Computer Aided Diagnosis (CAD) systems assist clinicians including radiologists and cardiologists to detect abnormalities and highlight conspicuous possible disease. Implementing a pre-operative CAD system contains a framework that accepts related technical as well as clinical parameters as input by analyzing the predefined method and demonstrates the prospective output. In this work we developed the Computer Aided Diagnostic System for biomedical imaging analysis of two applications on Breast Cancer and Atherosclerosis. The aim of the first CAD application is to optimize the registration strategy specifically for Breast Dynamic Infrared Imaging and to make it user-independent. Base on the fact that automated motion reduction in dynamic infrared imaging is on demand in clinical applications, since movement disarranges time-temperature series of each pixel, thus originating thermal artifacts that might bias the clinical decision. All previously proposed registration methods are feature based algorithms requiring manual intervention. We implemented and evaluated 3 different 3D time-series registration methods: 1. Linear affine, 2. Non-linear Bspline, 3. Demons applied to 12 datasets of healthy breast thermal images. The results are evaluated through normalized mutual information with average values of 0.70±0.03, 0.74±0.03 and 0.81±0.09 (out of 1) for Affine, BSpline and Demons registration, respectively, as well as breast boundary overlap and Jacobian determinant of the deformation field. The statistical analysis of the results showed that symmetric diffeomorphic Demons registration method outperforms also with the best breast alignment and non-negative Jacobian values which guarantee image similarity and anatomical consistency of the transformation, due to homologous forces enforcing the pixel geometric disparities to be shortened on all the frames. We propose Demons registration as an effective technique for time-series dynamic infrared registration, to stabilize the local temperature oscillation. The aim of the second implemented CAD application is to assess contribution of calcification in plaque vulnerability and wall rupture and to find its maximum resistance before break in image-based models of carotid artery stenting. The role of calcification inside fibroatheroma during carotid artery stenting operation is controversial in which cardiologists face two major problems during the placement: (i) “plaque protrusion” (i.e. elastic fibrous caps containing early calcifications that penetrate inside the stent); (ii) “plaque vulnerability” (i.e. stiff plaques with advanced calcifications that break the arterial wall or stent). Finite Element Analysis was used to simulate the balloon and stent expansion as a preoperative patient-specific virtual framework. A nonlinear static structural analysis was performed on 20 patients acquired using in vivo MDCT angiography. The Agatston Calcium score was obtained for each patient and subject-specific local Elastic Modulus (EM) was calculated. The in silico results showed that by imposing average ultimate external load of 1.1MPa and 2.3MPa on balloon and stent respectively, average ultimate stress of 55.7±41.2kPa and 171±41.2kPa are obtained on calcifications. The study reveals that a significant positive correlation (R=0.85, p<0.0001) exists on stent expansion between EM of calcification and ultimate stress as well as Plaque Wall Stress (PWS) (R=0.92, p<0.0001), comparing to Ca score that showed insignificant associations with ultimate stress (R=0.44, p=0.057) and PWS (R=0.38, p=0.103), suggesting minor impact of Ca score in plaque rupture. These average data are in good agreement with results obtained by other research groups and we believe this approach enriches the arsenal of tools available for pre-operative prediction of carotid artery stenting procedure in the presence of calcified plaques.

Preoperative Systems for Computer Aided Diagnosis based on Image Registration: Applications to Breast Cancer and Atherosclerosis / RIYAHI ALAM, MOHAMAD SADEGH. - (2015). [10.6092/polito/porto/2592170]

Preoperative Systems for Computer Aided Diagnosis based on Image Registration: Applications to Breast Cancer and Atherosclerosis

RIYAHI ALAM, MOHAMAD SADEGH
2015

Abstract

Computer Aided Diagnosis (CAD) systems assist clinicians including radiologists and cardiologists to detect abnormalities and highlight conspicuous possible disease. Implementing a pre-operative CAD system contains a framework that accepts related technical as well as clinical parameters as input by analyzing the predefined method and demonstrates the prospective output. In this work we developed the Computer Aided Diagnostic System for biomedical imaging analysis of two applications on Breast Cancer and Atherosclerosis. The aim of the first CAD application is to optimize the registration strategy specifically for Breast Dynamic Infrared Imaging and to make it user-independent. Base on the fact that automated motion reduction in dynamic infrared imaging is on demand in clinical applications, since movement disarranges time-temperature series of each pixel, thus originating thermal artifacts that might bias the clinical decision. All previously proposed registration methods are feature based algorithms requiring manual intervention. We implemented and evaluated 3 different 3D time-series registration methods: 1. Linear affine, 2. Non-linear Bspline, 3. Demons applied to 12 datasets of healthy breast thermal images. The results are evaluated through normalized mutual information with average values of 0.70±0.03, 0.74±0.03 and 0.81±0.09 (out of 1) for Affine, BSpline and Demons registration, respectively, as well as breast boundary overlap and Jacobian determinant of the deformation field. The statistical analysis of the results showed that symmetric diffeomorphic Demons registration method outperforms also with the best breast alignment and non-negative Jacobian values which guarantee image similarity and anatomical consistency of the transformation, due to homologous forces enforcing the pixel geometric disparities to be shortened on all the frames. We propose Demons registration as an effective technique for time-series dynamic infrared registration, to stabilize the local temperature oscillation. The aim of the second implemented CAD application is to assess contribution of calcification in plaque vulnerability and wall rupture and to find its maximum resistance before break in image-based models of carotid artery stenting. The role of calcification inside fibroatheroma during carotid artery stenting operation is controversial in which cardiologists face two major problems during the placement: (i) “plaque protrusion” (i.e. elastic fibrous caps containing early calcifications that penetrate inside the stent); (ii) “plaque vulnerability” (i.e. stiff plaques with advanced calcifications that break the arterial wall or stent). Finite Element Analysis was used to simulate the balloon and stent expansion as a preoperative patient-specific virtual framework. A nonlinear static structural analysis was performed on 20 patients acquired using in vivo MDCT angiography. The Agatston Calcium score was obtained for each patient and subject-specific local Elastic Modulus (EM) was calculated. The in silico results showed that by imposing average ultimate external load of 1.1MPa and 2.3MPa on balloon and stent respectively, average ultimate stress of 55.7±41.2kPa and 171±41.2kPa are obtained on calcifications. The study reveals that a significant positive correlation (R=0.85, p<0.0001) exists on stent expansion between EM of calcification and ultimate stress as well as Plaque Wall Stress (PWS) (R=0.92, p<0.0001), comparing to Ca score that showed insignificant associations with ultimate stress (R=0.44, p=0.057) and PWS (R=0.38, p=0.103), suggesting minor impact of Ca score in plaque rupture. These average data are in good agreement with results obtained by other research groups and we believe this approach enriches the arsenal of tools available for pre-operative prediction of carotid artery stenting procedure in the presence of calcified plaques.
2015
File in questo prodotto:
File Dimensione Formato  
tesi_dottorato_Sadegh.pdf

Open Access dal 25/02/2015

Tipologia: Tesi di dottorato
Licenza: Creative commons
Dimensione 6.01 MB
Formato Adobe PDF
6.01 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2592170
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo