Radiomic Quantification for MRI Assessment of Sacroiliac Joints of Patients with Spondyloarthritis.

Spondyloarthritis (SpA) is a group of diseases primarily involving chronic inflammation of the spine and peripheral joints, as evaluated by magnetic resonance imaging (MRI). Considering the complexity of SpA, we performed a retrospective study to discover quantitative/radiomic MRI-based features correlated with SpA. We also investigated different fat-suppression MRI techniques to develop detection models for inflammatory sacroiliitis. Finally, these model results were compared with those of experienced musculoskeletal radiologists, and the concordance level was evaluated. Examinations of 46 consecutive patients were obtained using SPAIR (spectral attenuated inversion recovery) and STIR (short tau inversion recovery) MRI sequences. Musculoskeletal radiologists manually segmented the sacroiliac joints for further extraction of 230 MRI features from gray-level histogram/matrices and wavelet filters. These features were associated with sacroiliitis, SpA, and the current biomarkers of ESR (erythrocyte sedimentation rate), CRP (C-reactive protein), BASDAI (Bath Ankylosing Spondylitis Activity Index), BASFI (Bath Ankylosing Spondylitis Functional Index), and MASES (Maastricht Ankylosing Spondylitis Enthesis Score). The Mann-Whitney U test showed that the radiomic markers from both MRI sequences were associated with active sacroiliitis and with SpA and its axial and peripheral subtypes (p < 0.05). Spearman's coefficient also identified a correlation between MRI markers and data from clinical practice (p < 0.05). Fat-suppression MRI models yielded performances that were statistically equivalent to those of specialists and presented strong concordance in identifying inflammatory sacroiliitis. SPAIR and STIR acquisition protocols showed potential for the evaluation of sacroiliac joints and the composition of a radiomic model to support the clinical assessment of SpA.

Machine learning techniques for computer-aided classification of active inflammatory sacroiliitis in magnetic resonance imaging.

BACKGROUND: Currently, magnetic resonance imaging (MRI) is used to evaluate active inflammatory sacroiliitis related to axial spondyloarthritis (axSpA). The qualitative and semiquantitative diagnosis performed by expert radiologists and rheumatologists remains subject to significant intrapersonal and interpersonal variation. This encouraged us to use machine-learning methods for this task. METHODS: In this retrospective study including 56 sacroiliac joint MRI exams, 24 patients had positive and 32 had negative findings for inflammatory sacroiliitis according to the ASAS group criteria. The dataset was randomly split with ~ 80% (46 samples, 20 positive and 26 negative) as training and ~ 20% as external test (10 samples, 4 positive and 6 negative). After manual segmentation of the images by a musculoskeletal radiologist, multiple features were extracted. The classifiers used were the Support Vector Machine, the Multilayer Perceptron (MLP), and the Instance-Based Algorithm, combined with the Relief and Wrapper methods for feature selection. RESULTS: Based on 10-fold cross-validation using the training dataset, the MLP classifier obtained the best performance with sensitivity = 100%, specificity = 95.6% and accuracy = 84.7%, using 6 features selected by the Wrapper method. Using the test dataset (external validation) the same MLP classifier obtained sensitivity = 100%, specificity = 66.7% and accuracy = 80%. CONCLUSIONS: Our results show the potential of machine learning methods to identify SIJ subchondral bone marrow edema in axSpA patients and are promising to aid in the detection of active inflammatory sacroiliitis on MRI STIR sequences. Multilayer Perceptron (MLP) achieved the best results.

Machine learning techniques for computer-aided classification of active inflammatory sacroiliitis in magnetic resonance imaging

Abstract Background: Currently, magnetic resonance imaging (MRI) is used to evaluate active inflammatory sacroiliitis related to axial spondyloarthritis (axSpA). The qualitative and semiquantitative diagnosis performed by expert radiologists and rheumatologists remains subject to significant intrapersonal and interpersonal variation. This encouraged us to use machine-learning methods for this task. Methods: In this retrospective study including 56 sacroiliac joint MRI exams, 24 patients had positive and 32 had negative findings for inflammatory sacroiliitis according to the ASAS group criteria. The dataset was randomly split with ∼ 80% (46 samples, 20 positive and 26 negative) as training and ∼ 20% as external test (10 samples, 4 positive and 6 negative). After manual segmentation of the images by a musculoskeletal radiologist, multiple features were extracted. The classifiers used were the Support Vector Machine, the Multilayer Perceptron (MLP), and the Instance-Based Algorithm, combined with the Relief and Wrapper methods for feature selection. Results: Based on 10-fold cross-validation using the training dataset, the MLP classifier obtained the best performance with sensitivity = 100%, specificity = 95.6% and accuracy = 84.7%, using 6 features selected by the Wrapper method. Using the test dataset (external validation) the same MLP classifier obtained sensitivity = 100%, specificity = 66.7% and accuracy = 80%. Conclusions: Our results show the potential of machine learning methods to identify SIJ subchondral bone marrow edema in axSpA patients and are promising to aid in the detection of active inflammatory sacroiliitis on MRI STIR sequences. Multilayer Perceptron (MLP) achieved the best results.(AU)

Shape, texture and statistical features for classification of benign and malignant vertebral compression fractures in magnetic resonance images.

PURPOSE: Vertebral compression fractures (VCFs) result in partial collapse of vertebral bodies. They usually are nontraumatic or occur with low-energy trauma in the elderly secondary to different etiologies, such as insufficiency fractures of bone fragility in osteoporosis (benign fractures) or vertebral metastasis (malignant fractures). Our study aims to classify VCFs in T1-weighted magnetic resonance images (MRI). METHODS: We used the median sagittal planes of lumbar spine MRIs from 63 patients (38 women and 25 men) previously diagnosed with VCFs. The lumbar vertebral bodies were manually segmented and statistical features of gray levels were computed from the histogram. We also extracted texture and shape features to analyze the contours of the vertebral bodies. In total, 102 lumbar VCFs (53 benign and 49 malignant) and 89 normal lumbar vertebral bodies were analyzed. The k-nearest-neighbor method, a neural network with radial basis functions, and a naïve Bayes classifier were used with feature selection. We compared the classification obtained by these classifiers with the final diagnosis of each case, including biopsy for the malignant fractures and clinical and laboratory follow up for the benign fractures. RESULTS: The results obtained show an area under the receiver operating characteristic curve of 0.97 in distinguishing between normal and fractured vertebral bodies, and 0.92 in discriminating between benign and malignant fractures. CONCLUSIONS: The proposed classification methods based on shape, texture, and statistical features have provided high accuracy and may assist in the diagnosis of VCFs.

Automatic detection of the nipple in screen-film and full-field digital mammograms using a novel Hessian-based method.

Automatic detection of the nipple in mammograms is an important step in computerized systems that combine multiview information for accurate detection and diagnosis of breast cancer. Locating the nipple is a difficult task owing to variations in image quality, presence of noise, and distortion and displacement of the breast tissue due to compression. In this work, we propose a novel Hessian-based method to locate automatically the nipple in screen-film and full-field digital mammograms (FFDMs). The method includes detection of a plausible nipple/retroareolar area in a mammogram using geometrical constraints, analysis of the gradient vector field by mean and Gaussian curvature measurements, and local shape-based conditions. The proposed procedure was tested on 566 mammographic images consisting of 372 randomly selected scanned films from two public databases (mini-MIAS and DDSM), and 194 digital mammograms acquired with a GE Senographe 2000D FFDM system. A radiologist independently marked the centers of the nipples for evaluation of the results. The average error obtained was 6.7 mm (22 pixels) with reference to the center of the nipple as identified by the radiologist. Only two out of the 566 detected nipples (0.35 %) had an error larger than 50 mm. The method was also directly compared with two other techniques for the detection of the nipple. The results indicate that the proposed method outperforms other algorithms presented in the literature and can be used to identify accurately the nipple on various types of mammographic images.

Content-based retrieval of mammograms using visual features related to breast density patterns.

This paper describes part of content-based image retrieval (CBIR) system that has been developed for mammograms. Details are presented of methods implemented to derive measures of similarity based upon structural characteristics and distributions of density of the fibroglandular tissue, as well as the anatomical size and shape of the breast region as seen on the mammogram. Well-known features related to shape, size, and texture (statistics of the gray-level histogram, Haralick's texture features, and moment-based features) were applied, as well as less-explored features based in the Radon domain and granulometric measures. The Kohonen self-organizing map (SOM) neural network was used to perform the retrieval operation. Performance evaluation was done using precision and recall curves obtained from comparison between the query and retrieved images. The proposed methodology was tested with 1,080 mammograms, including craniocaudal and mediolateral-oblique views. Precision rates obtained are in the range from 79% to 83% considering the total image set. Considering the first 50% of the retrieved mages, the precision rates are in the range from 78% to 83%; the rates are in the range from 79% to 86% considering the first 25% of the retrieved images. Results obtained indicate the potential of the implemented methodology to serve as a part of a CBIR system for mammography.