Comparison of Multispectral Image-Processing Methods for Brain Tissue Classification in BrainWeb Synthetic Data and Real MR Images.

Accurate quantification of brain tissue is a fundamental and challenging task in neuroimaging. Over the past two decades, statistical parametric mapping (SPM) and FMRIB's Automated Segmentation Tool (FAST) have been widely used to estimate gray matter (GM) and white matter (WM) volumes. However, they cannot reliably estimate cerebrospinal fluid (CSF) volumes. To address this problem, we developed the TRIO algorithm (TRIOA), a new magnetic resonance (MR) multispectral classification method. SPM8, SPM12, FAST, and the TRIOA were evaluated using the BrainWeb database and real magnetic resonance imaging (MRI) data. In this paper, the MR brain images of 140 healthy volunteers (51.5 ± 15.8 y/o) were obtained using a whole-body 1.5 T MRI system (Aera, Siemens, Erlangen, Germany). Before classification, several preprocessing steps were performed, including skull stripping and motion and inhomogeneity correction. After extensive experimentation, the TRIOA was shown to be more effective than SPM and FAST. For real data, all test methods revealed that the participants aged 20-83 years exhibited an age-associated decline in GM and WM volume fractions. However, for CSF volume estimation, SPM8-s and SPM12-m both produced different results, which were also different compared with those obtained by FAST and the TRIOA. Furthermore, the TRIOA performed consistently better than both SPM and FAST for GM, WM, and CSF volume estimation. Compared with SPM and FAST, the proposed TRIOA showed more advantages by providing more accurate MR brain tissue classification and volume measurements, specifically in CSF volume estimation.

Novel Automated Method for the Detection of White Matter Hyperintensities in Brain Multispectral MR Images.

BACKGROUND: According to the Standards for Reporting Vascular Changes on Neuroimaging, White Matter Hyperintensities (WMHs) are cerebral white matter lesions that are characterized by abnormal tissues of variable sizes and appear hyperintense in T2-weighted Magnetic Resonance (MR) measurements without cavitation (i.e., their tissue signals differ from those of Cerebrospinal Fluid or CSF). Such abnormal tissue regions are typically observed in the MR images of brains of healthy older adults and are associated with a number of geriatric neurodegenerative diseases. Explanations of the exact causes and mechanisms of these diseases remain inconclusive. Moreover, WMHs are typically identified by visual assessment and manual examination, both of which require considerable time. This brings up a need of developing a method for detecting WMHs more objectively and enabling patients to be treated early. As a consequence, damages on nerve cells can be limited and the severity of patients' conditions can be contained. AIMS: This paper presents a computer-aided technique for automatically detecting and segmenting anomalies in MR images. METHODS: The method has two steps: (1) a Band Expansion Process (BEP) to expand the dimensions of brain MR images nonlinearly and (2) anomaly detection algorithms to detect WMHs. Synthesized MR images provided by BrainWeb were used as benchmarks against which the detection performance of the algorithms was determined. RESULTS: The most notable findings are as follows: Firstly, compared with the other anomaly detection algorithms and the Lesion Segmentation Tool (LST), BEP-anomaly detection is shown to be the most effective in detecting WMHs. Secondly, across all levels of background noise and inhomogeneity, the mean Similarity Index (SI) produced by our proposed algorithm is higher than that produced by LST, indicating that the algorithm is more effective than LST in segmenting WMHs from brain MR images. CONCLUSION: Experimental results demonstrated a significantly high accuracy of the BEP-K/R-RX method in detection of synthetic brain MS lesion data. In the meantime, it also effectively enhances the detection of brain lesions.

Focal Liver Lesions: Real-time 3-Dimensional Contrast-Enhanced Ultrasonography Compared With 2-Dimensional Contrast-Enhanced Ultrasonography and Magnetic Resonance Imaging.

OBJECTIVES: This study sought to evaluate the application of real-time 3-dimensional (3D) contrast-enhanced ultrasonography (US) to diagnose focal liver lesions and to compare these results with those from 2-dimensional (2D) contrast-enhanced US and contrast-enhanced magnetic resonance imaging (MRI). METHODS: Patients with focal liver lesions were examined by 2D contrast-enhanced US, 3D contrast-enhanced US, and contrast-enhanced MRI for lesion characterization, and biopsies and comprehensive clinical diagnoses served as reference standards. The sensitivity, specificity, area under the receiver operating characteristic curve, and intermodality agreement were assessed. The number of contrast agent injections and lesions observed per injection were calculated for 3D and 2D contrast-enhanced US. The number and display quality of the feeding arteries observed with 3D and 2D contrast-enhanced US were assessed. RESULTS: A total of 117 patients with 151 focal liver lesions were enrolled, including 67 cases of hepatocellular carcinoma, 51 cases of liver metastasis, and 33 cases of benign liver lesions. No significant differences were found among the modalities. The sensitivity values for 3D contrast-enhanced US, 2D contrast-enhanced US, and contrast-enhanced MRI were 96%, 95%, and 93%, respectively; the specificity values were 87%, 84%, and 89%; and the area under the receiver operating characteristic curve values were 0.92, 0.90, and 0.92. The intermodality agreement was excellent (κ > 0.77). Fewer contrast agent injections were needed, and more lesions and feeding arteries were more clearly displayed on 3D than 2D contrast-enhanced US (P < .001). CONCLUSIONS: Real-time 3D contrast-enhanced US is useful for diagnosing focal liver lesions and for observing feeding arteries with fewer contrast agent injections.

Patient safety during radiological examinations: a nationwide survey of residency training hospitals in Taiwan.

OBJECTIVES: Variations in radiological examination procedures and patient load lead to variations in standards of care related to patient safety and healthcare quality. To understand the status of safety measures to protect patients undergoing radiological examinations at residency training hospitals in Taiwan, a follow-up survey evaluating the full spectrum of diagnostic radiology procedures was conducted. DESIGN: Questionnaires covering 12 patient safety-related themes throughout the examination procedures were mailed to the departments of diagnostic radiology with residency training programmes in 19 medical centres (with >500 beds) and 17 smaller local institutions in Taiwan. After receiving the responses, all themes in 2014 were compared between medical centres and local institutions by using χ(2) or 2-sample t-tests. PARTICIPANTS: Radiology Directors or Technology Chiefs of medical centres and local institutions in Taiwan participated in this survey by completing and returning the questionnaires. RESULTS: The response rates of medical centres and local institutions were 95% and 100%, respectively. As indicated, large medical centres carried out more frequent clinically ordered, radiologist-guided patient education to prepare patients for specific examinations (CT, 28% vs 6%; special procedures, 78% vs 44%) and incident review and analysis (89% vs 47%); however, they required significantly longer access time for MRI examinations (7.00±29.50 vs 3.50±3.50 days), had more yearly incidents of large-volume contrast-medium extravasation (2.75±1.00 vs 1.00±0.75 cases) and blank radiographs (41% vs 8%), lower monthly rates of suboptimal (but interpretable) radiographs (0.00±0.01% vs 0.64±1.84%) and high-risk reminder reporting (0.01±0.16% vs 1.00±1.75%) than local institutions. CONCLUSIONS: Our study elucidates the status of patient safety in diagnostic radiology in Taiwan, thereby providing helpful information to improve patient safety guidelines needed for medical imaging in the future.

Impact of positional difference on the measurement of breast density using MRI.

PURPOSE: This study investigated the impact of arms/hands and body position on the measurement of breast density using MRI. METHODS: Noncontrast-enhanced T1-weighted images were acquired from 32 healthy women. Each subject received four MR scans using different experimental settings, including a high resolution hands-up, a low resolution hands-up, a high resolution hands-down, and finally, another high resolution hands-up after repositioning. The breast segmentation was performed using a fully automatic chest template-based method. The breast volume (BV), fibroglandular tissue volume (FV), and percent density (PD) measured from the four MR scan settings were analyzed. RESULTS: A high correlation of BV, FV, and PD between any pair of the four MR scans was noted (r > 0.98 for all). Using the generalized estimating equation method, a statistically significant difference in mean BV among four settings was noted (left breast, score test p = 0.0056; right breast, score test p = 0.0016), adjusted for age and body mass index. Despite differences in BV, there were no statistically significant differences in the mean PDs among the four settings (p > 0.10 for left and right breasts). Using Bland-Altman plots, the smallest mean difference/bias and standard deviations for BV, FV, and PD were noted when comparing hands-up high vs low resolution when the breast positions were exactly the same. CONCLUSIONS: The authors' study showed that BV, FV, and PD measurements from MRI of different positions were highly correlated. BV may vary with positions but the measured PD did not differ significantly between positions. The study suggested that the percent density analyzed from MRI studies acquired using different arms/hands and body positions from multiple centers can be combined for analysis.

Robust volume assessment of brain tissues for 3-dimensional fourier transformation MRI via a novel multispectral technique.

A new TRIO algorithm method integrating three different algorithms is proposed to perform brain MRI segmentation in the native coordinate space, with no need of transformation to a standard coordinate space or the probability maps for segmentation. The method is a simple voxel-based algorithm, derived from multispectral remote sensing techniques, and only requires minimal operator input to depict GM, WM, and CSF tissue clusters to complete classification of a 3D high-resolution multislice-multispectral MRI data. Results showed very high accuracy and reproducibility in classification of GM, WM, and CSF in multislice-multispectral synthetic MRI data. The similarity indexes, expressing overlap between classification results and the ground truth, were 0.951, 0.962, and 0.956 for GM, WM, and CSF classifications in the image data with 3% noise level and 0% non-uniformity intensity. The method particularly allows for classification of CSF with 0.994, 0.961 and 0.996 of accuracy, sensitivity and specificity in images data with 3% noise level and 0% non-uniformity intensity, which had seldom performed well in previous studies. As for clinical MRI data, the quantitative data of brain tissue volumes aligned closely with the brain morphometrics in three different study groups of young adults, elderly volunteers, and dementia patients. The results also showed very low rates of the intra- and extra-operator variability in measurements of the absolute volumes and volume fractions of cerebral GM, WM, and CSF in three different study groups. The mean coefficients of variation of GM, WM, and CSF volume measurements were in the range of 0.03% to 0.30% of intra-operator measurements and 0.06% to 0.45% of inter-operator measurements. In conclusion, the TRIO algorithm exhibits a remarkable ability in robust classification of multislice-multispectral brain MR images, which would be potentially applicable for clinical brain volumetric analysis and explicitly promising in cross-sectional and longitudinal studies of different subject groups.

The outcome of a quality-controlled mammography screening program: experience from a population-based study in Taiwan.

BACKGROUND: A greater policy of emphasis on the early detection and treatment of breast cancer is prevalent among developed countries. To raise the screening performance with a potentially decreased mortality rate, it is crucial to evaluate and analyze the screening outcome after implementation. We report the clinical outcome of an 8-year nationwide mammography screening in Taiwan to help share our statistical information on breast screening worldwide, especially in Asia. METHODS: Taiwan has provided nationwide, free, biennial mammographic screening since 2004. A total of 2,392,789 consecutive screening mammography examinations were performed during this study period for women aged 50-69 years (2006-2009) and 45-69 years (from December 2009 onwards). The screening covers 33.2% of the target population in the most recent 2 years. The workload of every screening radiologist, the overall recall rate, positive predictive value (PPV1), cancer detection rate (CDR), cancer incidence rate (CIR) from the screening, 1-year interval cancer, sensitivity, and specificity of the screening mammography are calculated, and compared with the American College of Radiology (ACR) recommendation level and/or those of other screening mammographic series. RESULTS: The CDRs (%) and CIRs (%) increased from 3.94-4.08 and 4.80-5.04 to 4.71-5.04 and 5.71 after 2009, implying a high occurrence of breast cancer in the younger age group of 45-49 years. The recall rates (9.3-10.0%) in this review are within the ACR recommendation range (<10%) and the PPV1 has also reached the ACR recommended level (>5%) in the most recent 2 years. The improvement of the screening performance may be attributed to our peer auditing review and education program. The sensitivity of our screening mammography is slightly lower than that of the ACR recommended level (>85%), which is still comparable to the results of the Vermont area in the USA. Although the workload (screenees/screeners) for every radiologist each year has increased from 150 in 2004 to 1360 in 2012, it does not seem to worsen the quality outcome of this screening program. CONCLUSION: From the outcome review of this national mammography screening, there is still room to ameliorate our performance through comprehensive and continued education, to improve the competence of cancer detection and decrease false negative (FN) cases.

Peer reviewing of screening mammography in Taiwan: its reliability and the improvement.

BACKGROUND: Early detection with screening mammography can potentially reduce breast cancer mortality rates. To achieve an efficient screening, a peer review system provides a compensatory double-check reviewing, will hopefully to prevent the omission of detectable lesions and reduce unnecessary recall. METHODS: In 2009, 4643 initial mammographic screenings reported by 74 screening radiologists had negative results with a recall rate of less than 5%. In the same year, 2538 initial positives screened by 18 screening radiologists had a recall rate higher than 15%. Those 7181 randomized screenings were evenly distributed for reassessment by 39 reviewing radiologists. The disagreement of assessments between the reviewers and screening radiologists was recorded. The differential rate was defined as the number of the disagreements divided by the number of audited films reviewed by a screening radiologist. The equality of the differential rates for each screening radiologists with negative and positive assessments was compared by a Chi-square test. The performance of the 39 auditors was measured by the Kendall's tau statistic. P values less than 0.05 were considered statistically significant. RESULTS: The mean differential rate for screening radiologists of negative assessments was 6.7% (P = 0.588), while 35.0% for positive assessments were significant (P < 0.001). The result indicated that most of the initial negative assessments reported by the screening radiologists were generally accepted by the reviewers but not the positive assessments. With respect to the 39 reviewers, there was no significant evidence for the association of the difference rates between negative and positive assessments. Nine reviewers were found to have their differential rate for negative and positive assessments larger than the average of the population. Eleven reviewers were found to have their differential rates smaller than the average for both. Thirteen reviewers had their differential rates smaller than the average for negative assessments but larger than the average for positive assessments. The opposite condition was found for six reviewers. The Kendall's tau statistic was 0.038 (P = 0.735). CONCLUSIONS: Reviewers usually agreed with the opinion of the initial screening doctors who reported negative findings. Therefore, a 5% recall rate as the lower range of reviewing negatives may be still too high. The recall rate of more than 15% was significantly related to improper interpretation, especially when the differential rate is 25% or higher, a warning to the underperforming screening radiologist is recommended. An ideal reviewer should interpret films independently. Reviewers with tendencies to be followers or contrarians should not be enrolled in the reviewing system.

Can missed breast cancer be recognized by regular peer auditing on screening mammography?

BACKGROUND: This study was conducted to investigate whether detectable missed breast cancers could be distinguished from truly false negative images in a mammographic screening by a regular peer auditing. METHODS: Between 2004 and 2007, a total of 311,193 free nationwide biennial mammographic screenings were performed for 50- to 69-year-old women in Taiwan. Retrospectively comparing the records in Taiwan's Cancer registry, 1283 cancers were detected (4.1 per 1000). Of the total, 176 (0.6 per 1000) initial mammographic negative assessments were reported to have cancers (128 traditional films and 48 laser-printed digital images). We selected 186 true negative films (138 traditional films and 48 laser-printed ones) as control group. These were seeded into 4815 films of 2008 images to be audited in 2009. Thirty-four auditors interpreted all the films in a single-blind, randomized, pair-control study. The performance of 34 auditors was analyzed by chi-square test. A p value of < 0.05 was considered significant. RESULTS: Eight (6 traditional and 2 digital films) of the 176 false negative films were not reported by the auditors (missing rate of 4.5%). Of this total, 87 false negatives were reassessed as positive, while 29 of the 186 true negatives were reassessed as positive, making the overall performance of the 34 auditors in interpreting the false negatives and true negatives a specificity of 84.4% and sensitivity of 51.8%. The specificity and sensitivity in traditional films and laser-printed films were 98.6% versus 43.8% and 41.8% versus 78.3%, respectively. Almost 42% of the traditional false negative films had positive reassessment by the auditors, showing a significant difference from the initial screeners (p < 0.001). The specificity of their reinterpretation of laser-printed films was obviously low. CONCLUSION: Almost 42% of the false negative traditional films were judged as missed cancers in this study. A peer auditing should reduce the probability of missed cancers.

Correction of left ventricular wall thickening from short-axis cine MRI for basal-descent through-plane motion.

PURPOSE: To solve the problem of the basal descent movement in quantification of the regional left ventricular (LV) myocardial wall thickness (WTh) and wall thickening (%WT) in short-axis (SA) cine MRI for effectively assessing the regional wall motion of LV myocardium. MATERIALS AND METHODS: LV long-axis tagged MRI and SA cine MRI were performed to calculate the longitudinal translation and circumferential WTh of LV myocardium in eight normal volunteers. The new SA end-systolic thickness (EST) data were reconstructed from the original EST data, based on the quantified longitudinal translation of LV myocardium. RESULTS: The mean %WT of six segments in the basal section after correction was significantly different from that before correction in both intra- and inter-operator experiments. The polar map also showed the significant improvement of the variability of regional %WT and lack of quantification of %WT in the most basal SA slices after correction. CONCLUSION: The proposed technique demonstrated an important advantage to calculate the %WT in the most basal SA myocardial tissue, which was considered difficult to be achieved using cine MRI.

Quantitative analysis in clinical applications of brain MRI using independent component analysis coupled with support vector machine.

PURPOSE: To effectively perform quantification of brain normal tissues and pathologies simultaneously, independent component analysis (ICA) coupled with support vector machine (SVM) is investigated and evaluated for effective volumetric measurements of normal and lesion tissues using multispectral MR images. MATERIALS AND METHODS: Synthetic and real MR data of normal brain and white matter lesion (WML) data were used to evaluate the accuracy and reproducibility of gray matter (GM), white matter (WM), and WML volume measurements by using the proposed ICA+SVM method to analyze three sets of MR images, T1-weighted, T2-weighted, and proton density/fluid-attenuated inversion recovery images. RESULTS: The Tanimoto indexes of GM/WM classification in the normal synthetic data calculated by the ICA+SVM method were 0.82/0.89 for data with 0% noise level. As for clinical MR data experiments, the ICA+SVM method clearly extracted the normal tissues and white matter hyperintensity lesions from the MR images, with low intra- and inter-operator coefficient of variations. CONCLUSION: The experiments conducted provide evidence that the ICA+SVM method has shown promise and potential in applications to classification of normal and pathological tissues in brain MRI.

Health effects of medical radiation on cardiologists who perform cardiac catheterization.

BACKGROUND: We investigated the health effects of low-dose radiation on cardiologists exposed to scattered radiation while performing cardiac catheterization (CC) in a hospital setting from 2003 to 2006. METHODS: We performed a 4-year retrospective study on 2, 292 medical doctors, using claims data from all contracted hospitals of the Bureau of National Health Insurance, Taiwan. We gathered statistical data regarding radiation-related diseases using the International Classification of Diseases, 9(th) Revision, Clinical Modification record numbers of each doctor. RESULTS: Of the 2,292 doctors evaluated, 1,721 were aged 35-50 years and the remaining 571 were aged 51-65 years. There were 892 cardiologists who performed CC (experimental group), and the majority of these (733/892, 82.17%) were aged 35-50 years. There were 1,400 medical doctors who performed no CC from 2003 to 2006 (control group). A total of 988 of these belonged to the 35-50 years age group and 412 to the 51-65 years group. In the 35-50 years group, the controls had significantly more medical visits for hematological and thyroid cancer (p <0.05), skin disease (p <0.001), and acute upper respiratory tract infection (p <0.001) compared with the experimental group. In contrast, cardiologists who performed catheterization had more cataracts compared with the control group, but this difference was not significant. CONCLUSION: Doctors who did not perform CC had more visits for radiation-related diseases than those who performed catheterization. In the experimental group, cardiologists aged 35-50 years who were exposed to radiation during CC had more visits for cataracts than the control group. We recommend that radiation protection concepts be emphasized to cardiologists, and that hospital managers be obligated to upgrade angiography equipment because the newer models have less scattered radiation.

An automatic computer-aided detection system for meniscal tears on magnetic resonance images.

Knee-related injuries including meniscal tears are common in both young athletes and the aging population, and require accurate diagnosis and surgical intervention when appropriate. With proper techniques and radiologists' experienced skills, confidence in detection of meniscal tears can be quite high. This paper develops a novel computer-aided detection (CAD) diagnostic system for automatic detection of meniscal tears in the knee. Evaluation of this CAD system using an archived database of images from 40 individuals with suspected knee injuries indicates that the sensitivity and specificity of the proposed CAD system are 83.87% and 75.19%, respectively, compared to the mean sensitivity and specificity of 77.41% and 81.39%, respectively, obtained by experienced radiologists in routine diagnosis without using the CAD. The experimental results suggest that the developed CAD system has great potential and promise in automatic detection of both simple and complex meniscal tears of the knee.

Band expansion-based over-complete independent component analysis for multispectral processing of magnetic resonance images.

Independent component analysis (ICA) has found great promise in magnetic resonance (MR) image analysis. Unfortunately, two key issues have been overlooked and not investigated. One is the lack of MR images to be used to unmix signal sources of interest. Another is the use of random initial projection vectors by ICA, which causes inconsistent results. In order to address the first issue, this paper introduces a band-expansion process (BEP) to generate an additional new set of images from the original MR images via nonlinear functions. These newly generated images are then combined with the original MR images to provide sufficient MR images for ICA analysis. In order to resolve the second issue, a prioritized ICA (PICA) is designed to rank the ICA-generated independent components (ICs) so that MR brain tissue substances can be unmixed and separated by different ICs in a prioritized order. Finally, BEP and PICA are combined to further develop a new ICA-based approach, referred to as PICA-BEP to perform MR image analysis.

Reconstruction and rendering of microcalcifications from two mammogram views by modified projective grid space (MPGS).

Mammograms taken by two views: cranio-caudal (CC) and medio-lateral oblique (MLO) views provide only 2D projections of the microcalcifications, which lack the depth information. Thus, envisioning the relative lesion location from mammograms is a challenge for radiologists. To assist radiologists in locating and rendering lesion tissues, a modified projective grid space (MPGS) scheme is proposed to reconstruct 3D microcalcifications. The MPGS scheme reconstructs 3D microcalcifications in a unique space defined by corresponding points and the epipoles retrieved from the fundamental matrix of the CC and MLO views. Since only corresponding points of images are required in the proposed MPGS scheme, we can avoid the difficulty associated with most reconstruction approaches that require prior complicated calibration of X-ray machine. Considering the deformation of the breast, a new method based on the concept of bundle adjustment is proposed to rectify the 3D locations of reconstructed microcalcifications by uncompressed breast model reconstructed from the real patient body using MPGS scheme with iterative closest point (ICP). Then, the reconstructed microcalcifications are augmented in the real patient body model to show their relative positions.

3D localization of clustered microcalcifications using cranio-caudal and medio-lateral oblique views.

This paper presents a 3D localization method to register clustered microcalcifications on mammograms from cranio-caudal (CC) and medio-lateral oblique (MLO) views. The method consists of three major components: registration of clustered microcalcifications in CC and MLO views, 3D localization of clustered microcalcifications and 3D visualization of clustered microcalcifications. The registration is performed based on three features, gradient, energy and local entropy codes that are independent of spatial locations of microcalcifications in two different views and are prioritized by discriminability in a binary decision tree. The 3D localization is determined by a sequence of coordinate corrections of calcified pixels using the breast nipple as a controlling point. Finally, the 3D visualization implements a virtual reality modeling language viewer (VRMLV) to view the exact location of the lesion as a guide for needle biopsy. In order to validate our proposed 3D localization system, a set of breast lesions, which appear both in mammograms and in MR Images is used for experiments where the depth of clustered microcalcifications can be verified by the MR images.

Unilateral subcortical calcification: a manifestation of dural arteriovenous fistula.

We describe the imaging findings in a dural arteriovenous fistula (AVF) with unilateral subcortical calcification. A 50-year-old woman patient suffered from hypertension and chronic headache. Recently, marked headache and a changed consciousness level were noted. The imaging studies demonstrated left subcortical calcification and cerebral sulcus effacement. MR imaging and angiography revealed multiple abnormal tortuous vessels, mainly from left external carotid artery with left-sided transverse sinus occlusion. The final diagnosis was type II a + b dural AVF (classification of Djindjian and Merland), and the patient underwent endovascular embolization and radiosurgery.

Imaging characteristics of familial Wolfram syndrome.

Wolfram syndrome is a rare diffuse neurodegenerative disorder also known as DIDMOAD due to its characteristics of diabetes insipidus, diabetes mellitus, optic nerve atrophy and deafness. It is also associated with a wide variety of abnormalities of the central nervous system, urinary tract and endocrine glands. It may be familial or sporadic. Imaging findings include absence of the short T1 nature of the pituitary posterior lobe, atrophy of the optic nerve, chiasma, and tracts. Atrophy of the brain stem and cerebellum has also been reported. We describe a 15-year-old boy and an 11-year-old girl with Wolfram syndrome who were siblings from a diabetes mellitus family. They received regular insulin control at our hospital and had symptoms of frequent urinary tract infection and diabetes insipidus. Magnetic resonance imaging revealed marked pons and cerebellar atrophy. Optic nerve and chiasma atrophy was also noted.

3D ROC Analysis for Medical Imaging Diagnosis.

Receiver operating characteristics (ROC) has been widely used as a performance evaluation tool to measure effectiveness of medical modalities. It is derived from a standard detection theory with false alarm and detection power interpreted as false positive (FP) and true positive (TP) respectively in terms of medical diagnosis. The ROC curve is plotted based on TP versus FP via hard decisions. This paper presents a three dimensional (3D) ROC analysis which extends the traditional two-dimensional (2D) ROC analysis by including a threshold parameter in a third dimension resulting from soft decisions, (SD). As a result, a 3D ROC curve can be plotted based on three parameters, TP, FP and SD. By virtue of such a 3D ROC curve three two-dimensional (2D) ROC curves can be derived, one of which is the traditional 2D ROC curve of TP versus FP with SD reduced to hard decision. In order to illustrate its utility in medical diagnosis, its application to magnetic resonance (MR) image classification is demonstrated.