METHOD AND SYSTEM FOR THE EVALUATION OF THE RISK OF AORTIC RUPTURE OR DISSECTION IN AN INDIVIDUAL WITH AN ASCENDING THORACIC AORTIC ANEURYSM

Abstract

A method for calculating the risk of aortic rupture or dissection of an individual with an ascending thoracic aortic aneurysm, ATAA, is disclosed. The method includes the steps of obtaining a first data set linked to the clinical and/or demographic characteristics of the individual, obtaining a second data set linked to the biochemical characteristics of a biological sample of the individual, obtaining a third data set linked to the morphological and functional characteristics of the aorta and processing the third data set to obtain a fourth data set by computational modelling, integrating the first data set, the second data set, the third data set and the fourth data set in a predictive model to obtain a risk index (i) of aortic rupture or dissection, wherein the second data set includes expression values of at least one biomarker.

Claims

1. A method for calculating a risk index of aortic rupture or dissection of an individual with ascending thoracic aortic aneurysm, ATAA, the method comprising the steps of: obtaining a first data set linked to the clinical and/or demographic characteristics of the individual; obtaining a second data set linked to the biochemical characteristics of a biological sample of the individual; obtaining a third data set linked to the morphological and functional characteristics of the aorta and processing said third data set to obtain a fourth data set by computational modelling; and integrating the first data set, the second data set, the third data set and the fourth data set in a predictive model to obtain the risk index (i) of aortic rupture or dissection; wherein the second data set comprises expression values of at least one non-coding RNA biomarker chosen from the group consisting of: miR-16, miR-9 miR-101, miR-143, miR-19, miR-21, miR-29, and miR-423-5p.

2. The method according to claim 1, wherein the biomarker is chosen from the group consisting of: a metalloproteinase of the extracellular matrix, MMP, and a tissue inhibitor, TIMP.

3. The method according to claim 1, wherein the second data set further comprises expression values of at least one biomarker of non-coding RNA chosen from the group consisting of: miR-133a, miR-155, miR-320a, miR-34a, and miR-34a (MI0000268).

4. The method according to claim 2, wherein the metalloproteinase of the extracellular matrix is MMP-9 and the tissue inhibitor is TIMP-1.

5. The method according to claim 1, wherein the biomarker is chosen from the group consisting of: C-reactive protein, creatine kinase, Nt-proBNP, troponin, advanced glycation end product, AGE, and corresponding receptor, RAGE, transforming growth factor-beta, D-dimer and interleukin 6, IL-6.

6. The method according to claim 1, wherein the third data set comprises morphological data following a virtual reconstruction of the individual's aortic anatomy by a diagnostic imaging method.

7. The method according to claim 1, wherein the fourth data set comprises hemodynamic and structural parameters of the aorta estimated by a numerical simulation and wherein said hemodynamic and structural parameters are integrated in a bi-directional fluid-structure model.

8. The method according to claim 6, further comprising a processing of the numerical simulation results to display the hemodynamic and structural parameters superimposing them on the virtual reconstruction of the aortic anatomy and extrapolating said parameters in different anatomic positions of the aorta.

9. The method according to claim 7, wherein the hemodynamic and structural parameters comprise at least the blood pressure, shear stress, intramural stress and helicoidal flow index.

10. The method according to claim 1, wherein the fourth data set further comprises information relating to a deformation of the aorta and a time variation of said deformation obtained by applying a time tracking algorithm.

11. The method according to claim 1, further comprising an assessment of the weight of each datum belonging to the first, second, third or fourth data set on the risk index (i).

12. A system for calculating a risk index of aortic rupture or dissection of an individual with ascending thoracic aortic aneurysm, ATAA, the system comprising: first means to obtain a first data set linked to the clinical and/or demographic characteristics of the individual; second means to obtain a second data set linked to the biochemical characteristics of a biological sample of the individual; third means to obtain a third data set linked to the morphological and functional characteristics of the aorta; fourth means to obtain a fourth data set obtained from a processing of the third data set by means of computational modelling; and a computer having a data interface for receiving the first data set the second data set, the third data set and the fourth data set as input data and a processor for processing said data and issuing the risk index (i) of aortic rupture or dissection as output data, integrating the first, the second, the third and the fourth data set in a predictive model, wherein the second data set comprises expression values of at least one biomarker of non-coding RNA chosen from the group consisting of: miR-16, miR-9 miR-101, miR-143, miR-19, miR-21, miR-29, and miR-423-5p.

13. The system according to claim 12, wherein the second data set further comprises expression values of at least one biomarker of Non-coding RNA chosen from the group consisting of: miR-133a, miR-155, miR-320a, miR-34a (MI0001251), and miR-34a (MI0000268).

Description

[0052] These and other aspects of the present invention will become clearer in the light of the following description of some preferred embodiments described below.

[0053] FIG. 1 shows a flow diagram of the method according to the present invention;

[0054] FIG. 2 shows a schematic diagram in a block system of the system according to the present invention;

[0055] FIG. 3 shows a diagram of a parametric model of the ATAA;

[0056] FIG. 4 shows the steps for a numerical simulation of the ATAA;

[0057] FIG. 5A-B show the blood flows of an individual with an ATAA and TAV valve (a) of an individual with an ATAA and BAV valve (b);

[0058] FIG. 6A-B show the shear stress distributions of an individual with an ATAA and TAV valve (a) of an individual with an ATAA and BAV valve (b);

[0059] FIG. 7A-D show the values of shear stress (a) intramural stress (b) the helicoidal flow index (c) and the pressure index for an individual with an ATAA and TAV valve and an individual with an ATAA and BAV valve;

[0060] FIG. 8A-B show the correlation between shear stress (a) and intramural stress (b) with the aortic diameter of the vessel;

[0061] FIG. 9 shows a mapping of the distribution of strain in individuals with ATAA and BAV valve and individuals with ATAA and TAV valve;

[0062] FIG. 10 shows a diagram of the hierarchic CDSS structure according to the present invention.

[0063] FIG. 1 shows in a flow diagram the steps, which characterize the method according to the present invention. Assuming that an individual has been diagnosed with an ascending thoracic aortic aneurysm, the process starts by obtaining 10 a first data set 12 relating to the individual's clinical and/or demographic characteristics, obtaining 20 a second data set 22 relating to the biochemical characteristics of a biological sample taken from the individual and obtaining 30 a third data set 32 relating to the morphological and functional characteristics of the individual's aorta. The first data set 12 can be obtained immediately when the individual is admitted to hospital or to the clinic. With reference to the second data set 22, this can regard the creation of an epigenetic profile by special biomarkers (for example miRNA, MMP, TIMP) following a blood test carried out on the individual. To obtain the third data set 32 it is possible to use a CT, MR scan or an ultrasound of the individual's aorta. From the third data set 32, it is possible to obtain 40 a fourth data set 42. It is possible to estimate the hemodynamics and structural behaviour of the aneurysm by means of a computational simulation.

[0064] Then, the method comprises the step of integrating 50 the different variables relating to the first, the second, the third and the fourth data set 12,22,32,42 in a predictive model. In particular, variables are integrated, which include, amongst others, the individual's demographic data, a personalized computational modelling of the hemodynamics and mechanics of the aneurysm, as well as biomarkers (for example epigenetic biomarkers) obtained from the circulating blood.

[0065] In this way, it is possible to obtain 60 a risk index i based on a physiological behaviour of the vessel and not only on a purely epidemiological criterion. It is worth noting that the predictive model is opportunely calibrated on a previously observed population. Furthermore, since the method is based on an automatic learning process, it is possible to evaluate the weight of each single variable gathered in the clinical control phase in order to quantify the risk of complications of the aneurysm.

[0066] FIG. 2 shows a schematic diagram of the system 100 according to the present invention. The system 100 essentially comprises means 110 for obtaining a first data set 12, means 120 for obtaining a second data set 22, means 130 for obtaining a third data set 32 and means 140 for obtaining a fourth data set 42.

[0067] The data obtained from these means is entered in a computer 150 as input data, which processes it and issues a risk index i which can take a value from 0 to 1, wherein 0 represents no risk (0%) and 1 maximum risk (100%).

Materials and Methods

Epigenetic Profile

[0068] By analysing a sample of about 5 ml of blood taken by venipuncture, it is possible to evaluate the expression of the metalloproteinases (for example, MMP-1, -2, -3, -7, -8, -9), and the TIMP inhibitors thereof (for example, TIMP-1, -2, -3, -4), and the small endogenous molecules of non-coding RNA, in other words, miRNA. Examples of these molecules are miR-21 (access number: MI0000077) linked to endothelial damage, miR-143 (access number: MI0000459) and miR-145 (access number: MI0000461) linked to damage of the smooth-muscle tissue, miR-133a1 (access number: MI0000450) linked to cellular apoptosis, miR-155 (access number: MI0000681), and miR-16 (access number: MI0000070) linked to the inflammatory process of the aortic aneurysm, and finally miR-29b (access number: MI0000105) linked to tissue fibrosis.

[0069] The term biological sample refers in general to a sample of blood taken by venipuncture, but it can also include any other sample, now recognized or subsequently identified, which contains miRNA, MMP and TIMP such as (but not limited to) plasma, tissue or saliva or a combination thereof.

[0070] The identification of the expression of miRNA, MMP and TIMP according to the method described in the present invention is carried out by confirming the presence or absence of one or more miRNA in a biological sample. The miRNA, MMP and TIMP expression level of a patient with an ATAA can be determined using any method/technique recognized to-date as Next Generation Sequencing (for a broad spectrum analysis), polymerase chain reaction (PCR), PCR real-time or using a combination thereof. The difference in the expression of a miRNA, MMP and TIMP molecule between the biological sample of the individual with an ATAA and the biological sample of a healthy individual is indicative of the state of progress of the aneurysm.

[0071] A prospective study was carried out on a total of 32 patients with ATAAs and 16 controls (patients without an aneurysm) and subsequently on n.71 patients with an ATAA classified differently with BAV or TAV. All of the patients were followed at the hospital centre IRCCS Istituto Mediterraneo per i Trapianti e Terapie ad Alta Specializzazione (IRCCS ISMETT) (Mediterranean Institute for Highly Specialized Transplants and Therapies). Such study was approved by the local ethical committee with protocol IRB/04/14. After agreeing to take part in the study, blood samples were obtained from the patients by venipuncture and the expression of the serum levels of a kit of n.42 miRNA was assessed by PCR real-time with kit TaqMan Array MicroRNA A+B Cards. The MMP-2, -3 and -9 and the TIMP-1, -2, -3 and -4 were also assessed.

[0072] ANOVA statistical analysis was carried out followed by a Holm-Sidak post-hoc test to compare the results of the patients with aneurysms with the controls. A significance was adopted for =0.05.

[0073] Table 1 shows the average values (with the addition of standard deviation) for those miR (or miRNA) studied for the ATAAs, which showed a significant difference with the values of the control population. The miR were obtained by PCR real-time on aneurysms and controls. The units of measurement are 2{circumflex over ()}-ddCT, while U16 was chosen for reference.

[0074] In detail, of the n.42 miR studied, statistically significant differences were obtained for n.8 miR (table 1).

TABLE-US-00001 TABLE 1 Comparison of the statistically significant miR expression levels between ATAAs and controls (healthy individuals) Controls ATAA (n = 16) (n = 40) p miR-16 1.13 1.08 6.16 3.86 0.003 (MI0000070) miR-9 1.47 2.02 5.15 3.68 0.023 (MI0000466) miR-101 2.79 5.25 17.7 12.0 0.005 (MI0028598) miR-143 0.79 0.59 21.8 22.61 0.025 (MI0000459) miR-19 1.00 0.83 13.4 27.38 0.002 (MI0028679) miR-21 2.15 2.86 22.6 21.68 0.023 (MI0000077) miR-29 1.35 1.01 4.5 2.71 0.008 (MI0000105) miR-423-5p 1.64 1.49 6.00 5.1 0.042 (MI0001445)

[0075] Furthermore, a different expression was found for the following n.5 miR among patients with BAV versus TAV (table 2) on a new population of 71 patients with an ATAA.

TABLE-US-00002 TABLE 2 Statistically significant miR expression levels of ATAAs with TAV compared to ATAAs with BAV ATAA BAV ATAA TAV (n = 24) (n = 47) p miR-133 1.48 1.508 3.12 2.45 0.007 (MI0000450) miR-155 0.74 3.83 2.73 2.91 0.032 (MI0000681) miR-320a 1.08 2.32 0.32 2.11 0.033 (MI0000542) miR-34a 0.26 2.51 1.27 2.15 0.029 (MI0001251) miR-34a 0.12 2.98 2.20 1.69 0.002 (MI0000268)

[0076] Such miR distinguish the presence of an ATAA in an individual with a TAV from a BAV, with an increased risk of rupture/dissection of the vessel. The miR-34a (MI0000268) has been correlated significantly (R=0.843, p=0.001) with the elastic capacity of the vessel (i.e., stiffness) assessed as the variation in diameter of the vessel in a heartbeat in relation to the variation in blood pressure between systole and diastole. Logistic regression has shown that miR-34a (MI0000268) predicts the presence of an ATAA when age and smoking are used as confusing variables (=0.131, p=0.017, 95% CI upper bound=0.022 and lower bound=26.04).

[0077] Table 3 shows the differences for MMP and TIMP values, which showed a statistically significant difference. In particular, only MMP-9 and TIMP-1 are significant for the decision support. In this case, too, the average is reported with standard deviation and the units of measurement are pg/mL.

TABLE-US-00003 TABLE 3 Comparison of the statistically significant absolute values of the metalloproteinases and tissue inhibitors between controls (healthy individuals) and ATAAs Controls ATAA (n = 16) (n = 40) p MMP-9 (1 10.sup.2) 0.63 0.500 18.54 12.5 0.005 (AAD37404.1) TIMP-1 (1 10.sup.2) 33.28 14.76 52.08 13.55 0.001 (CAG46779.1)

[0078] The conclusion is that such identified molecules of miR, metalloproteinases and respective inhibitors are altered in patients with an aneurysm compared to healthy patients and that they thus have a prognostic significance in the state of progress of the disease.

[0079] C-reactive protein (access number: NP_000558; version: NP_000558.2) determined from serum or heparinized plasma. creatine kinase (access number: NP_001814; version: NP_001814.2), Nt-proBNP (access number: NP_002512; version: NP_002512.1), cardiac troponin I (access number: NP_000354; version: NP_000354.4) and interleukin 6 (access number: NP_000591; version: NP_000591.1) determined from serum or plasma. All of these proteins are evaluated using chemiluminescence technology.

[0080] Advanced glycation end product AGE (access number: P51606; version: P51606.2) and corresponding receptor RAGE (access number: ACF47656; version: ACF47656.1), transforming growth factor beta TGF-beta (access number: NP_000651; version: NP_000651.3), D-dimer (access number: 2Q9I_F; version: 2Q9I_F) determined by kit for enzyme linked (ELISA) immuno-absorbent assay.

Computational Modelling and Numerical Modelling

[0081] Using DICOM data (Digital Imaging and COmmunications in Medicine) of a CT or MR scan, a process of semi-automatic reverse engineering is carried out, based on operations of segmentation and thresholding to reconstruct the anatomy of the vessel including a) the aortic valve with the spatial position of the cusps of the valve, b) the ascending aorta, b) the aortic arch with the supra-aortic trunks, c) the aorta descending until iliac level. This process can be carried out using open-source software, such as, for example vascular tool ITK. For an accurate reconstruction of the morphology of the aortic valve, it is advisable to perform a CT angiography. However, in the absence of this, it is possible to use a parametric model to model the aortic valve, both for a TAV and a BAV, using echocardiographic measurements of the valvular orifice area, the size and spatial position of the cusps.

[0082] FIG. 3 shows the steps for obtaining a parametric model of the ATAA. The first step comprises the anatomical measurements of the aorta. The second step comprises geometrical modelling and the third step represents the final output parametric model.

[0083] It is possible to produce a virtual parametric geometry of the aorta and the valve using CAD modelling techniques (computer-aided design) based on NURBS surfaces. In this case, it is first necessary to have the echocardiographic measurements of the aortic valve and the aorta in different anatomical regions along the longitudinal direction thereof, as shown in FIG. 3. These measurements are: a) diameter of the annulus (D.sub.An); b) the diameter of the sinuses of Valsalva (D.sub.sin); c) the diameter of the tubular junction (D.sub.STJ); d) n.8 diameters of the aorta equally spaced along the longitudinal direction (D.sub.Aoi con i=1,8); e) the distance between the annulus and the sinuses (H.sub.sin); f) the distance between the annulus and the tubular junction (H.sub.STJ); g) the intercommissural distance of each sinus of Valsalva (a, b, c) and h) the corresponding corners on the valve plane (, , , , ).

[0084] In short, the aortic annulus and the sino-tubular junction can be described by a circular shape. The distance between the annulus and the sino-tubular junction is used to position these two circumferences in the space. Whereas, the sinuses of Valsalva can be described by semi-circumferences obtained by interpolating the end points of the intercommissural distance of the aortic valve.

[0085] To model a TAV, it is necessary to use three NURBS surfaces of the third order interpolation to model the leaflets or cusps of the valve. It is possible to adjust the convexity of such leaflets of the valve by check points of the NURBS surface based on the images shown by the transesophageal echocardiography. To model the BAV, two NURBS surfaces of the third order interpolation are needed. Three check points in a system of cylindrical coordinates are created mid-height of each cusp, to control the curvature of the valve cusps. These NURBS surfaces are constrained to the surface of the aortic sinus by morphological operations.

[0086] To generate the geometry of the ascending aorta, the spatial distribution of the diameters of the aorta is used measured along the central axis of the vessel based on the multi-planar views (in other words, sagittal, coronal and axial) of the echocardiographic image. The aorta is assumed to have a circular shape in the transversal plane along the central axis of the vessel. A loft protusion is used to generate a surface, which interpoles the n.8 diameters of the aorta along the central axis of the vessel previously measured by the transesophageal ultrasound. Whereas, the surfaces of the supra-aortic vessels are modelled by means of loft protrusions of the circle, which identifies the diameter of each supra-aortic vessel. This is shown in FIG. 3.

[0087] The numerical simulation according to the present invention adopts the finite elements method (FEM) as numerical technology for the solution of the differential equations for the partial derivatives, which govern the movement of the fluids and aortic mechanics. The numerical solution is carried out using commercial FEM packages.

[0088] The virtual geometry of the anatomy of the aorta is rendered discrete in small elements of finite volume (about 1 million tetrahedral elements for fluid dominion and about 30 thousand quadrilateral-shaped shell elements for structural dominion). The method according to the present invention is based on a bi-directional fluid-structure analysis using MpCCI software (Fraunhofer SCAI, Germany) for coupling the structural component (ABAQUS, SIMULIA Inc., Providence, R.I.), with the math solver of the fluid movement, (FLUENT, ANSYS Inc., Canonsburg, Pa.). The FLUENT and ABAQUS codes share a common border area where the data exchange takes place. The MpCCI algorithm allows data to be exchanged on meshes of non-corresponding elements, by interpolation on the nodes of data obtained from each code.

[0089] In ABAQUS, the biomechanical behaviour of the aorta is modelled as a hyper-plastic and homogeneous material and uses material parameters determined by mechanical tests on samples of patients with BAV or TAV, who have undergone surgical repair of the ATAA. Such biomechanical behaviour model of the aorta considers the dispersion of collagen fibres, which is typical of an aneurysm. The thickness of the vessel (about 1.8 mm for the BAVs and 2.0 mm for TAVs) are assumed as constant. A dynamic/implicit formulation is used to solve the math of the equations, which define the mechanical behaviour of the vessel because of the considerable deformation of the vessel itself. In order for the aorta to deform physiologically, the distal ends of the supra-aortic trunks, the aortic valve and the descending aorta are constrained in all directions.

[0090] In a FLUENT environment, a transitory analysis is carried out for the simulation of the fluid dynamics of the movement of the blood. The blood is assumed to be laminar, incompressible and Newtonian with a density of 1060 kg/m3 and a viscosity of 0.00371 Pas. PISO is used as a pressure-speed coupling algorithm to improve convergence in the immediate vicinity of the distorted elements and the PRESTO scheme as pressure interpolation method. The convergence of the solution is obtained when the remainder of the continuity equation reaches 10-5. The measurement of the transaortic flow is used as the inlet speed of the flow in the aortic valve, measured with a standard echocardiographic exam, which is carried out on the patient as part of their treatment. Whereas, a model with concentrated parameters of the systemic circulation is used for flows leaving the supra-aortic trunks and the abdominal aorta using the blood pressure measurements with a mercury sphygmomanometer. FIG. 4 shows the steps needed for the computational modelling. The first step consists of the acquisition or scanning of CT or MR images. The second step consists of the reconstruction of the virtual anatomy of the aorta in 3D. The third step comprises the inclusion and implementation of data relating to the specific individual like the valvular flow by eco-Doppler, the collagen fibre architecture and the biomechanical properties of the aortic wall. The fourth step includes the application of math equations (Navier-Stokes differential equations). The final step or fifth step consists of the hemodynamics projection, which constitutes the modelling output.

[0091] The post-processing of the results of the computational simulation consists of a) displaying the hemodynamic and structural parameters superimposing them on the patient's virtual anatomy and b) extrapolating these parameters in different anatomical positions of the vessel. Examples of such parameters are blood pressure, shear stress and intramural stress. A coloured map of the parameter of interest is used wherein the colour red indicates, for example a high value (significant risk) while blue indicates, for example a low value (negligible risk).

[0092] Furthermore, the average values of such parameters are extrapolated in different anatomical areas, including the sinus of Valsalva, the sino-tubular junction, the proximal part of the ascending aorta. In particular, the following hemodynamic and structural variables are assessed for every simulation a) shear stress; b) the pressure index described by the 95% highest value of the pressures normalized by the peak thereof; c) the helicoidal flow index as an indicator of three-dimensionality of the flow; d) intramural stress (in terms of Von Mises stress) for the layers of the internal and external tunica of the aorta.

[0093] A retrospective assessment was carried out on a total of 78 patients with ATAA BAV valve (n=42) and ATAA with TAV valve (n=36) followed at the IRCCS Mediterranean Institute for Highly Specialized Transplants and Therapies (ISMETT IRCCS). The study was approved by the ethical committee with protocol IRB/04/14. The following inclusion criteria was used: individuals with ATAAs aged >18 years old. The following exclusion criteria was used: severe high blood pressure; connective tissue disorders; clinical history of surgical operations; aortic stenosis (AS) or aortic insufficiency (AR) more than mild. The classification schemes of BAV aortic valve and the aorta morphology suggested by Schaefer et al were used. The bicuspid aortic valve: an integrated phenotypic classification of leaflet morphology and aortic root shape. Heart 2008; 94:1634-8. Table 4 summarizes the demographic data and the results of the descriptive statistics.

TABLE-US-00004 TABLE 4 Demographic characteristics of patients with an ATAA BAV ATAA TAV ATAA (n = 42) (n = 36) p-value age, years 58 13 65 9 0.061 Sex (%) 76 23 0.004 AR (%) 76 44 0.139 AS (%) 10 23 0.348 Aortic Diameters (mm) sinuses 37.1 4.7 38.8 2.6 0.676 Sino-tubular 35.4 6.2 38.9 6.6 0.132 Junction Aorta 42.7 5.3 45.4 10.0 0.451 Morphology of the aorta (n) Type N 2 3 Type A 15 7 Type E 4 3 BAV Morphology (n) Type 1 14 / Type 2 7 / Orifice Area (mm2) 346.2 88.6 447.8 75.8 0.003 Transaortic Jet 2.0 0.8 1.7 0.6 0.124 (m/s)

[0094] Computational modelling was carried out as described in this invention to assess the hemodynamics and structural mechanics of ATAAs.

[0095] FIG. 5 shows the hemodynamics of two patients with an aneurysm and different aortic valve morphology, while FIG. 6 shows the distribution of shear stress induced by the blood flow.

[0096] FIG. 7 shows the bar diagrams of the average values (plus standard deviation) of the computational variables for the two populations in question. In particular, FIG. 7A shows the shear stress values and FIG. 7B the intramural stress and strain values in different positions of the aorta. FIGS. 7C and 7D show the helicoidal and pressure flow index values respectively. The significant statistical difference is p<0.05.

[0097] The results highlight shear stress (WSS) of the ATAAs with a higher BAV than the one observed in patients with TAV in the sino-tubular junction (6.83.3 N/m2 for BAV and 3.91.3 N/m2 for TAV, p=0.006) and in the ascending aorta (9.83.3 N/m2 for BAV and 7.12.3 N/m2 for TAV, p=0.040, FIG. 7A). A statistically significant difference was observed in the BAVs compared to the TAVs for intramural stress along the ascending aorta (for example, 2.541050.32105 N/m2 for BAV and 2.041050.34105 N/m2 for TAV, p<0.001, FIG. 7B). The hemodynamics appears more disorganized for patients with BAV than in patients with TAV, although not statistically significant.

[0098] FIG. 8 shows the correlation between shear stress (FIG. 8A) and intramural stress (FIG. 8B) with the aortic diameter of the vessel. The results show a statistically significant Pearson correlation between shear stress and the diameter of the ascending aorta (R=0.76, p=0.002. Similarly, it is possible to note a significant correlation between intramural stress and the diameter of the ascending aorta (R=0.89, p=0.003). It is interesting to note that, patients with an ATAA and BAV valve who undergo surgery have higher intramural stress than patients who have not undergone an operation, suggesting that such parameter takes on prognostic significance to distinguish a malign aneurysm from a benign one, as shown by the indications in FIG. 8.

[0099] In conclusion, patients with an ATAA and BAV valve show significantly higher shear stress and intramural tension values than patients with an ATAA and TAV valve when these two groups of patients are compared at the same age and size of the aneurysm. This study was carried out by lowering the effect of the variables, which could confuse the progress of the aneurysm between the two populations and this shows that the differences in the parameters of shear stress and intramural stress are intrinsic for patients with a BAV valve compared to those with a morphologically normal valve. This highlights the importance of using the computational parameters to identify highly stressed areas of the aortic wall, which are consequently at a higher risk of developing complications and thus in need of more attention by the doctor.

Time Tracking Algorithm

[0100] A knowledge of the kinematics of the aortic wall is of considerable importance for assessing the physiopathology of the ATAA. Although a CT angiography is not the standard instrument for aneurysm imaging, this technology is advisable because it allows the size of the aneurysm to be measured both in diastole (in other words, when blood pressure is low and intramural stress is low) and in systole (in other words, when blood pressure is high and intramural stress is high). A CT angiography allows the quantification of parameters, such as deformation (also known as strain)a parameter correlated to cardiac dysfunction and the progression of the disease in some heart pathologies.

[0101] It is possible to extrapolate the morphology of the aorta in different instants of time of the heartbeat from a CT angiography and consequently apply an algorithm for the temporal monitoring of the wall based on analysis techniques of the movement and temporal recognition. This allows the field of movement of the aorta to be estimated for every heartbeat image. The deformation is thus expressed in relation to the initial configuration of the vessel, which is obtained from the image with the highest vessel contraction. It is also possible to calculate the deformation speed (in other words, the strain rate) in relation to the systole time. These parameters can be mapped with a coloured scale on the virtual geometry of the aorta as described previously.

[0102] A retrospective assessment was carried out on a sample of 14 patients with ATAAs both with BAV valve and TAV valve followed at the IRCCS Mediterranean Institute for Highly Specialized Transplants and Therapies (ISMETT). The study was approved by the ethical committee with protocol IRB/04/14.

[0103] Patients without aortic dilation were also enlisted as a negative control and to compare with the data of patients with an ATAA. The patients underwent a CT angiography, according to the radiologists' clinical indications, and not for the specific purpose of this invention. The time tracking algorithm was used to obtain the strain and strain rate of the patients enlisted. Table 5 shows the average values of these two parameters, while FIG. 9, the distribution of the deformation of the ATAA.

TABLE-US-00005 TABLE 5 Comparison of the strain parameter between controls (healthy individuals) and ATAA with BAV or TAV as an indicator of the elasticity of the pathological tissue Controls BAV ATAA TAV ATAA (n = 5) (n = 6) (n = 8) Strain, % 0.08 0.03 0.16 0.04 0.13 0.06 Strain Rate, 1/s.sup.1 0.44 0.09 0.53 0.14 0.49 0.19

[0104] FIG. 10 shows the hierarchical structure of the method or system according to the present invention. All of the data is represented in a hierarchy tree view, using colours and shapes to quickly distinguish the patient's condition and the importance of all of the variables entered. These are the variables relating to the first data set (12), the second data set (22), the third data set (32) and the fourth data set (42). In particular, the risk index of the aneurysm is highlighted in FIG. 10 for a clinical scenario of a 55-year-old patient with a bicuspid valve and aortic ectasia of 3.7 cm, in other words, an aortic dilation, which is not clinically significant and consequently does not require immediate surgery. Such aortic dilation was identified after a first echocardiographic exam, showing good functionality of the valve, which was subsequently confirmed by a CT angiography. Based on the individual's clinical history, the doctor recommends a radiological examination in six months to assess the progress of the aortic dilation. Whereas, the doctor adopts the CDSS presented in this invention to assess the risk for the patient in question. After collecting all of the data, FIG. 10 shows the result of the CDSS, which informs the doctor that there is a potentially high risk of complications related to aortic dilation (being 0.86 out of a maximum of 1). It can be noted from FIG. 10 that the weight of the computational data compared to, for example, the demographic data is high, probably due to considerable hemodynamic alterations induced by the conformation of the bicuspid valve. Therefore, the doctor, who, thanks to the CDSS, has much more information compared to that of the current criterion of the maximum diameter of the aorta, can review the therapy and decide to intervene immediately to avoid the risk of complications caused by the dilation.

[0105] In other words, the risk index obtained using the method according to the present invention uses variables to model the state of progress of the disease. The method transforms the multidisciplinary variables in a common space and aggregates them to obtain the end result. Therefore, an iterative calculation is used to improve the predictive capacity of the model itself and thus the output is represented by the risk index, which is represented in a hierarchy tree view to display the weight, which each variable has on the state of progress of the disease.

[0106] The highly innovative aspect of the present invention lies in the combination of data from biomarkers, such as, for example epigenetic data and computational calculations for a personalized stratification for the patient in question, proposing a method and system for a more rigorous decision-making, to distinguish a benign aneurysm from a malign one, reliably and accurately. Any type of radiological imaging can be used (for example, CAT, magnetic resonance or echocardiography), epigenetic biomarkers, analysis of deformations of the vessel in order to consider information heterogeneous and of any type or scale. Thus, a new paradigm of predictive medicine is proposed, wherein the doctor can benefit from the advantages of a decision instrument, which reconstructs a virtual model of the anatomy of the individual's aneurysm, based on radiological imaging and thus uses numerical simulation technology and biomarkers to quantify the physiological behaviour of the vessel and the risk associated with aneurysm.

[0107] A study, for example on the epigenetic profile of a sample of patients with an ATAA, showed that miRNA molecules obtained from peripheral blood are able to distinguish patients with an aneurysm from healthy patients. Thus, miRNA expression levels have a potential for clinically stratifying patients with an aneurysm. Similarly, degradation of the extracellular matrix is associated with the concentration of MMP, while TIMP inhibitors are important regulators of MMP activity. Thus, the assessment of the concentration of MMP and TIMPS from circulating blood represents a simple method for identifying and monitoring individuals with an ATAA and BAV valve. However, the heterogeneity of the disease does not allow only the measurements of the concentration of MMP and TIMP to be used for the clinical stratification of the aneurysm.

[0108] The computational study in a certain number of patients with ATAAs and different valve morphology showed that the method and system according to the present invention is able to provide important hemodynamic and structural parameters to identify areas of the aortic wall with a higher risk of complications. Computational modelling can allow the development of a new technology for a personalized approach in diagnosing and managing the diseases compared to the traditional guidelines based on epidemiological information. The great innovation of the computational method used is that of integrating structural characteristics of the behaviour of the vessel, specific of an aneurysm and the valve thereof, in a fluid-structure model, which are unique compared to the theoretical assumptions of the other computational analyses. To distinguish the differences between BAV and TAV, the computational model considers the inherent defect of the collagen fibres of the wall of the aneurysmatic aorta and the difference in the mechanical response of the vessel itself. Finally, an analysis of the deformations using a time tracking algorithm represents an in-vivo assessment of the mechanical behaviour of the wall of the vessel and a rapid way of quantifying the structural parameters linked to the presence of an aneurysm.

[0109] An important aspect of the method and system according to the present invention is that of defining artificial intelligence (in other words, an automatic learning system), which is able to provide a parameter essential for optimizing the therapeutic approach of a patient with an ATAA by integrating multidisciplinary and heterogeneous data. In the context of pathologies, for example cancer or Alzheimer's disease, studies have shown that such a support instrument can transform all of the information on a patient in a practical manner, generating knowledge, which can be applied in a clinical setting.

[0110] Numerous further modifications and variations can be made to the method and system described above by a person skilled in the art with the aim of satisfying further, contingent needs, all comprised within the protective scope of the present invention, as defined by the appended claims.