Fuzzy inference analysis of the arterial hypertension effect on aneurysm

Bouharati Imene; Boubendir Nasser-Eddine; Bouharati Khaoula; Laouari Slimane

doi:10.4103/2589-9686.348228

ORIGINAL ARTICLE

Year : 2022 | Volume : 5 | Issue : 2 | Page : 37-41

Fuzzy inference analysis of the arterial hypertension effect on aneurysm

Bouharati Imene¹, Boubendir Nasser-Eddine², Bouharati Khaoula², Laouari Slimane³
¹ Faculty of Medicine, Paris-Sorbonne University, Paris, France; Laboratory of Intelligent Systems, UFAS Ferhat Abbas Setif University, Setif, Algeria
² Faculty of Medicine, Constantine University, Constantine, Algeria
³ Faculty of Medicine, UFAS Ferhat Abbas Setif University, Setif, Algeria

Date of Submission	04-Jan-2022
Date of Decision	25-Jan-2022
Date of Acceptance	28-Jan-2022
Date of Web Publication	24-Jun-2022

Correspondence Address:
Dr. Bouharati Imene
Faculty of Medicine, Paris-Sorbonne University, Paris, France and Laboratory of intelligent systems, UFAS Ferhat Abbas Setif University

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/2589-9686.348228

Abstract

BACKGROUND: Several factors are at the root of the aneurysm. This disease is characterized by the loss of parallelism of the abdominal aorta, which widens to a breaking point. This anomaly is silent, and it is discovered by accident during the diagnosis of another patient. High blood pressure is often associated with the aneurysm. The age factor is directly linked to other factors that are often poorly understood.
SUBJECTS AND METHODS: This study proposes the analysis of these factors by artificial intelligence techniques, in particular fuzzy analysis. This mode of reasoning compensates for the uncertainties inherent in the process. A fuzzy system is established, allowing factors recorded from 100 diagnosed patients to be entered as input variables to the system to read the predicted aortic diameter.
RESULTS: The result at the output of the system will be as precise as possible, because, it is calculated from the aggregation of all the variables. This study compensates for imprecision and uncertainty by considering blood pressure as an imprecise and therefore fuzzy variable
CONCLUSIONS: The result at the output of the system will be as precise as possible, because it is calculated from the aggregation of all the variables. In the absence of a systematic screening program, this system can be a tool to help prevent this disease.

Keywords: Aneurysm, blood pressure, data analysis, fuzzy logic

How to cite this article:
Imene B, Nasser-Eddine B, Khaoula B, Slimane L. Fuzzy inference analysis of the arterial hypertension effect on aneurysm. Vasc Invest Ther 2022;5:37-41

How to cite this URL:
Imene B, Nasser-Eddine B, Khaoula B, Slimane L. Fuzzy inference analysis of the arterial hypertension effect on aneurysm. Vasc Invest Ther [serial online] 2022 [cited 2022 Aug 19];5:37-41. Available from: https://www.vitonline.org/text.asp?2022/5/2/37/348228

Introduction

The risk of aneurysm-related mortality is high due to the fact that it shows no signs of not being treated in time^[1],[2],[3] Enlargement of the abdominal aorta is often silent.^[4] This disease is often found when diagnosed with another disease.^[5] The aneurysm is the result of multiple factors mainly related to age.^{[6],[7],[8],[9]}

If the male sex is more concerned than the female sex, blood pressure remains the dominant factor. However, there has been a decrease in recent years.^[10],[11] Screening programs are established to monitor and follow the progress of this disease. In these programs, it is expected that optimal monitoring should take into account diagnostic intervals that are inversely proportional to the diameters of the abdominal aorta.^[12],[13]

At a certain threshold, the risk of rupturing the aorta is expected.^[14] This disease concerns geographical areas which remain to be defined.^[15],[16]

Given the complexity of the factors that cause this disease, this study is limited to analyzing the effect of high blood pressure. Also, since these factors are uncertain and imprecise, this study proposes their analysis using artificial intelligence techniques. The principles of fuzzy inference are applied. A population of 100 patients is diagnosed at the level of the radiology department of the Setif hospital in Algeria and various private clinics in the city during the period 2019–2020. The blood pressure taken from patients is related to the diameter of the abdominal aorta measured by computed tomography (CT) scan in radiological examination. A fuzzy analysis system for these factors is proposed. As hypertension is not the only factor implicated, the factors analyzed are considered fuzzy and therefore uncertain. By their fuzzyfication, the uncertainties are compensated. A basis for the rules is established from the actual measured values. The diameter of the aorta is mapped to the inputs. In the absence of a monitoring and screening program for these patients, this system makes it possible to predict this disease and therefore constitutes a diagnostic aid tool.

Arterial hypertension

The condition of the vascular walls is affected by the loss of elasticity with advanced age.^[17] No doubt, blood pressure is a function of age. This is reported by studies, which give as an example in the European continent, which has an aging population.^[18] While studies prove the direct link between high blood pressure and age, others refute it.^[19] This is to elucidate vascular physiology and its mechanism on the hypertension-aneurysm relationship. What is evident is that the pressure exerted on the internal walls of the vessels contributes to the extension of these. Since several factors can come into play, including high blood pressure, the system remains very complex to analyze or model.

Role of imaging

Before any surgery, it is necessary to take an X-ray reading. Doppler ultrasound can detect any abnormalities in vascular functioning. This imaging technique is used in preparatory diagnosis.^[20],[21] Doppler ultrasound has the advantage of being often available and less invasive.^{[22],[23],[24],[25]} In practice, to reveal more detail, CT imaging is used, which provides better image quality.^[26] For better anatomical vision, CT venography is proven in the presentation of anatomical structures.^[27],[28]

Materials and Methods

The study sample includes 100 patients diagnosed during the period from 2017 to 2020 at the Radiology Department of Setif University Hospital in Algeria as well as at other private clinics in the city. With each patient, in addition to other clinical examinations, the blood pressure is taken. Radiological imaging is obtained by ultrasound and CT scan. A database is established. In addition to blood pressure, this database contains other test results. Blood pressure is mapped to the diameter of the abdominal aorta. This collected data is analyzed by a fuzzy inference system that takes into account the uncertainty and imprecision of the data.

Fuzzy analysis

Given the complexity of the system, classical mathematical tools remain inapplicable to give exact results. This study offers intelligent data analysis. The principles of fuzzy logic are applied. A system is built with three input variables (Age, Gender, Blood pressure) and one output variable (Aortic diameter).

Each variable is considered fuzzy, which requires its fuzzyfication. This operation consists of its conversion into a linguistic variable expressed in human language. This way of operating compensates for inaccuracies. A rule base is built, connecting the input variables to the output variable according to the measured values. The output result is calculated by aggregating all the rules. By this, all possibilities are taken care of [Figure 1].

Figure 1: Block diagram of the system

Click here to view

Fuzzyfication of variables

Input variables

The input variable “Age” is fuzzified into three triangular-shaped membership functions:

Young: (0–30 years old); adult: (25–60 years old); Old: (55–100 years)
We see the creation of an overlap interval between two neighboring functions to compensate for the imprecision associated with the allocation of ages [Figure 2]
The input variable “Gender” is not fuzzified. Numerical values are assigned to each genre. Male: (1); Female: (2) [Figure 3]
The input variable “Blood pressure” is fuzzified into three triangular-shaped membership functions. Blood pressure is assigned numeric ranges based on their severity
Low: (0–2); Medium: (1–3); High: (2–4)
We note the creation of an overlap interval between two neighboring functions to compensate for the imprecision related to the assignment of the degrees of severity [Figure 4].

Figure 2: Fuzzyfication of the variable “Age”

Click here to view

Figure 3: Representation of the variable “Gender”

Click here to view

Figure 4: Fuzzyfication of the variable “Blood pressure”

Click here to view

Output variable

The output variable represents the diameter of the abdominal aorta. This variable is fuzzified into three triangular-shaped membership functions:

Normal: (15–25 mm); Risky: (20–40 mm); Serious: (35–60 mm) [Figure 5].

Figure 5: Fuzzyfication of the variable “Aorta diameter”

Click here to view

Code

[System]

Name='Blood Pressure-Aneurysm'

Type='mamdani'

Version = 2.0

NumInputs = 3

NumOutputs = 1

NumRules = 20

AndMethod='min'

OrMethod='max'

ImpMethod='min'

AggMethod='max'

DefuzzMethod='centroid'

[Input1]

Name='Age'

Range= [0 100]

NumMFs = 3

MF1='Young':'trimf',[0 15 30]

MF2='Adult':'trimf',[25 42.5 60]

MF3='Old':'trimf',[55 60 100000000]

[Input2]

Name='Gender'

Range= [0 3]

NumMFs = 2

MF1='Male':'trimf',[1 1 1]

MF2='Female':'trimf',[2 2 2]

[Input3]

Name='Bood. Pressure'

Range= [1 4]

NumMFs = 3

MF1='Low':'trimf',[1 1.75 2.5]

MF2='Medium':'trimf',[1.75 2.5 3.25]

MF3='High':'trimf',[2.5 3.25 4]

[Output1]

Name='Aorta. Diameter'

Range = [15 60]

NumMFs = 3

MF1='Normal':'trimf',[15 20 25]

MF2='Risky':'trimf',[20 30 40]

MF3='Serious':'trimf',[35 47.5 60]

Basis of the rules

This rule base consists of the mapping between the input variables and the output variable. Each rule is introduced from the measured values of each patient. The higher precision results from the larger number of rules. This is explained by the fact that the result at the output takes into consideration the combination of all the rules and therefore all the possibilities

The general form of a rule:

IF X1 is x11 AND X2 is x22 AND X3 is x33 THAN Y is Y1

Result and Discussion

Several studies have been conducted to investigate and highlight the effect of high blood pressure on aneurysms. Other studies also try to analyze and model the effect of the latter on the dilation of the abdominal aorta. However, since it is a complex system, the consideration of blood pressure as an imprecise and therefore fuzzy variable in this study compensates for the imprecision and uncertainty. Basically, age and other factors are involved in the process of the appearance of blood pressure. By establishing rules of inference from the values of the real cases, and by fuzzyfying them, this study compensates for the imprecision. The result obtained from the diameter of the aorta as a function of the input variables is presented as the most precise possible [Figure 6].

Figure 6: Example of application

Click here to view

Conclusions

Several age-related factors have been shown to cause aneurysm. This silent disease with no obvious signs is detected accidentally when diagnosed with another disease. Some countries have developed a systematic screening program for its detection and monitoring of its progress. The factors involved are poorly defined and complex to analyze; blood pressure has a lot to do with it. This study attempts to analyze the effect of blood pressure on the enlargement of the diameter of the abdominal aorta. Given the uncertainty and imprecision of the data, a fuzzy inference analysis is proposed. Since fuzzy logic deals with uncertainty, its application in this area is adequate. Considering age as a factor that affects all patients and sex as a variable because it affects men much more than women, blood pressure is also considered to be variable in this study. These factors are considered to be system input variables and fuzzified and are matched with the diameter of the abdominal aorta in diagnosed cases. The rule base established must take into account all possible combinations. Introducing the variables randomly at the inputs makes it possible to read the diameter of the corresponding aorta at the output [Figure 6]. This is therefore a preventive tool in the absence of a systematic screening program.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Upchurch GR Jr., Schaub TA. Abdominal aortic aneurysm. Am Fam Physician 2006;73:1198e204.
2.	Sampson UK, Norman PE, Fowkes FG, Aboyans V, Song Y, Harrell FE Jr., et al. Estimation of global and regional incidence and prevalence of abdominal aortic aneurysms 1990 to 2010. Glob Heart 2014;9:159-70.
3.	Cosford PA, Leng GC. Screening for abdominal aortic aneurysm. Cochrane Database Syst Rev 2007;(2):CD002945.
4.	Kuivaniemi H, Elmore JR. Opportunities in abdominal aortic aneurysm research: Epidemiology, genetics, and pathophysiology. Ann Vasc Surg 2012;26:862-70.
5.	Lederle FA, Simel DL. The rational clinical examination. Does this patient have abdominal aortic aneurysm? JAMA 1999;281:77-82.
6.	Mayer D, Rancic Z, Meier C, Pfammatter T, Veith FJ, Lachat M. Open abdomen treatment following endovascular repair of ruptured abdominal aortic aneurysms. J Vasc Surg 2009;50:1-7.
7.	Laine MT, Vänttinen T, Kantonen I, Halmesmäki K, Weselius EM, Laukontaus S, et al. Rupture of abdominal aortic aneurysms in patients under screening age and elective repair threshold. Eur J Vasc Endovasc Surg 2016;51:511-6.
8.	Nevidomskyte D, Shalhub S, Singh N, Farokhi E, Meissner MH. Influence of gender on abdominal aortic aneurysm repair in the community. Ann Vasc Surg 2017;39:128-36.
9.	Thompson SG, Brown LC, Sweeting MJ, Bown MJ, Kim LG, Glover MJ, et al. Systematic review and meta-analysis of the growth and rupture rates of small abdominal aortic aneurysms: Implications for surveillance intervals and their cost-effectiveness. Health Technol Assess 2013;17:1-118.
10.	Svensjö S, Björck M, Gürtelschmid M, Djavani Gidlund K, Hellberg A, Wanhainen A. Low prevalence of abdominal aortic aneurysm among 65-year-old Swedish men indicates a change in the epidemiology of the disease. Circulation 2011;124:1118-23.
11.	Lilja F, Wanhainen A, Mani K. Changes in abdominal aortic aneurysm epidemiology. J Cardiovasc Surg (Torino) 2017;58:848-53.
12.	Chaikof EL, Dalman RL, Eskandari MK, Jackson BM, Lee WA, Mansour MA, et al. The society for vascular surgery practice guidelines on the care of patients with an abdominal aortic aneurysm. J Vasc Surg 2018;67:2-77.e2.
13.	Wanhainen A, Verzini F, Van Herzeele I, Allaire E, Bown M, Cohnert T, et al. Editor's choice e European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms. Eur J Vasc Endovasc Surg 2019;57:8-93.
14.	Raimund E, Holger E. Aortic dimensions and the risk of dissection. Heart 2006;92:137-42. doi: 10.1136/hrt.2004.055111.
15.	Beck AW, Sedrakyan A, Mao J, Venermo M, Faizer R, Debus S, et al. Variations in abdominal aortic aneurysm care: A report from the international consortium of vascular registries. Circulation 2016;134:1948-58.
16.	Waton S, Johal A, Heikkila K, Cromwell D. Clinical Effectiveness Unit, The Royal College of Surgeons of England, National Vascular Registry 2015 Annual Report; 2015. Available from: https://www.vascularsociety.org.uk/_userfiles/pages/files/Resources/NVR-2015-Annual-Report.pdf. [Last accessed on 2019 Mar 22].
17.	Shuai T, Kan Y, Si Y, Fu W. High-risk factors related to the occurrence and development of abdominal aortic aneurysm. J Interv Med 2020;3:80-2.
18.	Li X, Zhao G, Zhang J, Duan Z, Xin S. Prevalence and trends of the abdominal aortic aneurysms epidemic in general population – A meta-analysis. PLoS One 2013;8:e81260.
19.	Schlösser FJ, Tangelder MJ, Verhagen HJ, van der Heijden GJ, Muhs BE, van der Graaf Y, et al. Growth predictors and prognosis of small abdominal aortic aneurysms. J Vasc Surg 2008;47:1127-33.
20.	Nwafor IA, Eze JC, Ezemba N, Chinawa JM, Idoko LF, Ngene CN. Management of varicose veins of the lower extremities: A 10-year institutional experience. J Vasc Med Surg 2017;5:309. [doi: 10.4172/2329-6925.1000309].
21.	Hardman RL, Rochon PJ. Role of interventional radiologists in the management of lower extremity venous insufficiency. RadioGraphics 2009;29:525-36.
22.	Mercer KG, Scott DJ, Berridge DC. Preoperative duplex imaging is required before all operations for primary varicose veins. Br J Surg 1998;85:1495-7.
23.	Jutley RS, Cadle I, Cross KS. Preoperative assessment of primary varicose veins: A duplex study of venous incompetence. Eur J Vasc Endovasc Surg 2001;21:370-3.
24.	Van der Heijden FH, Bruyninckx CM. Preoperative color-coded duplex scanning in varicose veins of the lower extremity. Eur J Surg 1993;159:329-33.
25.	Dixon PM. Duplex ultrasound in the pre-operative assessment of varicose veins. Australas Radiol 1996;40:416-21.
26.	Pleass HC, Holdsworth JD. Audit of introduction of hand-held Doppler and duplex ultrasound in the management of varicose veins. Ann R Coll Surg Engl 1996;78:494-6.
27.	Lee W, Chung JW, Yin YH, Jae HJ, Kim SJ, Ha J, et al. Three-dimensional CT venography of varicose veins of the lower extremity: Image quality and comparison with doppler sonography vascular and interventional radiology. AJR Am J Roentgenol 2008;191:1186-91.
28.	Kim R, Lee W, Park EA, Yoo JY, Chung JW. Anatomic variations of lower extremity venous system in varicose vein patients: Demonstration by three-dimensional CT venography. Acta Radiol 2017;58:542-9.