Magnetic Resonance Classification Research Articles

Recent Advances in Classification of Brain Tumor from MR Images - State of the Art Review from 2017 to 2021.

The task of identifying a tumor in the brain is a complex problem that requires sophisticated skills and inference mechanisms to accurately locate the tumor region. The complex nature of the brain tissue makes the problem of locating, segmenting, and ultimately classifying Magnetic Resonance (MR) images a complex problem. The aim of this review paper is to consolidate the details of the most relevant and recent approaches proposed in this domain for the binary and multi-class classification of brain tumors using brain MR images. In this review paper, a detailed summary of the latest techniques used for brain MR image feature extraction and classification is presented. A lot of research papers have been published recently with various techniques proposed for identifying an efficient method for the correct recognition and diagnosis of brain MR images. The review paper allows researchers in the field to familiarize themselves with the latest developments and be able to propose novel techniques that have not yet been explored in this research domain. In addition, the review paper will facilitate researchers who are new to machine learning algorithms for brain tumor recognition to understand the basics of the field and pave the way for them to be able to contribute to this vital field of medical research. In this paper, the review is performed for all recently proposed methods for both feature extraction and classification. It also identifies the combination of feature extraction methods and classification methods that, when combined, would be the most efficient technique for the recognition and diagnosis of brain tumor from MR images. In addition, the paper presents the performance metrics, particularly the recognition accuracy, of selected research published between 2017-2021.

Lumbar Intervertebral Disc Degeneration as a Common Incidental Finding in Young Pregnant Women as Observed on Prenatal Magnetic Resonance Imaging.

Background: Obstetric imaging, subserving fetal evaluation, may yield incidental maternal findings. Based on prenatal magnetic resonance (MR) imaging, this study aims to investigate incidental intervertebral disc degeneration and displacement in young, pregnant women. Methods: This retrospective study included the sagittal 1.5 Tesla, T2-weighted lumbar spine images of 943 pregnant Central Europeans (age range, 18-47 years), who initially had undergone MR imaging because of sonographically suspected fetal abnormalities. Qualitatively, 4715 lumbar intervertebral discs were evaluated for degeneration using a modified Pfirrmann MR classification (nondegenerated, low-grade, moderate, and high-grade degeneration), as well as for displacement. In addition to descriptive statistics, an ordinal regression analysis was performed to analyze the relationship between degeneration and the women's age, and body weight. Results: With regard to the highest degree of degeneration in each woman, 578 (61.3%) showed low-grade, 211 (22.4%) moderate, and 154 (16.3%) high-grade degeneration, and no woman had entirely nondegenerated discs. For the span from 18 to 47 years of age, moderate and high-grade degeneration increased from 6.7% to 36.7% and from 13.3% to 22.4%, respectively. Of 943 women, 57 (6%) had disc displacements, of which 97% were in conjunction with high-grade degeneration. There was a statistically significant relationship (p < 0.001) between degeneration and age, and between degeneration and body weight. Conclusions: In young pregnant women, lumbar intervertebral disc degeneration is a ubiquitous, incidental finding, increasing from the late second decade of life onward, which may be part of physiological aging, as opposed to a small percentage of incidental disc displacements.

Application of fast curvelet Tsallis entropy and kernel random vector functional link network for automated detection of multiclass brain abnormalities

Binary classification of brain magnetic resonance (MR) images has made remarkable progress and many automated systems have been developed in the last decade. Multiclass classification of brain MR images is comparatively more challenging and has great clinical significance. Hence, it has recently become an active area of research in biomedical image processing. In this paper, an automated multiclass brain MR classification framework is proposed to categorize the MR images into five classes such as brain stroke, degenerative disease, infectious disease, brain tumor, and normal brain. A texture based feature descriptor is proposed using curvelet transform and Tsallis entropy to extract salient features from MR images. The potential of Tsallis entropy features is compared with Shannon entropy features. A kernel extension of random vector functional link network (KRVFL) is used to perform multiclass classification and improve the generalization performance at faster training speed. To validate the proposed method, two standard multiclass brain MR datasets (MD-1 and MD-2) are used. The proposed system obtained classification accuracies of 97.33% and 94.00% for MD-1 and MD-2 datasets respectively using 5-fold cross validation approach. The experimental results demonstrated the effectiveness of our system compared to the state-of-the-art schemes and hence, can be utilized as a supportive tool by physicians to verify their screening.

Effects of anti-osteoporosis medications on radiological and clinical results after acute osteoporotic spinal fractures: a retrospective analysis of prospectively designed study.

Although anti-osteoporosis medications have beneficial effects on prevention of osteoporotic spinal fractures (OSFs), few studies have compared effects of medications on fracture healing following OSFs. Therefore, the purpose of this study was to elucidate the effects of different anti-osteoporosis medications on radiological and clinical outcomes after acute OSFs. A total of 132 patients diagnosed with acute OSFs were enrolled and allocated into three groups [group I (n = 39, no anti-osteoporosis medication), group II (n = 66, bisphosphonate), and group III (n = 27, parathyroid hormone (PTH)]. Radiological parameters including magnetic resonance (MR) classification, occurrence of intravertebral cleft (IVC), and clinical outcomes such as numerical rating scale (NRS) and Oswestry disability index were assessed. Risk analyses for IVC and progressive collapse were done along the related factors and medication type. IVC sign was observed in 30 patients. The rate of IVC sign was lower in group III (7.4%) than that in group I (20.5%) or group II (30.3%), although the difference was not statistically significant. Moreover, the degree of NRS improvement was better in group III than that in group I or group II (5.7 vs. 3.1 vs. 3.5, p < 0.001). On multiple regression analysis, mid-portion type fracture in MR classification was a significant risk factor for progressive OSFs. The use of PTH showed significant lower incidences of occurrence of IVC (odds ratio (OR) = 0.160) and increase in height loss (OR = 0.325). Different anti-osteoporosis medications presented different clinical and radiological results after acute OSFs. The use of anabolic agent significantly enhanced fracture healing, reduced progressive collapse, and presented better clinical outcomes. Proper selection of medication might improve initial management of acute OSFs.

MR Classification System Based on Axial Images for Cervical Compressive Myelopathy.

To propose a new magnetic resonance (MR) classification system based on axial images for cervical compressive myelopathy (CCM) (Ax-CCM system), to assess the interobserver agreement with the Ax-CCM system, and to evaluate the relationship between Ax-CCM patterns and the baseline severity of CCM and the subsequent surgical outcome. The institutional review board approved this retrospective study. A total of 202 patients (male-to-female ratio, 128:84; mean age ± standard deviation, 56.3 years ± 11.9; age range, 24-83 years) with CCM who underwent preoperative MR imaging and decompression surgery were retrospectively evaluated. The Ax-CCM pattern was based on the margin and extent of intramedullary hyperintensity on axial T2-weighted images, as follows; type 0 = normal, type 1 = diffuse, type 2 = fuzzy focal, and type 3 = discrete focal. Interobserver variability was analyzed by using the intraclass correlation coefficient across three readers. The modified Japanese Orthopedic Association (JOA) score and the postoperative improvement (good vs little improvement) were evaluated according to the Ax-CCM pattern by using one-way analysis of variance, the χ(2) test, and the Fisher exact test. The intraclass correlation coefficient for the Ax-CCM system was 0.83. The preoperative JOA score was significantly different according to Ax-CCM pattern across all readers (P < .05), with the type 2 pattern showing the worst preoperative JOA score (mean, 11.6 ± 3.1 for readers A and C and 11.7 ± 2.9 for reader B). The proportion of good improvement was significantly lower with the type 2 pattern (27 of 72 patients, 37%) than with the other patterns (64 of 123 patients, 52%) (P = .034). The Ax-CCM system showed good interobserver agreement, and the type 2 pattern was correlated with poor preoperative neurologic status and less postoperative improvement.

Can magnetic resonance imaging accurately predict concordant pain provocation during provocative disc injection?

To correlate magnetic resonance (MR) image findings with pain response by provocation discography in patients with discogenic low back pain, with an emphasis on the combination analysis of a high intensity zone (HIZ) and disc contour abnormalities. Sixty-two patients (aged 17-68 years) with axial low back pain that was likely to be disc related underwent lumbar discography (178 discs tested). The MR images were evaluated for disc degeneration, disc contour abnormalities, HIZ, and endplate abnormalities. Based on the combination of an HIZ and disc contour abnormalities, four classes were determined: (1) normal or bulging disc without HIZ; (2) normal or bulging disc with HIZ; (3) disc protrusion without HIZ; (4) disc protrusion with HIZ. These MR image findings and a new combined MR classification were analyzed in the base of concordant pain determined by discography. Disc protrusion with HIZ [sensitivity 45.5%; specificity 97.8%; positive predictive value (PPV), 87.0%] correlated significantly with concordant pain provocation (P < 0.01). A normal or bulging disc with HIZ was not associated with reproduction of pain. Disc degeneration (sensitivity 95.4%; specificity 38.8%; PPV 33.9%), disc protrusion (sensitivity 68.2%; specificity 80.6%; PPV 53.6%), and HIZ (sensitivity 56.8%; specificity 83.6%; PPV 53.2%) were not helpful in the identification of a disc with concordant pain. The proposed MR classification is useful to predict a disc with concordant pain. Disc protrusion with HIZ on MR imaging predicted positive discography in patients with discogenic low back pain.

MR classification of renal masses with pathologic correlation

To perform a feature analysis of malignant renal tumors evaluated with magnetic resonance (MR) imaging and to investigate the correlation between MR imaging features and histopathological findings. MR examinations in 79 malignant renal masses were retrospectively evaluated, and a feature analysis was performed. Each renal mass was assigned to one of eight categories from a proposed MRI classification system. The sensitivity and specificity of the MRI classification system to predict the histologic subtype and nuclear grade was calculated. Subvoxel fat on chemical shift imaging correlated to clear cell type (p < 0.05); sensitivity = 42%, specificity = 100%. Large size, intratumoral necrosis, retroperitoneal vascular collaterals, and renal vein thrombosis predicted high-grade clear cell type (p < 0.05). Small size, peripheral location, low intratumoral SI on T2-weighted images, and low-level enhancement were associated with low-grade papillary carcinomas (p < 0.05). The sensitivity and specificity of the MRI classification system for diagnosing low grade clear cell, high-grade clear cell, all clear cell, all papillary, and transitional carcinomas were 50% and 94%, 93% and 75%, 92% and 83%, 80% and 94%, and 100% and 99%, respectively. The MRI feature analysis and proposed classification system help predict the histological type and nuclear grade of renal masses.

Primary amenorrhea: evaluation with MR imaging.

To determine the accuracy of magnetic resonance (MR) imaging in the evaluation of patients with a history of primary amenorrhea. Twenty-nine patients with primary amenorrhea underwent T2-weighted MR imaging in the axial and sagittal planes. Contrast material-enhanced and unenhanced T1-weighted MR imaging was performed in selected cases. MR imaging findings were correlated with surgical findings in 23 patients and with clinical findings in six patients. Surgical correlation was available in 14 of 17 patients with female anatomic anomalies (Mayer-Rokitansky-Küster-Hauser syndrome [n = 9], transverse vaginal septum [n = 3], imperforate hymen [n = 1], cervical agenesis [n = 1]), which confirmed MR imaging findings. In nine of 12 patients with congenital disorders of sexual differentiation (testicular feminization [n = 5], true hermaphrodite [n = 2], gonadal dysgenesis [n = 2]) who underwent surgical correlation, 13 of 16 (81%) gonads were correctly localized with MR imaging. Overall, there was excellent correlation between the classification of patients with MR imaging versus classification with the combination of surgical and laboratory findings (K = 0.88). MR imaging is useful in the work-up of patients who present with primary amenorrhea both for accurate diagnosis of pathologic conditions and for surgical planning.