A system for predicting the severity of covid-19 disease progression in individuals

Abstract

A prediction system for predicting the severity of the course of the disease when individuals are infected with COVID-19 disease, comprising a processor unit, an input unit for inputting data to said processor unit, an output unit for outputting data to said processor unit and a memory unit associated with the processor unit for reading / writing data. The processor unit is configured to access a mathematical model in said memory unit which, when taking as input the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands, classifies the disease into one of at least two disease course categories; to receive the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands as input from an input unit, and to output the classification result produced by the said mathematical model in response to the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands as output from an output unit.

Description

[0001] A SYSTEM FOR PREDICTING THE SEVERITY OF COVID-19 DISEASE PROGRESSION IN INDIVIDUALS

[0002] TECHNICAL FIELD

[0003] The invention is a computer-based prediction system that provides findings that the suspected patient may have asymptomatic, mild-moderate, or severe disease before contracting the COVID-19 disease, thus providing a risk scoring function for the course of the disease. The prediction system of the invention will be carried out by in-vitro methods, and no treatment or diagnostic method is applied to the human body.

[0004] BACKGROUND

[0005] Despite intensive scientific research and great successes in vaccines, SARS-CoV-2 (hereinafter referred to as COVID-19) is still a major cause of morbidity and mortality worldwide. However, there is no study that has concluded that some individuals have asymptomatic disease while others develop severe symptomatic disease (Vetter et al. 2020). Vetter et al. mentioned that viral load, early intervention, gender, blood groups, comorbidities and hereditary genotypic variants may be important variables that determine a symptomatic outcome.

[0006] In the relevant technical field, there are studies on the preparation of tests and kits related to what symptomatic features COVID-19 disease will show in sick people. Vetter et al. in 2020, Wu et al. in 2021 , Van d. Made et al. in 2020, they obtained results related to testing and risk scoring with certain accuracy shares by considering various variables.

[0007] In 2020, Ovsyannikova et al. and Casanova et al. examined various genotype variants that were already found to be susceptible to viral infections with separate studies.

[0008] In the studies conducted by GWAS Group and its team in 2020, a gene cluster on chromosome-3 was identified as a genetic predisposition locus for COVID-19-related respiratory failure. Chromosome-3 contains a risk locus containing CCR9 , CXCR6 , XCR1 , CCR1 and CCR2 immune genes. The COVID Human Genetic Impact Consortium reported an overall estimated 3.5% incidence of TLR3, IRF7, and IFNAR1 variants in patients with severe disease suggesting that the innate immune response is defective. Made et al. in another study they conducted in 2020 identified rare assumed loss-of-function variants of X-chromosomal TLR7 associated with impaired type I and II IFN responses.

[0009] All these findings strongly point to the causal link between hereditary variants and the severity of COVID-19 infection.

[0010] There is evidence suggesting that “Killer Immunoglobulin-like Receptors” (KIR) genotypes have a role in antiviral immunity, as well as studies on the association of many other diseases with KIR (Khakoo and Carrington 2006 ). A more recent approach based on the analysis of the genomic organization of individual genes allows the identification of KIR haplotypes (Vendelbosch et al. 2015; Roe et al. 2017 ; Cisneros et al. 2020). Haplotypic analysis of KIR diversity allows for a comprehensive assessment of the concurrent effects of both inhibitory and activating KIR genotypes on the immune response.

[0011] Although there are groups reporting associations between viral infections and KIR gene and / or ligand frequencies, as well as A / B KIR haplotypes, there are no studies in the art focusing on telomeric / centromeric KIR motifs. Lu et al. in 2008, observed lower frequencies of haplotype A and higher frequencies of haplotype B in HBV-infected patients compared to healthy controls. They also found an association with certain KIR genotypes in Caucasian populations. However, they did not include telomeric / centromeric motifs in their study.

[0012] In the context of RNA virus infections, Influenza A (H1 N1 ), HIV, HCV, Ebola, Dengue virus (DENV) and Chikungunya virus (CHIKV) have been associated with specific KIR genes / ligands. KIR / HLA compound genotypes with an activation profile (presence of activating KIR or lack of inhibitory KIRs in the presence of ligands of the same origin) are associated with resistance to HCV (Khakoo et al. 2004) and HIV infections (Jennes et al.2006). On the other hand, Shan et al. found in their study in 2018 that chronic HCV infection was associated with KIR2DL3 / 2DL3, 2DL3 / 2DL3+HLA-C1 or C1 C1 genotypes. Similarly, the coexistence of KIR2DL3, which is included in the centromeric AA (cAA) motif and the same-origin ligand C1 or C2, has been found to be linked to rapid progression / susceptibility to HIV and Chikungunya virus (CHIKV) infection, respectively (Mori et al. 2019). Ahlenstiel et al. in 2008 and La et al. in their study in 2014, showed a possible association with NK cell dysfunction in patients with overactive immune responses to H1 N1 / 09 and leading to severe disease. Individuals with KIR3DL1 / S1 lacking Bw4 or KIR2DL1 lacking HLA-C2 ligands were detected at a high rate in intensive care patients during the 2009 influenza outbreak, individuals with KIR2DL2 / L3 and its corresponding ligand HLA-C1 . Aranda-Romo et al. in their study in 2012, found a relationship between severe H1 N1 infections and KIR2DL5, KIR2DS5 and KIR3DS1 genes. Although the studies did not include KIR haplotype analyses, along with these data telomeric AB1 (tAB1) without its ligand or centromeric AB1 (cAB1 ) with its ligand found to be associated with the need for intensive care during H1 N1 infection.

[0013] In the relevant technical field, there are only two published studies on the frequency and ligands of KIR receptors in relation to the severity of COVID-19. The Beijing Group, the first study of the National Research Project for SARS, reported that CD158b (KIR2DL3, a member of the centromeric A haplotype) expressed in NK cells was less frequent in patients with severe disease compared to mild cases, in line with findings previously reported for influenza infection (Ahlenstiel). The second study, Maucourant et al. is about a comprehensive phenotypic and functional analysis of KIR genotypes and KIR receptor expression in 2020 and the relationship between disease severity and KIR genotype / haplotype.

[0014] David J. Langton et al reported in the journal of "HLA Immune Response Genetics" on 25.04.2021 that the HLA-DRB1 *04:01 gene was more common and the DQA1 *01 :01- DQB1 *05:01-DRBr01 :01 haplotype was less common in asymptomatic cases. David J. Langton and his study team filed a patent application for these findings.

[0015] In studies to date, a risk analysis is carried out only on the basis of biochemical values and clinical parameters measured after the onset of the disease. There is no pre-infection risk analysis study that includes genetic factors affecting the individual immune system. However, it is clear that knowing such a risk in advance will be beneficial in matters such as who should be prioritized in vaccination, and who should take higher level measures to protect against the virus(es). In addition, an answer is sought to the question of why some individuals have had the disease without symptoms or very mild since the beginning of the pandemic, while others have filled the intensive care units with severe infection. It has become a must to carry out research and development activities in the relevant technical field to ensure all these. BRIEF DESCRIPTION OF THE INVENTION

[0016] In the current art, the risk analysis scoring of infectious diseases specific to COVID-19 is performed on the biochemical values and clinical parameters measured after the onset of the disease. A pre-infection risk analysis scoring study involving genetic factors affecting the individual immune system is not among the available methods. In order to provide a technical solution and advantage to the technical field related to the present invention, a risk probability calculation system that can be applied before contracting the infection is presented. This system is a mathematical probability calculation system that the person can have the disease with asymptomatic, mild-moderate symptoms or severe symptoms in case of infection using age, number of comorbidities, blood group and KIR genotype information, even if the individual is not yet infected. The individual KIR genotype can be determined by those who are experts in their field in immunogenetics laboratories. Knowing the age, number of comorbidities and blood group after determining the KIR genotype is sufficient to calculate the probability of how severe the person may have the symptoms of the disease if they have COVID-19 disease. While the results obtained provide information about which patient group people should be included in, they can also reveal the necessity of prioritization in vaccination. The invention is implemented on a computer-based basis, and a website has been created for the ease of implementation of the invention in greater detail. By filling in the aforementioned parameters in the relevant fields on this website, the individual's risk score for COVID-19 infection severity can be calculated automatically.

[0017] The invention is a prediction system for predicting the severity of the course of the disease when individuals are infected with COVID-19 disease, comprising a processor unit, an input unit that provides data entry to said processor unit, an output unit that enables said processor unit to output data, and a memory unit to which the processor unit is associated to read / write data. Accordingly, when the processor unit takes the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands as input, it will access a mathematical model in said memory unit that classifies them in at least one of two disease course categories; it is configured to take the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands from the input unit as input, and to output the classification result that said mathematical model produces in response to the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands as output from an output unit. Thus, when people have COVID-19, the severity of the course of the disease can be estimated and it can be ensured that people take precautions or check their hospital capacities accordingly. DETAILED DESCRIPTION OF THE INVENTION

[0018] In this detailed description, the subject matter of the invention is a risk test and prediction system for its implementation, which provides a risk scoring function for the course of the disease and thus a prediction system that provides evidence that the affected person may be asymptomatic, mild-moderate, or severe before the symptoms of COVID-19 are detected and is only described with examples that do not have any limiting effect for a better understanding of the subject matter.

[0019] The prediction system includes a processor, a memory unit associated with the processor, an input unit that provides data input to the processor, and an output unit where the processor outputs data. The prediction system may suitably be a general-purpose computer. The memory unit may include software consisting of command lines, including the process steps to be performed by the processor to ensure the realization of the test system of the invention. In more detail, the memory unit may include the mathematical model used in prediction. The mathematical model and its extraction are described in more detail later in this detailed description.

[0020] Taking as input parameter values for age, number of comorbidities, blood type, tAB1 and / or ligands, and tAA and / or ligands, the model herein provides a classification of the course of a person's viral infection. This classification is such that it is included in at least two course categories.

[0021] These categories can be asymptomatic, moderate, severe, etc.

[0022] The processor unit receives parameter values related to age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands of a person from the input unit as input, applies these parameter values to the mathematical module and outputs the course category obtained by the mathematical module as data output from the output unit. The output unit may be a user interface, display, etc.

[0023] The processor unit may be connected to a hospital host computer or server(s) in a possible embodiment of the invention. The patient may access the capacity databases in a possible embodiment of the invention. Hospital capacity databases refer to the current capacity of hospitals to intervene / care for course categories. This capacity may be the capacity to care for the patients with moderate illness, the capacity to care for the severe patients, etc. The capacity to care for severe patients can be, for example, the number of appropriate intensive care rooms available and the number of staff associated with them. When the processor unit detects that the number of categories obtained as a result of consecutive estimates does not match the capacity of the hospitals, it can send a warning signal to the hospitals. Thus, it is determined that hospitals can take a high number of COVID-19 patients who will have a severe course in the future, and it can be ensured that they take precautions accordingly.

[0024] The mathematical model is as follows in one embodiment of the invention:

[0025] Said model is a model in which the age data is shown with the coefficient of 1 for below 35 years, 2 for 35-50 years, 3 for 51-64 years, 4 for over 65 years old; blood group is shown with the coefficients 1 for group-A and 0 for the other; the number of comorbidities is shown with the coefficients 1 , 2, 3 or 4; genotype selected among the parameters of tAB1 with ligand and tAA with ligand is applied as 1 to the exponential function:

[0026] Exp(-3.52 + 1.56 Age Group - 2.74 Blood Group (Group-A vs. other) + 1.26 Number of comorbidities - 2.46 tAB1 with ligand + 3.17 tAA with ligand).

[0027] The following are details of obtaining the mathematical model.

[0028] Steps for obtaining information

[0029] First of all, age information is obtained for individuals. As it is known, increasing age causes a decrease in the incidence of symptoms and resistance to disease, and this has been proven by many studies (Geng et al. 2021). Accordingly, scoring was performed according to the meaning expression for the age groups determined in the risk test. Age grouping in the invention was made as below 35 years of age, 35-50 years of age, 51-64 years of age and 65 years of age or older.

[0030] The next step is to determine the blood group. Numerous studies have shown that different blood groups are decisive in terms of contracting the infection or affecting the severity of the disease after the infection, but there is no consistency between the results. In our study, it was found that blood type A was protective against the need for intensive care in COVID-19 disease. A study in Nature Communications reported that although blood type A was found to be protective against the need for intubation, it was also associated with an increased risk of contracting COVID-19 (Zietz et al. 2020).

[0031] As the next step, the number of comorbidities should be determined. As known in the art, "comorbidity" means one or more diseases or medical disorders in patients. Studies show that different concomitant diseases of individuals have significant effects on the course of COVID-19 infection. Diseases evaluated within the scope of comorbidity in the invention are determined as diabetes, obesity, hypertension, chronic obstructive pulmonary disease (COPD), coronary artery disease (CAD), chronic renal failure (CRF) and cancer. In the risk probability determination system, the number of these diseases the individual has should be entered.

[0032] Unlike the above information, which is easy to reach for each individual, the critical parameter for the invention is the KIR genotype. KIR genes stand for Killer Immunoglobulin- like Receptor genes in the art and are from the protein family that determine the function of natural killer cells of the immune system and present on the surface of these cells. As is known in the art, KIR genes are expressed in polymorphic regions that have been conserved for generations and have different types in different individuals. Existing studies in the literature point to strong relationships between different KIR genotypes, especially with cancer and infectious diseases. Our team contributed to this literature with two publications on the relationship between colon cancer and stem cell transplant success and KIR genotypes (Beksag et al. 2015; §ahin et al. 2017). With the pressure of balancing selection in humans, KIR haplotypes with inhibitory and activating effects have evolved. HLA ligands to which KIR receptors bind are similarly located in polymorphic regions and differ between individuals. KIR genes are grouped with different classifications. The genes, which are mainly grouped as haplotypes A and B and telomeric / centromeric haplotypes, have geographically different distribution characteristics. In the invention, it was determined that the coexistence of the tAA and tAB1 genotypes classified in the telomeric region and the corresponding ligands (Bw4+ABw4 and C2+Bw4, respectively) was one of the factors affecting the severity of COVID-19. Accordingly, the person's KIR genotype information should be entered into the risk probability calculation system as "tAA with ligand (Bw4+ABw4+)", "tAB1 with ligand (C2+Bw4+)" or "other".

[0033] A small amount of blood or saliva samples may be sufficient to determine the KIR genotypes of individuals. DNA isolation from the samples taken and then KIR genotyping and haplotype analysis can be performed by experts in this field in immunogenetic laboratories. DNA isolation from these samples is performed by methods known in the art. Although it is not within the scope of protection of the invention, DNA isolation in our unit can be performed in Easy 1 Advanced XL (Qiagen) device with DNA blood kit (Qiagen) or QIAmp DNA Investigator Kit (Qiagen). The DNA quality and quantities of the samples can be measured spectrophotometrically in the Nanodrop 1000 (Thermo Scientific, Massachusetts, USA) device. KIR genotyping and KIR-ligand genotyping processes are started from DNAs of sufficient quantity and quality. Although the invention is not within the scope of protection, KIR / KIR ligand genotyping processes in our unit are carried out using Qiagen brand SSP KIR Genotyping (Oleup, Stockholm, Sweden; Kat No: 104.101 -12U) and KIR HLA Ligand PCR SSP (Olerup, Stockholm, Sweden; SSP 04.201 -12U) commercial kits following DNA isolation. Different labs can also apply this method using their own probes with different brand kits. With the aforementioned commercial KIR / ligand genotyping kits, 14 different KIR genes (KIR2DL1 , 2DL2, 2DL3, 2DL4, 2DL5A / B, 2DS1 , 2DS2, 2DS3, 2DS4, 2DS5, 3DL1 , 3DL2, 3DL3, 3DS1 ) and 4 different KIR ligand genes (C1 , C2, Bw4, A-Bw4) are determined.

[0034] The result obtained after determining that the person has tAA (Bw4+ABw4+), tAB1 (C2+Bw4+) or any other genotype different from them with the KIR haplotype classification is ready to be entered into the probability calculation system.

[0035] After entering these four parameters (age, number of comorbidities, blood type, KIR genotype), the calculation of the probability of risk takes place with the following formula: Exp(-3.52 + 1.56 Age Group - 2.74 Blood Group (Group-A vs. other) + 1.26 Number of comorbidities - 2.46 with tAB1 ligand + 3.17 with tAA ligand)

[0036] The risk prediction system of the invention is implemented as an in-vitro method. No treatment and diagnostic method is performed on the patient. As is known, the expression "in-vitro" means carried out in laboratory or artificial conditions.

[0037] The KIR genotype, which is critical for the invention, is abbreviation of Killer Immunoglobulin- like Receptor in the art and is from the family of natural killer cells for the immune system. As is known in the art, the roles of KIR molecules in the reproductive and defense system have remained unchanged for generations. According to the studies in the literature, its relationship with some diseases, especially infectious diseases, is strong and it is a genetic indicator for these diseases. Group A and B KIR haplotypes, which have inhibitory and activating effects with balancing selection pressure in humans, have evolved, respectively. Polymorphism is the evolutionary adaptation of genes in the immune system, which are encountered with changing environmental conditions.

[0038] Although the present invention is related to determining the risk score for COVID-19 infection in sick people, it may also evolve into determining the risk score of similar infectious diseases in the future. The present invention combines the clinical information obtained from the patients who want to provide the relevant technical field and the KIR / KIR ligand genotyping information to obtain significant results for said diseases before infection. According to all that is mentioned herein, in the present invention, there are mainly three process steps. These three process steps are; obtaining clinical information such as age, comorbidity, blood group from the person, as the second process step, the person's KIR / KIR ligand genotyping, and as the last process step, applying this information to the mathematical formula given above and performing the probability calculation. The data entry and probability result through an example on the published website are presented below.

[0039] Example

[0040] A 53-year-old individual with blood type A, two different comorbidities (e.g., diabetes and hypertension), and tAB1 (C2+Bw4+) genotype was administered on a web address to demonstrate the probability calculation on a sample case. After entering the personal information in the specified areas, the probability of needing intensive care of this individual in case of getting infected was calculated through the mathematical formula as 17.8873% .

[0041] The fixed multipliers and coefficients in this mathematical formula within the scope of the invention were determined as a result of our study conducted at the emergence of the invention. The method performed in this study, the patient groups included in the study and the statistical analysis steps are presented under subheadings below.

[0042] Patient Selection Criteria and Obtaining Consents and Completing Questionnaires

[0043] A total of 132 patients who were diagnosed with COVID-19 in Ankara University Medical Faculty Hospital, including 56 asymptomatic patients, 51 outpatients with a positive PCR test without symptoms (asymptomatic) who were followed up in the hospital or at home with mild- moderate symptoms, but with no detected pneumonia, and 25 intensive care patients with pneumonia were included in the study conducted to obtain the reference values used in the creation of the mathematical model.

[0044] Patient groups composed of female or male individuals between the ages of 18-89. a) Asymptomatic patients diagnosed with COVID-19;

[0045] - A history of being tested and testing positive as a result of a relative's diagnosis

[0046] No or mild signs of COVID-19 such as cough, fever, weakness, vomiting, abdominal pain

[0047] • Frequent sense of taste loss in asymptomatic patients b) Patients diagnosed with COVID-19 and followed up at home and in the clinic; Carrying COVID-19 symptoms such as cough, fever, weakness, vomiting, abdominal pain

[0048] - No need for respiratory support

[0049] - No pneumonia developed c) Patients diagnosed with COVID-19 and hospitalized in the intensive care unit and progressing with a poor prognosis

[0050] - Needing intensive care support

[0051] Mostly with lung involvement.

[0052] The categories mentioned according to the severity of COVID-19 symptoms were created based on the standards determined by the World Health Organization. Gender, age, diabetes, hypertension, COPD, CAD, CRF, cancer, lung involvement, C-reactive protein (mg / gl), d-dimer (mg / mL), hospitalization status, and length of hospital stay, which are among the parameters prepared within the scope of the present invention and included in the health evaluation form, were compared for all patients, and evaluated between the groups with statistical methods.

[0053] Reference KIR / KIR ligand genotyping

[0054] DNA isolation was performed in the EZ1 Advanced XL (Qiagen) device using the DNA Blood kit (Qiagen) or the QIAamp DNA Investigator Kit (Qiagen) by taking 200 pl of blood samples collected from three different groups. The isolated DNA quality and quantity were measured spectrophotometrically in the Nanodrop 1000 (Thermo Scientific) device.

[0055] Using DNA samples found to be of appropriate quantity and quality, PCR was performed with KIR-specific primers for KIR genotyping and KIR ligand genotyping using CE-certified Olerup SSP KIR Genotyping (Oleup) and KIR HLA Ligand PCR SSP (Olerup) kits, respectively. The obtained PCR amplicons were imaged under UV in terms of amplicon bands specific to the KIR genes after agarose gel electrophoresis, and KIR genotype and ligand analysis were performed using the marker bands as reference. Within the scope of the research, a total of 14 different KIR genotypes (KIR2DL1 , 2DL2, 2DL3, 2DL4, 2DL5A / B, 2DS1 , 2DS2, 2DS3, 2DS4, 2DS5, 3DL1 , 3DL2, 3DL3, 3DS1) and 4 different KIR ligand genotypes (C1 , C2, Bw4, A-Bw4) were evaluated. Negative and positive controls were used in each experiment.

[0056] After determining the telomeric and centromeric haplotypes of the cases at each clinical disease stage, the differences between them were investigated by considering the possibility of the presence or absence of relevant ligands corresponding to the determined KIR genes. The distribution of KIR genotypes, haplotypes (with or without ligands), blood groups, gender, and co-morbidity frequencies among patient groups was evaluated using a chi- square or, where appropriate, two-tailed Fisher's exact test. Candidate variables for multivariate binary logistic regression analysis were selected based on univariate test results with a significance limit of 0.10. Candidate variables for the model were selected as age, blood type A, number of comorbidities, tAB1 (with C2+Bw4 ligands), tAA (with Bw4+ABw4 ligands), and cAB1 (with C1 +C2 ligands), and the analysis was performed with the backward elimination method. The present inventors have identified the protective effect of tAB1 and cAA for infections; however, after the analysis, only the tAB1 motif was found to have significant value. It was determined that only tAA reached the statistical significance level from the cAB1 and tAA genotypes, which were found to be related to the need for intensive care. In line with the results of the analysis, the risk scoring formula was calculated. In order to evaluate the ability of the risk prediction model between intensive care patients and asymptomatic patients to distinguish patient groups, the performances of our prediction models were evaluated by creating two ROC curves, one with KIR genotypes and the other with KIR genotypes. All statistical analyses were performed using IBM SPSS Statistics software.

[0057] Accordingly, it has been determined that age group, blood group A, number of comorbidities, C2+Bw4 ligands, tAB1 and Bw4+ABw4 ligands and tAA are important independent variables in predicting serious disease. The multivariate logistic model was used, and the number of comorbidities was coded as 0-4; age was coded as 1 under 35, age was coded as 2 between 35-50, age was coded as 3 between 51-64, and age was coded as 4 over 65. Accordingly, exp(-3.52 + 1.56 age group - 2.74 blood group A + 1.26 comorbidity number - 2.46 tAB1 w ligand + 3.17 tAA w ligand) equation was obtained with the multivariate logistic model obtained. In the results obtained with the equation obtained here, it was possible to determine the risk score with an accuracy of 92.86% for asymptomatic cases; in general, it was possible to determine the risk score with an accuracy of 87.65%.

[0058] Following the determination of the KIR genotypes of the person by experts in an immunogenetic laboratory, by entering the relevant parameters into the application on the website, information can be obtained about the possibility of the need for intensive care in case the person is infected.

[0059] In this way, with the information obtained before the disease, it will be possible to have preliminary information about the severity of the disease before the infection is transmitted. The present inventors provide a pre-disease risk score test that can be obtained by combining clinical information and genotype information in the relevant technical field. Another technical solution to be provided by this is to provide foresight about who can be subjected to priority vaccination procedures for said disease. In addition, although this system is only related to COVID-19 in its current form, it is an invention that can enable the risk score of similar infectious diseases to be determined in the future.

[0060] The scope of protection of the invention is specified in the attached claims and cannot be limited to those explained for sampling purposes in this detailed description. It is evident that a person skilled in the art may exhibit similar embodiments in light of above-mentioned facts without drifting apart from the main theme of the invention.

Claims

CLAIMS A prediction system for predicting the severity of the course of the disease when individuals are infected with COVID-19 disease, comprising a processor unit, an input unit for inputting data to said processor unit, an output unit for outputting data to said processor unit and a memory unit associated with the processor unit for reading / writing data; characterized in that the processor unit is configured to access a mathematical model in said memory unit which, when taking as input the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands, classifies the disease into one of at least two disease course categories; to receive the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands as input from an input unit, and to output the classification result produced by the said mathematical model in response to the parameter values of age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands as output from an output unit. A prediction system according to Claim 1 , characterized in that said mathematical model has been created as a result of multivariate binary logistic regression analysis using the data from persons with known disease course category and pre-registered age, number of comorbidities, blood group, tAB1 and / or ligands and tAA and / or ligands parameters. A prediction system according to Claim 1 , characterized in that said model is a model applied to the exponential function, such that the age data is represented with a coefficient of 1 for under 35 years, 2 for 35-50 years, 3 for 51-64 years and 4 for over 65 years; blood group is represented by a coefficient of 1 for group-A and 0 for the others; number of comorbidities is represented by a coefficient of 1 , 2, 3 or 4; genotype selected among tAB1 with ligand and tAA with ligand parameters with a coefficient of 1 . Exp(-3.52 + 1.56 Age Group - 2.74 Blood Group (Group-A vs. other) + 1.26 Number of comorbidities - 2.46 with tAB1 ligand + 3.17 with tAA ligand). A prediction system according to Claim 1 , characterized in that the processor unit is configured to access a hospital capacity database and query the hospital capacity suitable for the determined course category.

5. A prediction system according to Claim 4, characterized in that it is configured to generate a warning signal that there will be a number of patients in the first category exceeding the capacity of the hospital in the future if the result of a first course category in a predetermined number of criteria is obtained within a predetermined period of time after the consecutive course category of the persons is determined.

6. A prediction system according to Claim 1 , characterized in that the processor unit is configured to generate a warning signal in case the capacity of the hospital does not correspond to the determined categories and their numbers, after consecutive categorization for different persons.

7. A prediction system according to Claim 1 , characterized in that said output unit is a display.

8. A prediction system according to Claim 1 , characterized in that said input-output unit is provided on a web page.