Metal artifact reduction and image quality evaluation of lumbar spine CT images using metal sinogram segmentation.

Metal artifacts often appear in the images of computed tomography (CT) imaging. In the case of lumbar spine CT images, artifacts disturb the images of critical organs. These artifacts can affect the diagnosis, treatment, and follow up care of the patient. One approach to metal artifact reduction is the sinogram completion method. A mixed-variable thresholding (MixVT) technique to identify the suitable metal sinogram is proposed. This technique consists of four steps: 1) identify the metal objects in the image by using k-mean clustering with the soft cluster assignment, 2) transform the image by separating it into two sinograms, one of which is the sinogram of the metal object, with the surrounding tissue shown in the second sinogram. The boundary of the metal sinogram is then found by the MixVT technique, 3) estimate the new value of the missing data in the metal sinogram by linear interpolation from the surrounding tissue sinogram, 4) reconstruct a modified sinogram by using filtered back-projection and complete the image by adding back the image of the metal object into the reconstructed image to form the complete image. The quantitative and clinical image quality evaluation of our proposed technique demonstrated a significant improvement in image clarity and detail, which enhances the effectiveness of diagnosis and treatment.

Computed tomography evaluation of intracranial vascular calcification in major ischemic stroke patients (vascular territory)--its distribution and association with vascular risk factors: a retrospective trial.

OBJECTIVE: Our objective was to determine the distribution of intracranial atherosclerotic calcification, its association with risk factors, and cerebrovascular events in patients with major ischemic stroke. MATERIAL AND METHOD: In this retrospective study, 327 patients who underwent CT scan of brain were included and the clinical parameters were recorded. Two neuroradiologists evaluated the non-contrast axial CT images for any of intracranial arteries, based on a standard CT scoring system for extent (0-4) and thickness (0-4). The composite CT score for extent and thickness of these vascular segments or vessels were recorded on all patients. RESULTS: Based on of MDCT features, 155 major ischemic stroke and 172 non-ischemic stroke were enrolled The highest prevalence of calcification was seen in intracranial internal carotid artery (IICA) (73%), and less commonly in the vertebral artery (8%). There were higher prevalence of intracranial artery calcification in ischemic stroke patients than non-ischemic stroke patients (82% vs. 52%, p < 0.0001). Hypertension (OR = 1.903, 95% CI: 1.019-3.552, p < 0.05), intracranial artery calcification (OR = 2.147, 95% CI: 1.143-4.033, p < 0.05), moderate degree of calcification (OR = 2.631, 95% CI: 1.299-5.260, p < 0.05), and severe degree of calcification (OR = 3.479, 95% CI: 1.500-8.068, p < 0.05) were found to be independently associated with ischemic stroke. CONCLUSION: Significant intracranial atherosclerosis as determined by severe CT calcification had higher incidence in ischemic stroke patients. Intracranial artery calcification with moderate and severe degree of calcification and hypertension were independently significant associated with ischemic stroke. CT calcification score might serve as an indicator of intracranial atherosclerotic disease and might be useful in predicting ischemic stroke.

Comparison of 18F-FDG Pet/CT and CT: diagnosis performance in lymphoma patient after treatment.

OBJECTIVE: Retrospectively comparing 18F-FDG PET/CT and CT findings at the same anatomic locations in patients with lymphoma by using a combined PET/CT scanner and to analyze the lesions on both metabolic and anatomic bases to evaluate their sensitivity specificity positive predictive value (PPV), negative predictive value (NPV), and accuracy. We analyzed all studies, all patients, common cell type in this study such as diffuse large B cell lymphoma (DLBCL) and Hodgkin's lymphoma and indication of the study such as restaging for recurrence post-therapy and evaluate residual disease within two months after chemotherapy. MATERIAL AND METHOD: Sixty-seven lymphoma patients were studied PET/CT between January 2007 and December 2012 in Siriraj Hospital. We excluded six patients due to no medial report in our hospital. Sixty-one patients (29 male, 32 female, mean age 46.6 +/- 17.7 years, range 8-75) with NHL and with HL) were analyzed for the result of dual-modality PET/CT They underwent 77 18F-FDG PET/CT studies for restaging, for recurrence post-therapy based on 41 studies and evaluation of residual disease within two months after chemotherapy in 36 studies. RESULTS: The statistical parameters of 18F-FDG PET/CT imaging of lymphoma patients after treatment show significantly better specificity than CT and insignificant high accuracy for all studies, all patients, histology of DLBCL, indication of evaluation of active lymphoma within two months after chemotherapy. The 18F-FDG PET/CT parameters of accuracy and PPV are higher than CT without statistical significance. The 18F-FDG PET/CT is not significantly better than CT for histology of Hodgkin's lymphoma and indication of restaging for recurrence post-therapy Nevertheless, the 18F-FDG PET/CT shows slightly improved specificity PPV and accuracy than CT The sensitivity of CT in this study is high and may be from most of our cases selected post-treatment lymphoma that had a residual mass after treatment. Therefore, the sensitivity of PET scan is not significantly higher when compared with CT scan. CONCLUSION: The PET/CT is better than CT for post-treatment lymphoma patient particularly for cell type of DLBCL and indication for evaluation of active lymphoma within two months after chemotherapy.

Lateral cephalometric analysis and the risks of moderate to severe obstructive sleep-disordered breathing in Thai patients.

PURPOSES: This cross-sectional study was aimed to compare lateral cephalometric parameters among patients with different severities of obstructive sleep-disordered breathing and to determine if there are radiographic variables that increase risk of moderate to severe obstructive sleep apnea (OSA). METHODS: Ten linear and 5 angular parameters of lateral cephalometry in 188 adult subjects which included 47 controls and 141 patients with OSA classified by apnea-hypopnea index (AHI) from level-I polysomnography, were analyzed. All radiographs were done under standardized processes and measured twice on separate occasions. RESULTS: There were statistically significant differences between controls and patients with AHI ≥ 15 in parameters such as a distance from mandibular plane to hyoid (MP-H), posterior airway space (PAS), skull base angle (NSBA), a distance from posterior nasal spine to posterior pharyngeal wall (PNS-PP), and soft palate length (PNS-P), (p < 0.01). However, no significant difference was found between controls and mild OSA. Logistic regression analysis showed that only MP-H with cutoff point of ≥ 18 millimeters, NSBA ≤ 130°, and PAS ≤ 10 millimeters were independent cephalometric variables that increased risk of having AHI ≥ 15 compared to controls with adjusted odds ratio of 17.1, 8.3 and 4.2, respectively. Gender did not significantly associate with OSA severity in this analysis. CONCLUSIONS: Patients who had longer MP-H, narrower PAS and narrower NSBA than specific cutoff points significantly increased risks of moderate to severe OSA. Treatments that effectively improve these parameters, particularly at tongue base level (MP-H and PAS), may decrease the severity of disease.

Detection of recurrent colorectal cancer by 18F-FDG PET/CT comparison with contrast enhanced CT scan.

OBJECTIVE: To evaluate diagnostic performance of 18F-FDG PET/CT in assessment of recurrence colorectal cancer after treatment in Siriraj Hospital. MATERIAL AND METHOD: The authors retrospectively studied 48 treated colorectal cancer patients with suspected recurrence who underwent 18F-FDG PET/CT and contrast-enhanced CT (CECT). Clinical information, image follow-up for at least one year, and pathological reports of the patients were reviewed for gold standard RESULTS: Recurrent or metastatic disease was found in 36 of 48 patients. Calculated sensitivity, specificity and accuracy of 18F-FDG PET/CT were 94.4%, 66.7%, and 87.5%. 18F-FDG PET/CT can reduce false positive results of CECT in six patients, thus specificity of 18F-FDG PET/CT was statistically significantly better than that of CECT. Using lesion-based analysis with 65 recurrent sites and 26 non-recurrent lesion, 18F-FDG PET/CT showed better sensitivity 87.7%, specificity 61.5%, and accuracy 80.2 than CECT without statistical significance. CONCLUSION: 18F-FDG PET/CT overall showed higher sensitivity, specificity, and accuracy than CECT.

Reliability and validity of the Canadian neurological scale, Thai version.

BACKGROUND: The Canadian Neurological Scale (CNS) is one of the most reliable stroke severity assessment scales. There is a strong need for a simple and well validated stroke severity assessment scale among Thais. OBJECTIVE: To translate and perform a reliability and validity study of the Canadian Neurological Scale, Thai version (CNS-T). MATERIAL AND METHOD: Forward and backward translations of the original CNS version were independently performed. The final version of the CNS-T was prospectively tested for reliability and validity in acute ischemic stroke setting. Consecutive series of acute stroke patients were assessed by one of the six raters from three different types of healthcare providers: 2 stroke nurses, 2 internal medicine residents and 2 stroke fellows. Each patient was independently assessed twice at 3 weeks interval using video tape by all raters. Extent of infarction was measured by MRI lesion volume. Clinical outcome at 3 months was measured using modified Rankin Score (mRS). Correlation among the CNS-T and 3-mo mRS and MRI lesion volume were assessed. Inter and intra-observer reliabilities were evaluated. RESULTS: A total of 38 patients were enrolled. Median CNS-T was 8.5. Intra-observer reliability demonstrated a high agreement with an intraclass correlation (ICC) of 0.99, 0.97, 0.98, 0.96, 0.93 and 0.98 for 2 stroke fellows, 2 internal medicine residents and 2 stroke nurses respectively. Inter-observer reliability between the 6 raters was excellent: ICC 0.87 (95% CI; 0.81-0.92). The Spearman rank correlation coefficient was -0.55 (p = 0.001) between the initial CNS-T score versus initial MRI lesion volume and -0.61 (p < 0.001) between the initial CNS-T score versus 3-mo mRS. CONCLUSION: The CNS-T can be performed by trained nurses, internists and neurologists with an excellent reliability. The CNS-T is a valid and simple clinical tool for stroke severity assessment among Thais.

Selective ophthalmic arterial infusion of chemotherapeutic drugs for recurrent retinoblastoma.

Introduced in 1988 by Kaneko and colleagues, selective ophthalmic arterial infusion of chemotherapeutic drug has recently gained more interest among retinoblastoma experts worldwide. The report showed that the procedure could be repeated up to 12 treatments without serious side effects. We report a 4-year-old girl with bilateral retinoblastoma. The left eye was enucleated for the group E disease. The right eye started with 3 retinal tumors (group C) was treated with systemic chemotherapy plus local therapy. Seven months after the last cycle of chemotherapy, the tumor recurred close to the fovea. Systemic chemotherapy was reinitiated without success. To avoid aggressive cryotherapy and external-beam radiotherapy, selective ophthalmic arterial infusion of chemotherapeutic drugs was performed for 15 sessions. The tumor responded partially without evidence of drug-induced retinal toxicity by the electroretinogram. Minor irregularities of the inner wall of supraclinoid portion of the internal carotid artery were observed only at the sixth session. Narrowing of the vascular lumen occurred on the last 2 sessions. We demonstrate that this technique when performed repeatedly could result in the anatomic changes of the small blood vessel. Careful follow-up is necessary for early detection of any serious consequences.

Acute disseminated encephalomyelitis in Siriraj Hospital: clinical manifestations and short-term outcome.

OBJECTIVE: To describe clinical manifestations, neuroimaging findings, and clinical outcomes in children with acute disseminated encephalomyelitis (ADEM). MATERIAL AND METHOD: Children with a diagnosis of ADEM who were less than 15 years of age at Siriraj Hospital between January 2002 and December 2008 were retrospectively reviewed. Clinical symptoms and signs as well as cerebrospinal fluid analysis, neuroimaging findings and clinical outcomes were extracted from medical records using a standard form. RESULTS: During the present study period, 14 children were diagnosed with ADEM. Median age was 7.2 years (range, 1.25-13 years). The most common presenting symptoms were decreased mental status (93%), weakness (71%), and fever (50%). Cranial MRI was abnormal in all patients. All but one patient received high dose intravenous methylprednisolone and a course of tapered oral prednisolone. After a mean follow-up period of 28.6 +/- 19.8 months, 13 patients were classified as monophasic ADEM and one progressed to have multiple sclerosis. Eleven patients recovered completely while one was left with mild hemiparesis and the other two (one with final diagnosis of MS) with severe psycho-neurological disturbances. CONCLUSION: There are no specific symptoms and signs in children with ADEM. Multifocal neurological deficits along with encephalopathy and abnormal MRI findings lead to correct diagnosis. Treatment with corticosteroid may improve clinical outcomes. Some children may progress to MS. Long-term clinical and neuroimaging studies in these children are needed.

Vertebral body compression fracture: discriminating benign from malignant causes by diffusion-weighted MR imaging and apparent diffusion coefficient value.

OBJECTIVE: To evaluate the diagnostic performance of apparent diffusion coefficient (ADC) value in discriminating benign from malignant vertebral compression fracture. MATERIAL AND METHOD: 22 symptomatic patients with compression fracture of vertebra referred for conventional MRI spines during January 2009-March 2010 underwent additional diffusion weighted MR techniques. Evaluation of diffusion weighted MR imaging and quantified ADC value from reconstructed ADC map were performed. The accuracy, sensitivity, specificity, positive predictive value and negative predictive value of apparent diffusion coefficient (ADC) value were calculated. RESULTS: A total of 39 vertebral fractures; 7 malignant compression fractures and 32 benign compression fractures were evaluated. The difference between ADC values of malignant, benign compression fracture and normal vertebrae were statistically significant (p < 0.0001). The accuracy, sensitivity and specificity were 89.7%, 85.7% and 90.6% respectively with the ADC threshold of 0.89 to discriminate malignancy. CONCLUSION: The ADC promises to be an effective implement for characterization of vertebral body compression fracture in differentiating benign and malignant compression fractures.

Predictive value of magnetic resonance imaging in cervical spondylotic myelopathy in prognostic surgical outcome.

OBJECTIVE: Establish the predictive value of magnetic resonance imaging (MRI) for cervical spondylotic myelopathy as being a good operative outcome. MATERIAL AND METHOD: A retrospective study of the 52 consecutive patients with cervical spondylotic myelopathy underwent both magnetic resonance imaging (MRI) cervical spines examination at Siriraj Hospital between January 2005 and June 2007. Surgery was divided into two groups: "Good" operative outcome (35 patients) and "No improvement group" (17 patients). Two neuroradiologists independently identified the MR images data that showed the maximum stenosis on sagittal and axial sections and recorded predictive MRL parameters: T2-weighted signal change of the spinal cord, cross-sectional area of the spinal cord, anteroposterior (AP) diameter of the spinal canal and the spinal cord and AP-compression ratio (AP diameter/transverse diameter of the spinal cord). RESULTS: There were no statistically significant differences between both groups in all parameters. CONCLUSION: The AP-diameter of the spinal canal and spinal cord, AP-compression ratio and signal change of the spinal cord are not useful in predicting prognosis outcome in patients with cervical spondylotic myelopathy. In addition, cross-sectional area of the spinal cord cannot confidentially be used as predictive factor in CSM patients due to many influent factors of surgical outcome. A further prospective study without patient selective bias may offer more definite results to confirm these findings.

Accuracy of Support-Vector Machines for Diagnosis of Alzheimer's Disease, Using Volume of Brain Obtained by Structural MRI at Siriraj Hospital.

Background: The determination of brain volumes using visual ratings is associated with an inherently low accuracy for the diagnosis of Alzheimer's disease (AD). A support-vector machine (SVM) is one of the machine learning techniques, which may be utilized as a classifier for various classification problems. This study exploratorily investigated the accuracy of SVM classification models for AD subjects using brain volume and various clinical data as features. Methods: The study was designed as a retrospective chart review. A total of 201 eligible subjects were recruited from the Memory Clinic at Siriraj Hospital, Thailand. Eighteen cases were excluded due to incomplete MRI data. Subjects were randomly assigned to a training group (AD = 46, normal = 46) and testing group (AD = 45, normal = 46) for SVM modeling and validation, respectively. The results in terms of accuracy and a receiver operating characteristic curve analysis are reported. Results: The highest accuracy for brain volumetry (62.64%) was found using the hippocampus as a single feature. A combination of clinical parameters as features provided accuracy ranging between 83 and 90%. However, a combination of brain volumetry and clinical parameters as features to the SVM models did not improve the accuracy of the result. Conclusions: In our study, the use of brain volumetry as SVM features provided low classification accuracy with the highest accuracy of 62.64% using the hippocampus volume alone. In contrast, the use of clinical parameters [Thai mental state examination score, controlled oral word association tests (animals; and letters K, S, and P), learning memory, clock-drawing test, and construction-praxis] as features for SVM models provided good accuracy between 83 and 90%.

Screening may not be accurate word to represent the cases submitted to PET/CT evaluation for primary tumor in a patient who has abnormal serum tumor marker.

OBJECTIVE: Determine the value of PET/CT in unknown primary cancer patient with high tumor marker and negative study for clinical and conventional imaging. MATERIAL AND METHOD: A retrospective database review of 417 patients who received PET/CT between July 2006 and August 2007 in National cyclotron and PET center at Chulabhorn cancer center was done. Patients were included in this study if the diagnosis were unknown primary cancer and rising tumor marker. Twelve patients were included in this study. Data included age, gender, tumor marker rising, anatomical imaging finding (CT and MRI), PET finding and clinical follow-up. RESULTS: Nine cases had normal PET/CT. This showed that PET/CT does not get more information than conventional imaging. The PET scan showed positive in three cases, #5, #6 and #10. Two cases were false positive, #5 and #6. Case #5 had clinical follow-up for one year and revealed to be normal. Case #6 PET showed markedly glucose avid lesion at tumor thrombus but contrast CT confirm blood clot and the patient was treat with wafarin and claxane. The follow-up clinical showed improvement. The high serum CA 125 explained by lung infarction caused the false positive. In case #10, the PET/CT suggested lung cancer at basal segment of LLL. CONCLUSION: Screening 18F FDG PET/CT is not appropriate in unknown primary with rising tumor marker and normal conventional imaging is required.

Degree of midline shift from CT scan predicted outcome in patients with head injuries.

OBJECTIVE: To evaluate the relationship between the degree of midline shift by Computed Tomography (CT) finding and Glasglow Coma Score (GCS) as a predictive of clinical outcome in patients after head injury. METHOD AND METHOD: The present study was performed by retrospectively reviewing 216 consecutive cases of traumatic head injury admitted to the trauma center in Siriraj Hospital from 1999 until 2004. All patients were evaluated for level of consciousness by a neurosurgeon determining by GCS and underwent CT brain for evaluation of intracranial hemorrhage and midline shift. The final clinical outcome was also divided into two groups; good outcome for the patients who recovered well with moderate disability and the poor outcome for the patients who suffered severe disability, vegetative status and death. Then, the authors compared midline shift vs. GCS and midline shift vs. clinical outcomes. RESULTS: Total of 216 cases, the three most common types of head injury were motorcycle accident, fall or assault and car accident. 96 of 216 patients had midline shifting, 53 of 96 patients had CT scan of midline shifting less than 10 mm whereas 37 of 96 patients had a CT scan of greater than 10 mm of midline shifting. 63.3% with midline shifting up to 10 mm had severe head injury and up to 81% with brain shifting greater than 10 mm had severe head injury. The clinical outcome also showed that poor clinical outcomes correlated to midline shifting greater than 10 mm. CONCLUSION: The increased degree of midline shift in patients with head injuries by CT scan was related to the severity of head injury (GCS = 3-12) and was significantly related to poor final clinical outcome.

Computed tomographic findings in non-traumatic hemorrhagic stroke.

OBJECTIVE: To analyze CT findings in hemorrhagic stroke patients correlation with clinical outcome and assess the interobserver agreement of hemorrhagic stroke identification on CT imaging. MATERIAL AND METHOD: CT imaging features of 131 cases and clinical data were verified and collected at Siriraj Hospital from Jan 2004 to Dec 2005 and retrospectively analyzed for type, location, mass effect, size of hemorrhage, intraventricular extension, initial level of consciousness (GCS), hospital length of stay and patient outcome. The percentages, predictive values, kappa were calculated. RESULTS: From all types of hemorrhagic stroke, intracerebral hemorrhage remains a common and devastating clinical problem. The most common site was the thalamus and basal ganglia. In the present study, the authors found that fifty-three cases (53/131 cases, 40.5%) with thalamic-ganglionic hemorrhage, nineteen cases (19/131 cases, 14.5%) in lobar hemorrhage, five cases (5/131 cases, 3.8%) in cerebellum, five cases (5/131 7 cases, 3.8%) in brainstem and eight cases (8/131 cases, 6.1%) occurred in multiple locations. There were twenty-five cases (25/131 cases, 19.1%) of subarachnoid hemorrhage, thirteen cases (13/131 cases, 9.9%) of subdural hemorrhage and three cases (3/131 cases, 2.3%) of intraventricular hemorrhage. Two variables on CT imaging, identified as significant as early mortality predictors, were hematoma volume more than 60 cm3, and presence of intraventricular hemorrhage extension (p < 0.05). The mass effect defined as midline and/or enlargement of contralateral ventricle was not significant (p = 0.067). The present study found concordance between CT brain interpretation by two neuroradiologists for the type of hemorrhagic stroke was very good, Kappa = 0.861 as well as for location was 0.866. CONCLUSION: CT imaging is an imaging instrument for early identification of hemorrhagic stroke patients and providing imaging evidence of high mortality risk.

Evaluation of image quality and lens's radiation dose of a low-dose cranial CT scan.

OBJECTIVE: To determine the lowest miliampere-second (mAs) of a cranial computed tomography (CT) scan that can maintain acceptable image quality on cranial CT scan which might help reducing the risk of cataract formation. MATERIAL AND METHOD: The present study was performed on the 148 patients in routine daily practice sent for diagnosis of intracranial conditions by a cranial CT scan. During the cranial CT scanning, each lens's radiation dose on patient's eyes was measured by a thermoluminescent dosimeter Clinical image quality, particularly in analysis of gray and white matter differentiation, was evaluated independently using a 5 point scale by two radiologists. RESULTS: During standard cranial CT scan protocol with 250 miliampere-second (mAs), the total dose of left and right lens's dose were about 50.93 miligray (mGy) and 51.66 mGy, respectively. When applying low dose cranial CT scan protocols by decreasing mAs to 200, 150 and 100 mAs, the total dose ofright and left lens were of 45.68 mGy and 46.04 mGy for 200 mAs, 34.65 mGy and 34.77 mGy for 150 mAs, 28.73 mGy and 29.25 mGy for 100 mAs respectively. CONCLUSION: A low dose cranial CT scan at 100 miliampere-second provides not only an acceptable clinical image quality, but also decreases the lens's radiation dose by 43%.

Does magnetic resonance imaging give value-added than bone scintigraphy in the detection of vertebral metastasis?

OBJECTIVE: To determine the role of Magnetic Resonance (MR) imaging for the investigation ofpatients with suspected metastasis to the spine by bone scintigraphy. MATERIAL AND METHOD: Retrospectively reviewed with comparison was made between Technetium-99m Methylene Diphosphonate (99(m)Tc-MDP) bone scintigraphy and corresponding spine MR images in 48 cases of vertebral metastasis at Siriraj Hospital. The intervals between bone scintigraphy and MR images did not exceed 1 month. The authors studied between January 2005 and December 2006 Bone scintigraphy were performed with planar imaging of the entire body and MR imaging was performed with the 1.5 tesla and 3.0 tesla scanner using standard techniques with T1-, T2-weighted images and fat-suppressed T1-weighted images with intravenous administration of gadopentetate dimeglumine. The MR imaging findings were studied: location (cervical or thoracic or lumbar or sacrum spine), number of lesions (solitary or multiple lesions), pattern of enhancement (homogeneous or inhomogeneous), involvement of spinal canal, compression of spinal cord, extradural extension, other incidental findings such as pulmonary metastasis, pleural effusion, lymphadenopathy The final diagnosis was confirmed clinically and followed-up for further management (radiation or surgery) or followed-up by MR imaging (1 month-16 months) and bone scintigraphy (5 months-12 months). RESULTS: Forty-eight cases (80 lesions) of vertebral metastasis were identified (25 men and 23 women; mean age 61 years and range 8-84 years). Primary neoplasms include breast cancer (n=11), colorectal cancer (n=7), lung cancer (n=6), prostate cancer (n=5), nasopharyngeal cancer (n=5), head and neck cancer (n=3), thyroid cancer (n=2), liver cancer (n=2), esophagus cancer (n=1), bladder cancer (n=1), retroperitoneum cancer (n=1), medulloblastoma (n=1), cervical cancer (n=1), ovarian cancer (n=1), malignant melanoma (n=1). The result of bone scintigraphy and MR imaging is used to evaluate vertebral metastasis: in 44 lesions of bone scintigraphy positive for vertebral metastasis, 40/44 lesions (91%) which MR imaging reveal vertebral metastasis. This group may not benefit for further investigation by MR imaging. In 24 lesions of negative of bone scintigraphy for vertebral metastasis, the authors found that 14/24 lesions (58%) showed positive of vertebral metastasis from MR imaging. In this group, the authors recommended a further investigation because 58% of negative bone scintigraphy lesions are depicted by only MR imaging. MR imaging demonstrated metastatic cord compression in 16 cases. Extradural extension causes spinal canal narrowing in 30 cases. CONCLUSION: The authors conclude that the MR imaging is more efficient than the bone scintigraphy in detecting vertebral metastasis, especially in the cases that bone scintigraphy are equivocal or negative for vertebral metastasis in high clinical suspicion. Furthermore, MR imaging is important for the further treatment planning such as radiation therapy or systemic chemotherapy. Although MR imaging is useful in the detection of early metastasis that are localized completely in the bone marrow cavity routinely bone scintigraphy remains that most cost-effective method for examination of the entire skeleton.

Diagnostic accuracy of MR imaging in tuberculous spondylitis.

OBJECTIVE: To systemically evaluate MR imaging features of tuberculous spondylitis and to find features that may help differentiating tuberculosis from other spinal diseases. MATERIAL AND METHOD: Retrospective review of 65 MR imaging of two groups of patients between January 2002 and December 2005. Thirty-one patients were diagnosed as tuberculosis spondylitis and the rest were a randomly selected group of 34 patients with other spinal diseases. All images were reviewed by two neuroradiologists blinded to clinical data. Sensitivity and specificity of each MR imaging features were calculated. RESULTS: Three most useful MR imaging features with high sensitivity and specificity (> 80%) were endplate disruption (100%, 81.4%), paravertebral soft tissue (96.8%, 85.3%), and high signal intensity of intervertebral disc on T2W (80.6%, 82.4%). High sensitivity but low specificity signs in MRI included bone marrow edema (90.3%, 76.5%), bone marrow enhancement (100%, 42.5%), posterior element involvement (93.5%, 76.5%), canal stenosis (87.1%, 26.5%), and spinal cord or nerve root compression (80.6%, 38.2%). Low sensitivity but high specificity features in MRI were intervertebral disc enhancement (63.3%, 84.2%), vertebral collapse (58.1%, 85.3%), and kyphosis deformity (67.7%, 82.4%). Overall, the sensitivity and specificity of MRI for spinal tuberculosis were 100% and 88.2% respectively. CONCLUSION: The authors presented three good to excellent sensitivity and specificity MR imaging features for spinal tuberculosis, end plate disruption, paravertebral soft tissue formation, and high signal of intervertebral disc on T2W. In contrast to a previous study, most of the presented cases still presented with classic radiological pictures of "two vertebral disease with the destruction of the intervertebral disc". Only a small portion of the patients revealed sparing intervening disc or isolated single vertebral body involvement, which possibly reflected the early stages of the disease process.

The limited protocol MRI in diagnosis of lumbar disc herniation.

OBJECTIVE: To assess agreement in detection of lumbar disc herniation (LDH) between limited and full protocol MRI. MATERIAL AND METHOD: 123 patients who requested lumbar MRI for diagnosis of disc herniation were assessed The full protocol MRI composed of sagittal T1-wi, sagittal T2-wi and axial T2-wi was performed on each patient. The sagittal T2-wi was selected as the limited protocol MRI. The limited and full protocols MRI of each patient were separately interpreted by three neuroradiologists to assess disc herniation and nerve root compression. The consensus results of limited and full protocol were compared. The diagnostic performance of each protocol was analyzed using surgery as the gold standard. RESULTS: There were 62 females and 61 males enrolled into the study between the age of 21-60 years old (means = 42.91 years). The duration of pain was 1-204 months (mean = 31.20 months). The degree of severity was mild in 23.58%, moderate 45.52% and severe 30.89% of cases. Thirty-three cases were operated on. For detection of LDH, the limited protocol gave the same interpretation results as the full protocol (0 = 1.04, 95%CI = 0.94, 1.14 with accepted range of 0.95-1.05). In nerve root compression, the limited protocol was not accurate as the full protocol (0 = 0.75, 95%CI = 0.87, 0.63). The sensitivity, specificity, accuracy, PPV, NPV, and LR+ in the surgical group of limited protocol in diagnosis of LDH were 82.61%, 80%, 81.82%, 90.48%, 60.67%, 4.13 and in nerve root compression were 54.84%, 100%, 57.58%, 100%, 12.5% respectively. The same statistics of full protocol MRI in diagnosis of LDH were 82.61%, 70.80%, 78.79%, 86.36%, 63.64%, 2.75 and of nerve root compression were 80.65%, 100%, 81.82%, 100%, 25% respectively. The sensitivity and specificity in diagnosis of LDH were not different in both protocols but the sensitivity of nerve root compression was statistically significant different (p < 0.013, 95%CI = -0.33, -0.25). CONCLUSION: The authors concluded that limited protocol MRI may replace full protocol MRI in diagnosis of LDH but not in nerve root compression.

Reliability of intracerebral hemorrhage classification systems: A systematic review.

BACKGROUND: Accurately distinguishing non-traumatic intracerebral hemorrhage (ICH) subtypes is important since they may have different risk factors, causal pathways, management, and prognosis. We systematically assessed the inter- and intra-rater reliability of ICH classification systems. METHODS: We sought all available reliability assessments of anatomical and mechanistic ICH classification systems from electronic databases and personal contacts until October 2014. We assessed included studies' characteristics, reporting quality and potential for bias; summarized reliability with kappa value forest plots; and performed meta-analyses of the proportion of cases classified into each subtype. SUMMARY OF REVIEW: We included 8 of 2152 studies identified. Inter- and intra-rater reliabilities were substantial to perfect for anatomical and mechanistic systems (inter-rater kappa values: anatomical 0.78-0.97 [six studies, 518 cases], mechanistic 0.89-0.93 [three studies, 510 cases]; intra-rater kappas: anatomical 0.80-1 [three studies, 137 cases], mechanistic 0.92-0.93 [two studies, 368 cases]). Reporting quality varied but no study fulfilled all criteria and none was free from potential bias. All reliability studies were performed with experienced raters in specialist centers. Proportions of ICH subtypes were largely consistent with previous reports suggesting that included studies are appropriately representative. CONCLUSIONS: Reliability of existing classification systems appears excellent but is unknown outside specialist centers with experienced raters. Future reliability comparisons should be facilitated by studies following recently published reporting guidelines.