Detection of Connective Tissue Disorders from 3D Aortic MR Images Using Independent Component Analysis

Abstract

The current report concerns methods of early detection of connective tissue disorders leading to aortic aneurysms and dissections. Automated and accurate segmentation of the aorta in 4D (3D + time) MR image data is reviewed, and a computer-aided diagnosis (CAD) method using independent component analysis is reported. This admits the objective identification of subjects with connective tissue disorders from 4D aortic MR images.The majority of the presented work is concentrated on independent component analysis(ICA), estimating sources to be used for the diagnosis task. Prior knowledge of the source distribution is utilized using an ordering of the components. Two new ordering measures are introduced in current work. A novel approach to constrained dimensionality reduction in ICA is developed. A new idea of time-invariant independent components is introduced, and assists in the disease detection in the presence of sparse data.4D MR image data sets acquired from 21 normal and 10 diseased subjects are used to evaluate the efficiency of the method. The automated 4D segmentation result produces accurate aortic surfaces. The ICA results are validated by a leave-one-out classification test, and are further substantiated by visual inspection of the components. Using a single phase of the cardiac cycle, 8 out of 10 diseased subjects are identified and the specificity is 100 %, classifying all 21 healthy subjects correctly. These results are obtained using components showing correspondence to clinical observations. With 4D information included, the CAD method classifies 9 out of 10 diseased correctly, and still the specificity is 100 %. ii ResuméDen indevaerende rapport vedrører metoder til tidlig detektering af bindevaevssygdomme, som fører til aortic aneurysms og dissections. En automatisk og praecis metode til segmentering af aorta i 4D (3D + tid) MR data er refereret og en computerassisteret diagnose (CAD) metode, der involverer brugen af independent component analysis, er rapporteret. Dette muliggør en objektiv identificering af subjekter med bindevaevssygdomme, udfra 4D MR billeder af aorta.Hovedparten af det fremlagte arbejde er koncentreret omkring independent component analysis (ICA), som estimerer kilder, der bruges under diagnose opgaven. A priori viden om kildernes fordeling er udnyttet til udformningen af en sortering af de fundne komponenter. To nye sortereringsmål er fremført i det indevaerende arbejde. En ny tilgang til dimsionsreducering under bibetingelser i ICA er udviklet. Et nyt koncept om en tidsinvariant independent component er desuden introduceret, hvilket assisterer til sygdomsdetekteringen, når der kun er en staerkt begraenset maengde data til rådighed.4D MR billedsaet, optaget af 21 normale og 10 syge subjekter, er brugt til at evaluere effektiviteten af den udviklede metode. Den automatiserede 4D segmentering giver en nøjagtig aorta overflade. ICA resultaterne er valideret ved en leave-one-out klassificeringstest, og er yderligere underbygget ved visuel inspektion af de fundne komponent...

Estimation of Independent Non-Linear Deformation Modes for Analysis of Craniofacial Malformations in Crouzon Mice

“…ICA has gained increasing popularity, and due to its inherent maximization of sparsity, it has previously been used for shape analysis, e.g. Hansen et al [6]. To our knowledge this is the first time ICA has been applied to statistical analysis of a deformation model.…”

Scoliosis Follow-Up Using Noninvasive Trunk Surface Acquisition

“…Recently, ICA has been used for image segmentation [30] and medical image analysis [8,19,41]. Boquete et al [8] have proposed a thermographic image analysis based on ICA for automated detection of high tumor risk areas.…”

“…Boquete et al [8] have proposed a thermographic image analysis based on ICA for automated detection of high tumor risk areas. Hassen et al [19] used ICA to built a cardiovascular disease diagnosis system based on magnetic resonance imaging. ICA in a high-dimensional space with sparse data was applied to landmarked 3D shapes resulting from the aortic segmentation.…”

Analysis of scoliosis trunk deformities using ICA

“…Among them, we have feature extraction or data representation, which consist of computing a small vector representation of data (sound, image, etc) [11,18,4]. Recently, ICA is used for image segmentation task [17] and medical image analysis [6,10,21]. Boquete et al [6] have proposed a thermographic image analysis based on ICA for automated detection of high tumor risk areas.…”

“…Boquete et al [6] have proposed a thermographic image analysis based on ICA for automated detection of high tumor risk areas. Hassen et al [10] used ICA to built cardiovascular disease diagnosis based on magnetic resonance imaging. ICA in a high-dimensional space with sparse data was applied to landmarked 3D shapes resulting from the aortic segmentation.…”