Anne-Marie Jouannic

Computer Aided Detection and Measurement of Abdominal Aortic Aneurysm Using Computed Tomography Digital Images

Computer-aided detection (CAD) systems, which automatically detect and indicate location of potential abnormalities in scan digital images, have the capacity to increase the accuracy of the radiologists’ interpretations and finding. This paper presents an efficient new CAD .for automatic and accurate detection and quantification of Abdominal Aortic Aneurysm (AAA). The system first detects and extracts the lumen and then identifies the location of the abdominal aortic from the total lumen. The extracted abdominal aortic lumen is then used as an initial surface to segment the abdominal aorta which might contain aneurysm. The geometrical and morphological features of both lumen and aorta are examined for the presence of aneurysm based on predefined criteria set by incorporating prior understanding of the normal expected variation of aorta. The experimental result of the proposed system on 60 CTA datasets indicated a 98% success in detection (CAD) and a 95% in segmentation results (CAM).

Percutaneous endovascular biopsy of intravascular masses: efficacy and safety in establishing pre-therapy diagnosis

AbstractPurposeTo evaluate the efficacy and safety of percutaneous endovascular biopsy (PEB) in intravascular filling-defect lesions (IVLs) of the great vessels.Material and methodsWe retrospectively reviewed 19 patients (age 65 ± 12 years), 11 men and eight women, who underwent PEB for IVLs, between March 2004 and November 2014. All PEBs were performed for early diagnosis and/or characterization of the IVL, or in case of reasonable doubt about the IVL nature. Pre-intervention imaging work-up included CT, MRI and/or PET-CT. PEBs were obtained with a 7F biopsy forceps device. Clinical profile, procedure technical success and safety, and clinical success were evaluated.ResultsPEB was technically successful in all patients (mean of two samples per IVL). No intra- or post-procedural complications were reported. Histopathological analysis provided a diagnosis in all PEBs with a clinical success of 100%. Of the 19 IVLs, 14 were malignant (74%). The most frequent malignant lesion observed was leiomyosarcoma (29%). Benign lesions (26%) included three thrombi (pulmonary artery) and two myxomas.ConclusionPEB is a safe and efficient procedure providing the most effective technique to obtain a tissue sample of high diagnostic quality, which serves to establish early diagnosis in patients with suspected malignant lesions.Key Points• Intravascular filling-defect lesions are related to both benign conditions and malignant tumours. • Endovascular biopsy is indicated in case of doubt about intravascular lesion nature. • Percutaneous endovascular biopsy is a safe technique. • Endovascular biopsy provides tissue samples leading to correct histopathological analysis. • Percutaneous endovascular biopsy provides early diagnosis of malignant intravascular lesions.

A Fuzzy Logic System for Classification of the Lung Nodule in Digital Images in Computer Aided Detection

Digital image analysis technology suffers from imperfection, imprecision and vagueness of the input data and its propagation in all individual components of the technology including image enhancement, segmentation and pattern recognition. Furthermore, a Medical Digital Image Analysis System (MDIAS) such as computer aided detection (CAD) technology deals with another source of uncertainty that is inherent in an image-based practice of medicine. While there are several technology-oriented studies reported in developing CAD applications, no attempt has been made to address, model and integrate these types of uncertainty in the design of the system components even though uncertainty issues directly affect the performance and its accuracy. In order to tackle the problem of uncertainty in the classification design of the system two fuzzy methods are employed and are evaluated for the lung nodule CAD application. The Mamdani model and the Sugeno model of the fuzzy logic system are implemented and the classification results are compared and evaluated through ROC curve analysis and root mean squared error methods. The novelty of the study is to investigate the effect of training algorithms on the performance of the CAD system. The results reveal that the fuzzy logic system with hybrid-training is superior to the other models in terms of root-mean-squared error and ROC curve sensitivity and specificity rates.

Automated Detection of Peripheral Arteries in CTA Datasets

Peripheral artery disease is a chronic disease that manifests in insufficient blood supply to the legs due to narrowing of the arteries. Fully automated detection, segmentation and measurement of stenosis of peripheral vessels from CTA datasets would be highly desirable but has yet to be realized. A key component of this procedure is the development of an automated and accurate method for the segmentation of the peripheral vessel, which would be a major step towards the automated detection of stenosis. We propose a Computer Aided Detection (CAD) algorithm, with which to detect and segment the peripheral vessels directly from 3D data. In order to create a good delineation of arteries in the image, and as to improve the sensitivity for detection and measurement of stenosis, a differential geometry-based approach is employed. This approach serves as an enhancement filter and, further, provides information about the geometry of the structures in the image: the tubular objects representing the interest (arteries). Having enhanced the arteries, a 3D region growing method is employed, utilizing voxel-based geometrical features. With this proposed region growing method the initial seed point is represented by the common iliac arteries junction, and it is thus automatically selected. The method has been successfully implemented on 15 datasets and the evaluation was carried out by the visual judgment of 2 experienced radiologists.

An adaptive 3D region growing algorithm to automatically segment and identify thoracic aorta and its centerline using Computed Tomography Angiography Scans

Thoracic Aortic Aneurysm (TAA) is a localized swelling of the thoracic aorta. The progressive growth of an aneurysm may eventually cause a rupture if not diagnosed or treated. This necessitates the need for an accurate measurement which in turn calls for the accurate segmentation of the aneurysm regions. Computer Aided Detection (CAD) is a tool to automatically detect and segment the TAA in the Computer tomography angiography (CTA) images. The fundamental major step of developing such a system is to develop a robust method for the detection of main vessel and measuring its diameters. In this paper we propose a novel adaptive method to simultaneously segment the thoracic aorta and to indentify its center line. For this purpose, an adaptive parametric 3D region growing is proposed in which its seed will be automatically selected through the detection of the celiac artery and the parameters of the method will be re-estimated while the region is growing thorough the aorta. At each phase of region growing the initial center line of aorta will also be identified and modified through the process. Thus the proposed method simultaneously detect aorta and identify its centerline. The method has been applied on CT images from 20 patients with good agreement with the visual assessment by two radiologists.

Percutaneous intentional intra-luminal-assisted recanalization (PILAR technique) of challenging chronic total occlusions using a high-frequency vibration device

ObjectivesRecanalization of peripheral chronic total occlusions (CTO) is technically challenging especially in cases of in-stent and/or pre-stent and heavily calcified lesions. A high-frequency vibrational device (HFVD) was first used as a secondary-intention device in CTO recanalizations when they were refractory to a guidewire. The aim of this study was to assess the safety and efficacy of the HFVD as a first-line treatment for challenging CTOs and thus to define the percutaneous intentional intraluminal-assisted recanalization (PILAR) technique.MethodsFifty-two patients were treated with the HFVD. Only challenging CTOs were included: 7 pre-stent, 7 in-stent, and 38 highly calcified CTOs. Technical success was defined as the ability to cross the CTO using the HFVD. Secondary outcome was defined as successful intraluminal crossing. Safety endpoints were procedure-related thromboembolism or perforation. Patients were followed up at 3 months and 1 year.ResultsThe technical success rate for recanalization was 90%, of which 83% were intraluminal. The mean recanalized length was 91 ± 44 mm. One thromboembolic complication occurred, which was subsequently treated with thromboaspiration. Three-month and 1-year primary patency rates were 92% and 79%, respectively.ConclusionsHFVD-based PILAR is a safe and effective technique for in-stent or pre-stent CTO recanalization of long and calcified lesions.Key Points• Intraluminal recanalization is the preferred procedure in heavily calcified or pre-/in-stent CTO.• First-line use of assisted intraluminal recanalization for CTO defines the PILAR technique.• HFVD-based PILAR is safe and provides a high success rate for challenging CTO recanalization.