AngioSupport: the design of an interactive tool to allow numerical modelling in clinical decision making

Research output: ThesisPd Eng ThesisAcademic

Abstract

Coronary Artery Disease (CAD) is the most common type of heart disease. It is caused by the buildup of plaque in the arterial walls, which narrows the vessel and reduces blood flow to the heart muscle. The remodeling of the heart can account for these changes, however, eventually increased plaque formation can result in insufficient oxygen in the heart muscle. Over time, CAD can cause heart failure, arrhythmias, ischemia, or a heart attack. CAD often develops slowly over time and the first emerging symptom is chest pain, often only during physical activity.
Treatment of CAD is either by medical therapy or revascularization. The two main revascularization techniques are coronary artery bypass graft (CABG) and percutaneous coronary intervention (PCI). To visualize the coronary vasculature, a coronary angiogram (CA) is made and the Fractional Flow Reserve (FFR) is measured invasively, which has shown to be an excellent indication of the presence of ischemia resulting from blocked blood flow. Currently, for patients with multivessel CAD, clinical outcomes following invasive revascularization depend on many characteristics, such as patient age, cardiac function, coronary disease distribution, previous interventions, diabetes, and/or the number of diseased vessels. Complicated cases are therefore discussed during a heart team meeting, where at least one cardiac surgeon and one intervention-cardiologist discuss each patient. However, choices (in the position, length or diameter) for a CABG or PCI can still be very difficult based on only experience, FFR measurement, and CA.
The aim of this project was to create a patient specific model based interactive tool, named AngioSupport, that could help the heart team in this process by simulating the outcome of PCI and CABG. AngioSupport is designed to fit in the current health care procedure of the Catharina Hospital Eindhoven and allows the heart team to use the models that have been developed at the Technical University of Eindhoven. In this design project, a ‘proof of concept’ has been given with a first prototype of AngioSupport, being developed at LifeTec Group.
In AngioSupport, a 3D representation of the full coronary vasculature of the patient is created by segmentation and pre-processing, using the angiogram images from the patient. Segmentation of the coronary arteries is performed with the help of CAAS software from Pie Medical Imaging. This 3D representation gives a more realistic insight of the patient specific vasculature and an easier interpretation of the blockages. It also allows the use of a one dimensional computational fluid dynamics code (1D CFD code), which can calculate the patient specific pressure throughout the coronary vasculature and is originally developed at the Eindhoven University of Technology (TU/e). With this 1D CFD code, the virtual FFR (vFFR) of the patient can be calculated, which is used to assess the assess whether the heart muscle receives enough blood. AngioSupport also consists of an interface to be used during heart team meetings. The heart team only needs to load in the (pre-processed) patient data and can then start the 1D CFD code themselves. The interface also allows the heart team to virtually perform interventions and simulating the results, which greatly supports in deciding a treatment plan for each patient.
AngioSupport was tested on retrospective data from 10 patients during a user survey with 9 cardiologists. The cardiologists all stated that AngioSupport gives them more insight in each patient and that the AngioSupport interface is easy to use. However, improvements are still needed in AngioSupport, especially for the segmentation of the coronary vasculature. Also the accuracy of the 1D CFD code needs to be increased and validated if AngioSupport is to be used in clinical decision making.
During this project, a patient-specific model-based interactive tool has been developed that can support the heart team in clinical decision making. AngioSupport allows the heart team to use numerical models from TU/e to calculate the pre- and post-operative vFFR to support them in deciding a treatment plan for each patient.
LanguageEnglish
Supervisors/Advisors
  • Lammerts, Ivonne, Supervisor
  • Stijnen, Marco, Supervisor
  • van de Vosse, Frans, Supervisor
  • Huberts, Wouter, Supervisor
  • van 't Veer, Marcel, Supervisor
  • Tonino, W.A.L. (Pim), Supervisor, External person
Award date6 Sep 2019
Place of PublicationEindhoven
Publisher
StatePublished - 6 Sep 2019

Fingerprint

Coronary Artery Disease
Percutaneous Coronary Intervention
Coronary Artery Bypass
Myocardium
Angiography
Clinical Decision-Making
Transplants
Ischemia
Diagnostic Imaging
Hydrodynamics
Therapeutics
Chest Pain
Coronary Disease
Blood Vessels
Cardiac Arrhythmias
Heart Diseases
Coronary Vessels
Software
Heart Failure
Myocardial Infarction

Bibliographical note

PDEng thesis

Cite this

@phdthesis{558128c9bd2f46acb4b7912a3651a06c,
title = "AngioSupport: the design of an interactive tool to allow numerical modelling in clinical decision making",
abstract = "Coronary Artery Disease (CAD) is the most common type of heart disease. It is caused by the buildup of plaque in the arterial walls, which narrows the vessel and reduces blood flow to the heart muscle. The remodeling of the heart can account for these changes, however, eventually increased plaque formation can result in insufficient oxygen in the heart muscle. Over time, CAD can cause heart failure, arrhythmias, ischemia, or a heart attack. CAD often develops slowly over time and the first emerging symptom is chest pain, often only during physical activity.Treatment of CAD is either by medical therapy or revascularization. The two main revascularization techniques are coronary artery bypass graft (CABG) and percutaneous coronary intervention (PCI). To visualize the coronary vasculature, a coronary angiogram (CA) is made and the Fractional Flow Reserve (FFR) is measured invasively, which has shown to be an excellent indication of the presence of ischemia resulting from blocked blood flow. Currently, for patients with multivessel CAD, clinical outcomes following invasive revascularization depend on many characteristics, such as patient age, cardiac function, coronary disease distribution, previous interventions, diabetes, and/or the number of diseased vessels. Complicated cases are therefore discussed during a heart team meeting, where at least one cardiac surgeon and one intervention-cardiologist discuss each patient. However, choices (in the position, length or diameter) for a CABG or PCI can still be very difficult based on only experience, FFR measurement, and CA.The aim of this project was to create a patient specific model based interactive tool, named AngioSupport, that could help the heart team in this process by simulating the outcome of PCI and CABG. AngioSupport is designed to fit in the current health care procedure of the Catharina Hospital Eindhoven and allows the heart team to use the models that have been developed at the Technical University of Eindhoven. In this design project, a ‘proof of concept’ has been given with a first prototype of AngioSupport, being developed at LifeTec Group.In AngioSupport, a 3D representation of the full coronary vasculature of the patient is created by segmentation and pre-processing, using the angiogram images from the patient. Segmentation of the coronary arteries is performed with the help of CAAS software from Pie Medical Imaging. This 3D representation gives a more realistic insight of the patient specific vasculature and an easier interpretation of the blockages. It also allows the use of a one dimensional computational fluid dynamics code (1D CFD code), which can calculate the patient specific pressure throughout the coronary vasculature and is originally developed at the Eindhoven University of Technology (TU/e). With this 1D CFD code, the virtual FFR (vFFR) of the patient can be calculated, which is used to assess the assess whether the heart muscle receives enough blood. AngioSupport also consists of an interface to be used during heart team meetings. The heart team only needs to load in the (pre-processed) patient data and can then start the 1D CFD code themselves. The interface also allows the heart team to virtually perform interventions and simulating the results, which greatly supports in deciding a treatment plan for each patient.AngioSupport was tested on retrospective data from 10 patients during a user survey with 9 cardiologists. The cardiologists all stated that AngioSupport gives them more insight in each patient and that the AngioSupport interface is easy to use. However, improvements are still needed in AngioSupport, especially for the segmentation of the coronary vasculature. Also the accuracy of the 1D CFD code needs to be increased and validated if AngioSupport is to be used in clinical decision making.During this project, a patient-specific model-based interactive tool has been developed that can support the heart team in clinical decision making. AngioSupport allows the heart team to use numerical models from TU/e to calculate the pre- and post-operative vFFR to support them in deciding a treatment plan for each patient.",
author = "{van den Boom}, Tim",
note = "PDEng thesis",
year = "2019",
month = "9",
day = "6",
language = "English",
series = "PDEng report",
publisher = "Technische Universiteit Eindhoven",

}

AngioSupport : the design of an interactive tool to allow numerical modelling in clinical decision making. / van den Boom, Tim.

Eindhoven : Technische Universiteit Eindhoven, 2019. 47 p.

Research output: ThesisPd Eng ThesisAcademic

TY - THES

T1 - AngioSupport

T2 - the design of an interactive tool to allow numerical modelling in clinical decision making

AU - van den Boom,Tim

N1 - PDEng thesis

PY - 2019/9/6

Y1 - 2019/9/6

N2 - Coronary Artery Disease (CAD) is the most common type of heart disease. It is caused by the buildup of plaque in the arterial walls, which narrows the vessel and reduces blood flow to the heart muscle. The remodeling of the heart can account for these changes, however, eventually increased plaque formation can result in insufficient oxygen in the heart muscle. Over time, CAD can cause heart failure, arrhythmias, ischemia, or a heart attack. CAD often develops slowly over time and the first emerging symptom is chest pain, often only during physical activity.Treatment of CAD is either by medical therapy or revascularization. The two main revascularization techniques are coronary artery bypass graft (CABG) and percutaneous coronary intervention (PCI). To visualize the coronary vasculature, a coronary angiogram (CA) is made and the Fractional Flow Reserve (FFR) is measured invasively, which has shown to be an excellent indication of the presence of ischemia resulting from blocked blood flow. Currently, for patients with multivessel CAD, clinical outcomes following invasive revascularization depend on many characteristics, such as patient age, cardiac function, coronary disease distribution, previous interventions, diabetes, and/or the number of diseased vessels. Complicated cases are therefore discussed during a heart team meeting, where at least one cardiac surgeon and one intervention-cardiologist discuss each patient. However, choices (in the position, length or diameter) for a CABG or PCI can still be very difficult based on only experience, FFR measurement, and CA.The aim of this project was to create a patient specific model based interactive tool, named AngioSupport, that could help the heart team in this process by simulating the outcome of PCI and CABG. AngioSupport is designed to fit in the current health care procedure of the Catharina Hospital Eindhoven and allows the heart team to use the models that have been developed at the Technical University of Eindhoven. In this design project, a ‘proof of concept’ has been given with a first prototype of AngioSupport, being developed at LifeTec Group.In AngioSupport, a 3D representation of the full coronary vasculature of the patient is created by segmentation and pre-processing, using the angiogram images from the patient. Segmentation of the coronary arteries is performed with the help of CAAS software from Pie Medical Imaging. This 3D representation gives a more realistic insight of the patient specific vasculature and an easier interpretation of the blockages. It also allows the use of a one dimensional computational fluid dynamics code (1D CFD code), which can calculate the patient specific pressure throughout the coronary vasculature and is originally developed at the Eindhoven University of Technology (TU/e). With this 1D CFD code, the virtual FFR (vFFR) of the patient can be calculated, which is used to assess the assess whether the heart muscle receives enough blood. AngioSupport also consists of an interface to be used during heart team meetings. The heart team only needs to load in the (pre-processed) patient data and can then start the 1D CFD code themselves. The interface also allows the heart team to virtually perform interventions and simulating the results, which greatly supports in deciding a treatment plan for each patient.AngioSupport was tested on retrospective data from 10 patients during a user survey with 9 cardiologists. The cardiologists all stated that AngioSupport gives them more insight in each patient and that the AngioSupport interface is easy to use. However, improvements are still needed in AngioSupport, especially for the segmentation of the coronary vasculature. Also the accuracy of the 1D CFD code needs to be increased and validated if AngioSupport is to be used in clinical decision making.During this project, a patient-specific model-based interactive tool has been developed that can support the heart team in clinical decision making. AngioSupport allows the heart team to use numerical models from TU/e to calculate the pre- and post-operative vFFR to support them in deciding a treatment plan for each patient.

AB - Coronary Artery Disease (CAD) is the most common type of heart disease. It is caused by the buildup of plaque in the arterial walls, which narrows the vessel and reduces blood flow to the heart muscle. The remodeling of the heart can account for these changes, however, eventually increased plaque formation can result in insufficient oxygen in the heart muscle. Over time, CAD can cause heart failure, arrhythmias, ischemia, or a heart attack. CAD often develops slowly over time and the first emerging symptom is chest pain, often only during physical activity.Treatment of CAD is either by medical therapy or revascularization. The two main revascularization techniques are coronary artery bypass graft (CABG) and percutaneous coronary intervention (PCI). To visualize the coronary vasculature, a coronary angiogram (CA) is made and the Fractional Flow Reserve (FFR) is measured invasively, which has shown to be an excellent indication of the presence of ischemia resulting from blocked blood flow. Currently, for patients with multivessel CAD, clinical outcomes following invasive revascularization depend on many characteristics, such as patient age, cardiac function, coronary disease distribution, previous interventions, diabetes, and/or the number of diseased vessels. Complicated cases are therefore discussed during a heart team meeting, where at least one cardiac surgeon and one intervention-cardiologist discuss each patient. However, choices (in the position, length or diameter) for a CABG or PCI can still be very difficult based on only experience, FFR measurement, and CA.The aim of this project was to create a patient specific model based interactive tool, named AngioSupport, that could help the heart team in this process by simulating the outcome of PCI and CABG. AngioSupport is designed to fit in the current health care procedure of the Catharina Hospital Eindhoven and allows the heart team to use the models that have been developed at the Technical University of Eindhoven. In this design project, a ‘proof of concept’ has been given with a first prototype of AngioSupport, being developed at LifeTec Group.In AngioSupport, a 3D representation of the full coronary vasculature of the patient is created by segmentation and pre-processing, using the angiogram images from the patient. Segmentation of the coronary arteries is performed with the help of CAAS software from Pie Medical Imaging. This 3D representation gives a more realistic insight of the patient specific vasculature and an easier interpretation of the blockages. It also allows the use of a one dimensional computational fluid dynamics code (1D CFD code), which can calculate the patient specific pressure throughout the coronary vasculature and is originally developed at the Eindhoven University of Technology (TU/e). With this 1D CFD code, the virtual FFR (vFFR) of the patient can be calculated, which is used to assess the assess whether the heart muscle receives enough blood. AngioSupport also consists of an interface to be used during heart team meetings. The heart team only needs to load in the (pre-processed) patient data and can then start the 1D CFD code themselves. The interface also allows the heart team to virtually perform interventions and simulating the results, which greatly supports in deciding a treatment plan for each patient.AngioSupport was tested on retrospective data from 10 patients during a user survey with 9 cardiologists. The cardiologists all stated that AngioSupport gives them more insight in each patient and that the AngioSupport interface is easy to use. However, improvements are still needed in AngioSupport, especially for the segmentation of the coronary vasculature. Also the accuracy of the 1D CFD code needs to be increased and validated if AngioSupport is to be used in clinical decision making.During this project, a patient-specific model-based interactive tool has been developed that can support the heart team in clinical decision making. AngioSupport allows the heart team to use numerical models from TU/e to calculate the pre- and post-operative vFFR to support them in deciding a treatment plan for each patient.

M3 - Pd Eng Thesis

T3 - PDEng report

PB - Technische Universiteit Eindhoven

CY - Eindhoven

ER -