one health state hypothesis based on the patient-related processed data set Comprising And ​ Here, the system is configured to predict at least one health state based on the at least one health state hypothesis and to implement the method according to any one of claims 1 to 8, a system.

10. The system is at least one memory component configured to store data related to the at least one health state of the patient; at least one computing component configured to implement a dynamic model for predicting the at least one health state, where the at least one health state hypothesis has a correlation with at least one medical condition of the patient, comprising The system according to claim 9, configured to predict the dynamic behavior of the at least one health state of the patient.

11. The system is configured to generate at least one threshold value, where the at least one threshold value represents an indicator of at least one potential medical condition; output at least one potential medical condition and is configured as such, where the at least one potential medical condition includes at least one of seizure; respiratory; cardiovascular; blood dysfunction; endocrine; kidney; liver; uteroplacental dysfunction; fetal growth retardation; unplanned preterm birth; placental abruption; hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome; and eclampsia, the system according to claim 9.

12. The at least one analysis component is determining a minimum threshold value for the at least one threshold value, determining a maximum threshold value for the at least one threshold value, correlating each at least range of the at least one intermediate threshold value with at least one medical condition, generating an interpreted data set based on the correlating step, outputting an automatic report indicating at least one potential medical condition, determining at least one medical condition change indicator, monitoring changes in the at least one medical condition change indicator, generating at least one medical condition change indicator trend, predicting the progression of at least one of the at least one medical condition based on the at least one medical condition change indicator trend and is configured as such, the system according to claim 9.

13. The system comprises at least one monitoring component configured to monitor at least one value change of at least one of the at least one patient-related characteristic, and the at least one monitoring component further Record an initial value of the at least one patient-related characteristic, Record at least one subsequent value of the at least one patient-related characteristic, Compare the initial value with at least one of the at least one subsequent value, Generate comparison value data, Output a hypothesis of a patient-related characteristic based on the comparison value data, Record a current value of one of the at least one patient-related characteristic, where the current value is different from the initial value, The system according to claim 9, configured as such.

14. The system, Supply data to the at least one server, Train a computer-implemented dynamic model based on the data supplied to the at least one server, Generate an adjustment function based on the data for training, where the adjustment function is suitable for adjusting any configuration of the system described in claim 9 above; Trigger at least one action proposal based on the at least one health state hypothesis The system according to claim 9, configured to display the at least one action proposal to the user and prompt the user to input at least one of acceptance of at least one of the at least one action proposal and rejection of at least one of the at least one action proposal.

15. The system, Capture at least one image data of the patient, and Receive at least one image data of the patient Comprising at least one imaging component configured to perform at least one of these, where The at least one image data includes data related to at least one medical condition and / or at least one potential medical condition of the patient. The system according to claim 9.

16. A method of treatment for treating a medical condition of a patient, the treatment comprising generating a treatment protocol having at least one therapeutic agent and a treatment regimen, where the treatment regimen is based on at least one health state hypothesis, and where the at least one health state hypothesis is provided by the method according to any one of claims 1 to 8.

17. A diagnostic method for diagnosing a patient's medical condition, the diagnosis comprising the step of generating at least one diagnostic finding having at least one medical condition of the patient, wherein the at least one diagnostic finding is based on at least one health state hypothesis, and wherein the at least one health state hypothesis is provided by the method according to any one of claims 1 to 8.

Images