Supporting content-based image retrieval and computer-aided diagnosis systems with association rule-based techniques

In this work, we take advantage of association rule mining to support two types of medical systems: the Content-based Image Retrieval (CBIR) systems and the Computer-Aided Diagnosis (CAD) systems. For content-based retrieval, association rules are employed to reduce the dimensionality of the feature vectors that represent the images and to improve the precision of the similarity queries. We refer to the association rule-based method to improve CBIR systems proposed here as Feature selection through Association Rules (FAR). To improve CAD systems, we propose the Image Diagnosis Enhancement through Association rules (IDEA) method. Association rules are employed to suggest a second opinion to the radiologist or a preliminary diagnosis of a new image. A second opinion automatically obtained can either accelerate the process of diagnosing or to strengthen a hypothesis, increasing the probability of a prescribed treatment be successful. Two new algorithms are proposed to support the IDEA method: to pre-process low-level features and to propose a preliminary diagnosis based on association rules. We performed several experiments to validate the proposed methods. The results indicate that association rules can be successfully applied to improve CBIR and CAD systems, empowering the arsenal of techniques to support medical image analysis in medical systems. (C) 2009 Elsevier B.V. All rights reserved.

Whole-Brain Voxel-Based Morphometry in Kallmann Syndrome Associated with Mirror Movements

BACKGROUND AND PURPOSE: There are 2 main hypotheses concerning the cause of mirror movements (MM) in Kallmann syndrome (KS): abnormal development of the primary motor system, involving the ipsilateral corticospinal tract, and lack of contralateral motor cortex inhibitory mechanisms, mainly through the corpus callosum. The purpose of our study was to determine white and gray matter volume changes in a KS population by using optimized voxel-based morphometry (VBM) and to investigate the relationship between the abnormalities and the presence of MM, addressing the 2 mentioned hypotheses. MATERIALS AND METHODS: T1-weighted volumetric images from 21 patients with KS and 16 matched control subjects were analyzed with optimized VBM. Images were segmented and spatially normalized, and these deformation parameters were then applied to the original images before the second segmentation. Patients were divided into groups with and without MM, and a t test statistic was then applied on a voxel-by-voxel basis between the groups and controls to evaluate significant differences. RESULTS: When considering our hypothesis a priori, we found that 2 areas of increased gray matter volume, in the left primary motor and sensorimotor cortex, were demonstrated only in patients with MM, when compared with healthy controls. Regarding white matter alterations, no areas of altered volume involving the corpus callosum or the projection of the corticospinal tract were demonstrated. CONCLUSION: The VBM study did not show significant white matter changes in patients with KS but showed gray matter alterations in keeping with a hypertrophic response to a deficient pyramidal decussation in patients with MM. In addition, gray matter alterations were observed in patients without MM, which can represent more complex mechanisms determining the presence or absence of this symptom.

Magnetic Resonance Imaging Assessment of Subchondral Bone and Soft Tissues in Knee Osteoarthritis

Knee osteoarthritis (OA) has to be considered a whole joint disease. Magnetic resonance imaging (MRI) allows superior assessment of all joint tissues that may be involved in OA, such as the subchondral bone, synovium, ligaments, and periarticular soft tissues. Reliable MRI-based scoring systems are available to assess and quantify these structures and associated pathology. Cross-sectional and longitudinal evaluation has enabled practitioners to understand their relevance in explaining pain and structural progression.

Perinatal outcome of fetal atrioventricular block - One-hundred-sixteen cases from a single institution

Background-Fetal atrioventricular (AV) block is an uncommon lesion with significant mortality. Because of the rarity of this disorder, the natural course, extensive evaluation of untreated fetuses, and late follow-up remain unclear. Methods and Results-Of the 116 consecutive cases of fetal AV block studied from 1988 to 2006, only 1 was terminated, and 75% were live births. Fifty-nine cases of AV block were associated with major structural heart disease, mainly left atrial isomerism (n = 40), with only 26% of neonatal survivors. Of the 57 fetuses with normal cardiac anatomy, 41 (72%) were positive for maternal antinuclear antibodies, and 32 of these seropositive mothers did not receive any treatment. This untreated group had live-birth and 1-year infant survival rates of 93% and 90%, respectively. Five fetuses from seronegative mothers showed regression to sinus rhythm during pregnancy. The presence of major structural heart disease, hydrops, an atrial rate <= 120 bpm, and a ventricular rate <= 55 bpm were identified as risk factors for mortality. Logistic regression analysis of the whole group showed that the presence of structural heart disease was the only independent predictor of death (P < 0.001). Conclusions-This long-term study confirms that fetal AV block has a poor outcome when associated with structural heart disease and that spontaneous regression of AV block is possible in seronegative forms. The survival rate of >90% of our untreated patients with isolated forms of AV block raises concerns about any decision to intervene with immunosuppressive agents.

Assessment of synovitis with contrast-enhanced MRI using a whole-joint semiquantitative scoring system in people with, or at high risk of, knee osteoarthritis: the MOST study

Objectives To introduce a comprehensive and reliable scoring system for the assessment of whole-knee joint synovitis based on contrast-enhanced (CE) MRI. Methods Multicenter Osteoarthritis Study (MOST) is a cohort study of people with, or at high risk of, knee osteoarthritis (OA). Subjects are an unselected subset of MOST who volunteered for CE-MRI. Synovitis was assessed at 11 sites of the joint. Synovial thickness was scored semiquantitatively: grade 0 (< 2 mm), grade 1 (2-4 mm) and grade 2 (> 4 mm) at each site. Two musculoskeletal radiologists performed the readings and inter-and intrareader reliability was evaluated. Whole-knee synovitis was assessed by summing the scores from all sites. The association of Western Ontario and McMaster Osteoarthritis Index pain score with this summed score and with the maximum synovitis grade for each site was assessed. Results 400 subjects were included (mean age 58.8 +/- 7.0 years, body mass index 29.5 +/- 4.9 kg/m(2), 46% women). For individual sites, intrareader reliability (weighted kappa) was 0.67-1.00 for reader 1 and 0.60-1.00 for reader 2. Inter-reader agreement (kappa) was 0.67-0.92. For the summed synovitis scores, intrareader reliability (intraclass correlation coefficient (ICC)) was 0.98 and 0.96 for each reader and inter-reader agreement (ICC) was 0.94. Moderate to severe synovitis in the parapatellar subregion was associated with the higher maximum pain score (adjusted OR (95% CI), 2.8 (1.4 to 5.4) and 3.1 (1.2 to 7.9), respectively). Conclusions A comprehensive semiquantitative scoring system for the assessment of whole-knee synovitis is proposed. It is reliable and identifies knees with pain, and thus is a potentially powerful tool for synovitis assessment in epidemiological OA studies.