ABSTRACT
OBJECTIVE: The aim of this study is to provide an automatic framework for computer-aided analysis of multiparametric magnetic resonance (mp-MR) images of prostate. METHOD: We introduce a novel method for the unsupervised analysis of the images. An evidential C-means classifier was adapted for use with a segmentation scheme to address multisource data and to manage conflicts and redundancy. RESULTS: Experiments were conducted using data from 15 patients. The evaluation protocol consisted in evaluating the method abilities to classify prostate tissues, showing the same behaviour on the mp-MR images, into homogeneous classes. As the actual diagnosis was available, thanks to the correlation with histopathological findings, the assessment focused on the ability to segment cancer foci. The method exhibited global sensitivity and specificity of 70 and 88 %, respectively. CONCLUSION: The preliminary results obtained by these initial experiments showed that the method can be applied in clinical routine practice to help making decision especially for practitioners with limited experience in prostate MRI analysis.
Subject(s)
Diagnosis, Computer-Assisted/methods , Diffusion Magnetic Resonance Imaging/methods , Image Processing, Computer-Assisted/methods , Magnetic Resonance Imaging/methods , Prostate/pathology , Prostatic Neoplasms/pathology , Algorithms , Computer Simulation , Contrast Media/chemistry , Humans , Male , Models, Statistical , Pattern Recognition, Automated , Predictive Value of Tests , Sensitivity and Specificity , SoftwareABSTRACT
Computer-aided medical interventions and medical robotics for prostate cancer have known an increasing interest and research activity. However before the routine deployment of these procedures in clinical practice becomes a reality, in vivo and in silico validations must be undertaken. In this study, we developed a digital morphologic atlas of the prostate. We were interested by the gland, the peripheral zone and the central gland. Starting from an image base collected from 30 selected patients, a mean shape and most important deformations for each structure were deduced using principal component analysis. The usefulness of this atlas was highlighted in two applications: image simulation and physical phantom design.
Subject(s)
Image Interpretation, Computer-Assisted/methods , Imaging, Three-Dimensional/methods , Magnetic Resonance Imaging/methods , Models, Anatomic , Prostate/anatomy & histology , Aged , Humans , Image Enhancement/methods , Male , Middle Aged , Reproducibility of Results , Sensitivity and SpecificityABSTRACT
PURPOSE: To investigate the performance of a new method of automatic segmentation of prostatic multispectral magnetic resonance images into two zones: the peripheral zone and the central gland. METHODS: The proposed method is based on a modified version of the evidential C-means clustering algorithm. The evidential C-means optimization process was modified to introduce spatial neighborhood information. A priori knowledge of the prostate's zonal morphology was modeled as a geometric criterion and used as an additional data source to enhance the differentiation of the two zones. RESULTS: Thirty-one clinical magnetic resonance imaging series were used to validate the method, and interobserver variability was taken into account in assessing its accuracy. The mean Dice Similarity Coefficient was 89% for the central gland and 80% for the peripheral zone, as validated by a consensus from expert radiologist segmentation. CONCLUSIONS: The method was statistically insensitive to variations in patient age, prostate volume and the presence of tumors, which increases its feasibility in a clinical context.
Subject(s)
Image Processing, Computer-Assisted/methods , Magnetic Resonance Imaging/methods , Prostate , Humans , Male , Middle Aged , Prostatic Neoplasms/diagnosisABSTRACT
PURPOSE: Computerized detection of prostate cancer on T2-weighted MR images. METHODS: The authors combined fractal and multifractal features to perform textural analysis of the images. The fractal dimension was computed using the Variance method; the multifractal spectrum was estimated by an adaptation of a multifractional Brownian motion model. Voxels were labeled as tumor/nontumor via nonlinear supervised classification. Two classification algorithms were tested: Support vector machine (SVM) and AdaBoost. RESULTS: Experiments were performed on images from 17 patients. Ground truth was available from histological images. Detection and classification results (sensitivity, specificity) were (83%, 91%) and (85%, 93%) for SVM and AdaBoost, respectively. CONCLUSIONS: Classification using the authors' model combining fractal and multifractal features was more accurate than classification using classical texture features (such as Haralick, wavelet, and Gabor filters). Moreover, the method was more robust against signal intensity variations. Although the method was only applied to T2 images, it could be extended to multispectral MR.
Subject(s)
Fractals , Image Processing, Computer-Assisted/methods , Magnetic Resonance Imaging/methods , Prostatic Neoplasms/diagnosis , Algorithms , Artificial Intelligence , Humans , Male , ROC CurveABSTRACT
This paper describes the methodology and the evaluation of a 3D skeletonization algorithm applied on brain vascular structure. This method is based on the application of the minimum cost-spanning tree using Dijkstra's algorithm and seems well appropriate to tubular objects. We briefly describe the different steps, from the segmentation to the skeleton analysis. Besides, we propose an original evaluation scheme of the method based on digital phantom and clinical data. The final aim of this work is to provide a symbolic description framework applied to cerebro-vascular structures.
Subject(s)
Brain/blood supply , Electroencephalography/methods , Imaging, Three-Dimensional/methods , Algorithms , Cerebrovascular Circulation , Humans , Image Processing, Computer-Assisted/methods , Magnetic Resonance Imaging/methods , Pattern Recognition, Automated/methods , Phantoms, Imaging , Programming Languages , SoftwareABSTRACT
Multimodality image registration is a critical issue in image-guided cancer ablation techniques. Focal therapies of prostate cancer are usually monitored using ultrasound imaging, while the dose planning is performed on MRI. In this study, a new multimodality images registration and deformation method, based on the Thin Plate Splines -Rigid Point Matching (TPS-RPM) algorithm, is introduced. The Method combines non-rigid mapping and interpolation to deform the images. Preliminary results obtained on phantom and clinical images showed that the registration is accurate and robust against landmarks initialization.
Subject(s)
Algorithms , Magnetic Resonance Imaging/methods , Pattern Recognition, Automated/methods , Prostatic Neoplasms/therapy , Subtraction Technique , Therapy, Computer-Assisted/methods , Ultrasonography/methods , Humans , Image Enhancement/methods , Image Interpretation, Computer-Assisted/methods , Male , Pilot Projects , Reproducibility of Results , Sensitivity and SpecificityABSTRACT
INTRODUCTION: Many algorithms exist for 3D reconstruction of data from freehand 2D ultrasound slices. These methods are based on interpolation techniques to fill the voxels from the pixels. For quantification purposes, segmentation is involved to delineate the structure of interest. However, speckle and partial volume effect errors can affect quantification. OBJECTIVE: This study aimed to assess the effect of the combination of a fuzzy model and 3D reconstruction algorithms of freehand ultrasound images on these errors. METHODS: We introduced a fuzzification step to correct the initial segmentation, by weighting the pixels by a distribution function, taking into account the local gray levels, the orientation of the local gradient, and the local contrast-to-noise ratio. We then used two of the most wide-spread reconstruction algorithms (pixel nearest neighbour (PNN) and voxel nearest neighbour (VNN)) to interpolate and create the volume of the structure. Finally, defuzzification was used to estimate the optimal volume. VALIDATION: B-scans were acquired using 5 MHz and 8 MHz ultrasound probes on ultrasound tissue-mimicking phantoms. Quantitative evaluation of the reconstructed structures was done by comparing the method output to the real volumes. Comparison was also done with classical PNN and VNN algorithms. RESULTS: With the fuzzy model quantification errors were less than 4.3%, whereas with classical algorithms, errors were larger (10.3% using PNN, 17.2% using VNN). Furthermore, for very small structures (0.5 cm(3)), errors reached 24.3% using the classical VNN algorithm, while they were about 9.6% with the fuzzy VNN model. CONCLUSION: These experiments prove that the fuzzy model allows volumes to be determined with better accuracy and reproducibility, especially for small structures (<3 cm(3)).
Subject(s)
Algorithms , Artificial Intelligence , Fuzzy Logic , Image Interpretation, Computer-Assisted/methods , Imaging, Three-Dimensional/methods , Pattern Recognition, Automated/methods , Ultrasonography/methods , Image Enhancement/methods , Neural Networks, Computer , Reproducibility of Results , Sensitivity and SpecificityABSTRACT
Accurate localization and contouring of prostate are important issues in prostate cancer diagnosis and/or therapies. Although several semi-automatic and automatic segmentation methods have been proposed, manual expert correction remains necessary. Our paper introduces an original method for automatic 3D segmentation of the prostate gland from Magnetic Resonance Imaging data. We use a statistical shape model as a priori knowledge, and we model gray levels distribution by fitting histogram modes with a Gaussian mixture. Markov fields are used to introduce contextual information regarding voxels neighbourhood. Final labelling optimization is based on Bayesian a posteriori classification, estimated with the Iterative Conditional Mode algorithm (ICM). We compared the accuracy of this method, free from any manual correction, with contours outlined by an expert radiologist. In 6 random cases, including prostates with cancer and benign prostatic hypertrophy (BPH), mean Hausdorff distance (HD) and Overlap Ratio (OR) were 9.94 mm and 0.83, respectively. Beyond fast computing times, this new method showed satisfying results, even at prostate's base and apex.
Subject(s)
Imaging, Three-Dimensional/methods , Magnetic Resonance Imaging/methods , Prostatic Neoplasms/pathology , Radiotherapy Planning, Computer-Assisted/methods , Algorithms , Humans , Male , Markov Chains , Normal Distribution , Pelvis/anatomy & histology , Probability , Prostate/anatomy & histology , Prostatic Neoplasms/radiotherapy , Software , Technology, Radiologic/methods , Tumor BurdenABSTRACT
A new automatic method for tumor volume segmentation on PET images has been developed. The method introduced in this paper is based on previous works in MRA segmentation and involves both fuzzy set and possibility theories. Visual results prove the method efficiency which is confirmed by obtained Jaccard index.
Subject(s)
Diagnosis, Computer-Assisted/methods , Image Enhancement/methods , Neoplasms/diagnosis , Neoplasms/pathology , Positron-Emission Tomography/methods , Algorithms , Automation , Carcinoma, Hepatocellular/pathology , Fuzzy Logic , Humans , Image Processing, Computer-Assisted , Imaging, Three-Dimensional , Liver Neoplasms/pathology , Models, Theoretical , Probability , Reproducibility of ResultsABSTRACT
It is thought that incisor protrusion could be ethnic-related because Tunisians, as Africans, might reflect the genetic influence of the incisal protrusion of their darker skinned neighbors. We first compiled local standards by determining the normal incisor positions of children of the Sfax region with balanced facial patterns. We selected 95 teenagers from 5 different high schools in Sfax (average age: 16 years 9 months), all having a normal dentition and a balanced facial pattern. This first study showed that upper and lower incisors were protruded 3 mm beyond American standards. Second, we studied incisor stability after four-first bicuspid extraction treatment. We selected 26 patients all having Class I molar relationships and upper and lower incisor protrusion. We compared x-ray incisor measurements of 3 radiographs: before treatment, after treatment and 2 or 3 years post-treatment. This study shows that incisor relapse ranges between 20 and 30%.
Subject(s)
Malocclusion/ethnology , Malocclusion/therapy , Adolescent , Adult , Cephalometry , Child , Female , Humans , Incisor/growth & development , Male , Orthodontics, Corrective , Recurrence , TunisiaABSTRACT
INTRODUCTION: Keratoacanthoma centrifugum marginatum is a rare variety of commonly isolated keratoacanthoma. The size of the lesions and prolonged evolution often raise therapeutic problems. CASE REPORT: A 63-year-old man presented with multiple keratoacanthoma centrifugum marginatum evolving for 6 months and predominating on the lower limbs. Because of the multiplicity and size of the lesions, the patient was treated at acitretine at the dose of 1 mg/kg/day for five months, leading to the regression of all the lesions. DISCUSSION: Keratoacanthoma centrifugum marginatum has rarely been reported in the literature. Our observation remains exceptional in the multiplicity and size of the lesions. The efficacy of retinoids, previously reported in the treatment of solitary giant keratoacanthoma, multiple keratoacanthoma and keratoacanthoma centrifugum marginatum, was demonstrated even in this handicapping form of keratoacanthoma.
Subject(s)
Keratoacanthoma/pathology , Leg Dermatoses/pathology , Humans , Male , Middle AgedABSTRACT
From an analysis of 1,546 cases of cutaneous leishmaniasis collected throughout Tunisia three clinico-epidemiological forms could be determined. These forms are due to three different species of Leishmania and are observed in different geographical areas. Sporadic cutaneous leishmaniasis (SCL), 63 cases. This form is found in Northern Tunisia, in known foci of Kala-azar. 96% of the patients have a single, small, ulcerated and crusty lesion on the face surrounded by an important zone of infiltration. Amastigotes are less than 4 microns in diameter, and it is very difficult to maintain the parasites in cultures. This form is probably due to Leishmania infantum, as has been demonstrated in Algeria in a similar form evolving in the same biotope as the Tunisian form. The vector and reservoirs of SCL are unknown. Zoonotic cutaneous leishmaniasis (ZCL), 1,412 cases. The disease is epidemic in Central and Southern Tunisia. The lesions are multiple; they affect the limbs more frequently than the face and heal in less than 8 months. They are polymorphous, usually large and superinfected. There is a seasonal occurrence of the outbreaks (summer and autumn). Amastigotes are large (4 to 6.5 microns in diameter) and the parasites easily grow in cultures. This form is caused by Leishmania major, and its vector is Phlebotomus papatasi. Three species of rodents harbour the parasite. Anthroponotic cutaneous leishmaniasis (ACL), 71 cases. This form is endemic in South-Eastern Tunisia. 70% of the patients present with single lesions, one half of which affect the face. Dry and proliferating lesions are the most frequent.(ABSTRACT TRUNCATED AT 250 WORDS)