Subjective Visual Evaluation Research Articles

One-dimensional quantitative evaluation of peripheral lung adenocarcinoma with or without ground-glass opacity on thin-section CT images using profile curves

The purpose of our investigation was to compare the usefulness of the subjective visual assessment of ground-glass opacity (GGO) with a quantitative method that used a profile curve to determine prognosis. 96 adenocarcinomas were studied. Three diameters ([D1]-[D3]) were defined for estimating the diameter of tumours on the monitor: the distance between two points was measured using software that displays a CT density profile across the tumour. One experienced and one less experienced radiologist independently evaluated the following six parameters: the three diameters [D1]-[D3]; the solid portion of total tumour in the two different ratios ([D2]/[D1], [D3]/[D1]); and the area ratio of GGO for total opacity to subjective visual evaluation. Interobserver agreement between the two radiologists of the diameters (mean bias+/- 1.96 standard deviations) was as follows: [D1], -0.7 +/- 6 mm; [D2], 0.4 +/- 4.4 mm; and [D3], -0.1 +/- 4.2 mm (Bland and Altman's method). Interobserver agreement was fair in evaluating the area ratio of GGO (kappa test, kappa = 0.309). Univariate logistic regression analysis revealed that two ratios ([D2]/[D1], [D3]/[D1]) might be significantly useful in estimating lymph node metastasis (p < 0.026), lymph duct invasion (p < 0.001) and recurrence (p < 0.015). Observation of the area ratio of GGO by an experienced radiologist would be necessary for estimating lymph node metastasis (p = 0.04) and lymph duct invasion (p < 0.001). We concluded that the ratio of solid component to total tumour, which is obtainable in a more objective and simple way using profile curves obtained by software, is a more useful method of estimating prognosis than is visual assessment.

Spatial error concealment for H.264 using sequential directional interpolation

Error concealment at the decoder restores erroneous macroblocks (MBs) caused by channel errors. In this paper, we propose a novel spatial error concealment algorithm based on prediction modes of intra-blocks which are included in a H.264-coded stream and highly correlated to the direction of local edge within the block. The key contribution is to sequentially interpolate each pixel in a lost MB by utilizing edge directions and strengths efficiently estimated from the neighboring blocks, preserving local edge continuity for more visually acceptable images. The proposed scheme is simple to implement and more reliably recover high-detailed content in corrupted MBs. The experimental results shows the proposed method achieves reduction in speed by 14% <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">~</sub> 39% as compared to existing method, and outperforms them in PSNR by 0.5 <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">~</sub> 1 dB as well as in subjective visual evaluation.

Comparing two objective methods for the aesthetic evaluation of breast cancer conservative treatment

Two programs were recently developed for the aesthetic evaluation of results in breast cancer conservative treatment: the Breast Cancer Conservative Treatment cosmetic results (BCCT.core) and the Breast Analyzing Tool (BAT). Both make use of a face-only photographic view of the patient and were developed to overcome the lack of reproducibility observed with subjective visual evaluation. The BCCT.core analyses several parameters related to asymmetry, color differences and scar appearance, while the BAT considers only asymmetry measurements. The purpose of this study was to compare the performance of these two methods. Digital pictures of 59 patients from Porto and 60 from Vienna were evaluated subjectively by two panels using the four-class Harris scale. The Porto photographs had a similar backlight and better quality, and were evaluated by an international panel of 23 experts. The Vienna photographs had different backlight and lower quality, and were evaluated by four students and two breast cancer specialists. All 119 cases were submitted to analysis using the BCCT.core and BAT. Agreement between the software programs and the subjective evaluation was calculated using kappa (k), weighted kappa statistics (wk) and error rate (er). In overall analysis, BCCT.core program obtained a better agreement with the subjective evaluation (k = 0.56; wk = 0,64; er = 0.20) than the BAT software (k = 0.39; wk = 0.46; er = 0.42) (P < 0.0007). Results were again better for the BCCT.core program, when analysing the photographs obtained in Porto (k = 0.71; wk = 0.78; er = 0.14) than for the BAT (k = 0.35; wk = 0.41; er = 0.51) (P < 0.0003) while no significant differences in agreement were obtained regarding the Vienna images (P > 0.1). The results suggest that the inclusion of multiple parameters in image analyses of aesthetic results has the potential to improve results. However, picture quality is probably important for analysis of other features besides asymmetry.

The thorny subject of insular endemic taxonomy: morphometrics reveal no evidence of speciation between Coenonympha corinna and Coenonympha elbana butterflies (Lepidoptera: Nymphalidae)

Species demarcation between mainland and island endemic populations is often subjective according to the biological concept of species. However, the recognition of the specific status of island endemics is decisive for biogeography and conservation studies. In this paper we applied rigorous morphometric analyses instead of subjective visual evaluation to solve the long debated question whether the endemic butterfly from Tuscany Coenonympha elbana should be considered as a separated entity from the endemic C. corinna from Sardinia and Corsica. We carried out traditional and geometric morphometric analyses of wing pattern and male genitalia comparing the populations from Corsica, Sardinia, Tuscan Archipelago and Tuscany. We found that Sardo-Corsican C. corinna and Tuscan C. elbana can be distinguished only on the basis of the dimension of ventral wing ocelli. However, this characteristic shows a clinal pattern from Sardinia to Tuscany. Moreover, there is not any clear diversification in the shape of male genitalia between the two entities and each islet population can not be unambiguously attributed to one of the two taxa. In conclusion, we found no decisive evidence for a specific diversification between these two entities and the Coenonympha corinna/elbana complex, occurring both in Sardinia-Corsica and in the Italian mainland, should not be strictly considered as an endemic insular taxa.

Comparing two objective methods for the aesthetic evaluation of breast cancer conservative treatment

Two programs were recently developed for the aesthetic evaluation of results in breast cancer conservative treatment: the Breast Cancer Conservative Treatment cosmetic results (BCCT.core) and the Breast Analyzing Tool (BAT). Both make use of a face-only photographic view of the patient and were developed to overcome the lack of reproducibility observed with subjective visual evaluation. The BCCT.core analyses several parameters related to asymmetry, color differences and scar appearance, while the BAT considers only asymmetry measurements. The purpose of this study was to compare the performance of these two methods. Material and methods Digital pictures of 59 patients from Porto and 60 from Vienna were evaluated subjectively by two panels using the four-class Harris scale. The Porto photographs had a similar backlight and better quality, and were evaluated by an international panel of 23 experts. The Vienna photographs had different backlight and lower quality, and were evaluated by four students and two breast cancer specialists. All 119 cases were submitted to analysis using the BCCT.core and BAT. Agreement between the software programs and the subjective evaluation was calculated using kappa (k), weighted kappa statistics (wk) and error rate (er). Results In overall analysis, BCCT.core program obtained a better agreement with the subjective evaluation (k = 0.56; wk = 0,64; er = 0.20) than the BAT software (k = 0.39; wk = 0.46; er = 0.42) (P 0.1). Conclusions The results suggest that the inclusion of multiple parameters in image analyses of aesthetic results has the potential to improve results. However, picture quality is probably important for analysis of other features besides asymmetry.

「錯視現象」が胸部ＣＴ画像の気胸の診断能に及ぼす影響 ‐視覚心理学的シミュレーションによる実証‐

Some cases have been reported in which an optical illusion of lightness perception influences the detectability in diagnosis of low-density hematoma in head CT images in addition to the visual impression of the photographic density of the brain. Therefore, in this study, the author attempted to compare the detectability in diagnosis for chest images with pneumothorax using visual subjective evaluation, and investigated the influence of optical illusion on that detectability in diagnosis. Results indicated that in the window setting of lung, on such an occasion when the low-absorption free space with pneumothorax forms a crescent or the reduced lung borders on the chest-wall, an optical illusion in which the visual impression on the difference of the film contrast between the lung and the low-absorption free space with pneumothorax was psychologically emphasized when contrast was observed. In all cases the detectability in diagnosis for original images with the white thorax and mediastinum was superior to virtual images. Further, in case of the virtual double window setting of lung, thorax, and mediastinum, under the influence of the difference in the radiological anatomy of thorax and mediastinum as a result of the grouping theories of lightness computation, an optical illusion different from the original images was observed.

Image restoration using L1 norm penalty function

The process of estimating an original image from a given blurred and noisy image is known as image restoration. It is an ill-posed inverse problem, since one of the ways of solving it requires finding a solution to a Fredholm integral equation of convolution type in two-dimensional space. The focus of the article is to achieve a quality edge preserving image restoration using a less expensive (fast) regularization technique with L1 norm penalty function. L1 norm based approaches do not penalize edges or high frequency contents in the restored image compared to L2 norm based approaches. Total variation (TV) is an established L1 norm regularization approach that performs edge preserving image restoration, but at a high computational cost. TV regularization requires linearization of a highly nonlinear penalty term, which increases the restoration time considerably for large scale images. In order to reduce the computational cost, we extend least absolute shrinkage and selection operator (LASSO), an L1 norm minimization statistical modeling technique to image restoration. The penalty function of LASSO is an identity matrix so it is computationally fast. The metrics, like, residual error, peak signal to noise ratio (PSNR), restoration time, edge map of the restored image, and subjective visual evaluation are used to assess the performances of both methods. Based on our experimental results, we show that LASSO achieves similar quality of edge preserving restoration as TV regularization, and is approximately two times faster in computation compared to TV regularization on the same set of images. We also analyze the impact of the different degree of blurring caused by point spread functions (PSFs) corrupted by different signal to noise ratios (SNRs) on image restoration.

Performance differences in laparoscopic surgical skills between true high-definition and three-chip CCD video systems

Laparoscopic surgery requires surgeons to rely on visual clues for discrimination among differing tissues and for depth of field on a two-dimensional screen. High definition (HD) provides a superior image. If there is a measurable advantage with HD television (TV), the increase in the cost of the technology would be justified. A digital three-chip CCD camera with a standard monitor (SD system) and a true HD camera (1,080 pixels) with a 16:9-ratio HD monitor (HD system) were compared in clinical and laboratory settings. Three experiments were performed: (1) subjective visual evaluation of the HD and SD systems during actual surgical cases, (2) subjective visual evaluation in a controlled laboratory surgical setting with simultaneous parallel recording, and (3) three laparoscopic surgical task evaluations in a laboratory setting, namely, task A (metric analysis of participants on the surgical simulator), task B (simple eye-hand coordination performance), and task C (knot tying). All 53 participants subjectively evaluated HD as superior to SD in the laboratory setting and during actual surgery. In task B, there was no significant difference between SD and HD (dominant hand: p = 0.19; nondominant hand: p = 0.07). In task C, the knot-tying time was significantly less when performed with HD (mean, 173 +/- 84 s vs 214 +/- 107 s; p = 0.003). Most importantly, subjects with less skill (more documented time required in the basic module on a surgical simulator) improved significantly in the knot-tying task with the HD system (R = 0.631; p = 0.005). All the participants preferred HD to SD. High definition significantly improved laparoscopic knot tying, which requires precise depth perception, proving that HD is more than just a pretty picture.

Subjective visual evaluation vs. traditional and geometric morphometrics in species delimitation: a comparison of moth genitalia

Abstract Species‐level taxonomic studies usually include detailed morphological descriptions of taxa. Traditionally, species descriptions have been based on nonmetric, subjective, visual comparisons of morphological traits regardless of whether diagnostic characters between species are quantitative or qualitative in nature. In difficult cases, traditional morphometrics (morphometrics based on linear measurements or ratios) and appropriate multivariate statistics have been applied to validate species delimitations. Modern geometric morphometrics, a new powerful method to quantify shapes, is increasingly being used in taxonomy and systematics. We compared the results from the three morphology‐based methods (subjective visual differentiation and differentiation by traditional morphometrics and geometric morphometrics) using male genital morphology for this purpose. Five species complexes of Lepidoptera Tortricidae and Geometridae were analysed, each having two to four species, and each species being readily identifiable by their wing patterns. The between‐species differences were visualized with thin‐plate spline deformation grids using average landmark configurations of each species. The results from exploratory and confirmatory statistics of geometric data in a taxonomic context were compared. Morphometrics provided more accurate identification than subjective visual differentiation and the best result was achieved by combined size and shape data. Furthermore, species delimitation using exploratory statistics of landmark data is often problematic because of overlap in scatters. We conclude that geometric morphometrics provides a powerful way to search for differences between taxa and serves as an objective, useful and novel way to visualize morphological variation in shape in insect taxonomy. We recommend more extensive use of geometric morphometric tools in taxonomy.

THE PREDICTION OF FAT PERCENTAGE IN THE LONGISSIMUS DORSI MUSCLE IN FED BEEF USING NUCLEAR MAGNETIC RESONANCE

ABSTRACT The objective of this study was to develop predictive equations to estimate the percentage of crude fat in beef longissimus dorsi muscles from subjective marbling evaluations or objective visual evaluations. Forty beef ribs and loins representing four U.S. Department of Agriculture (USDA) quality grades – USDA Prime (n = 10); USDA Premium Choice [upper 2/3 of Choice] (n = 10); USDA Low Choice (n = 10); and USDA Select (n = 10) – were fabricated into 3.17‐cm transverse slices for subjective and objective visual evaluations. Additional 0.63‐cm slices were removed for crude fat analysis. Mean crude fat percentages were as follows: USDA Prime = 11.90, USDA Premium Choice = 9.78, USDA Low Choice = 7.08 and USDA Select = 5.29. Regression analysis indicated that crude fat can be estimated with the following equation using a subjective marbling score: = (–0.58) + 0.17 × marbling score (R2 = 0.55); furthermore, crude fat = (3.47) + 0.73 × objective visual percent fat (R2 = 0.54).

Single-Blind Evaluation of Post-Tonsillectomy Pain Treatment with an Eutectic Mixture of Local Anesthetics

The efficacy of an eutectic mixture of local anesthetics (lidocaine and prilocaine) in ointment for topical use (EMLA) in the treatment of post-tonsillectomy pain in both pediatric and adult patients was evaluated. Ninety patients affected by recurring tonsillitis and refractory to antibiotic therapy were submitted to tonsillectomy by dissection. Of the ninety patients 45 random subjects received topical treatment and the remaining 45 did not receive any treatment in the emptied tonsillar compartment. Pain intensity was evaluated at 3, 6 and 9 hours after surgical treatment using a visual subjective evaluation scale (VAS for adults and FES for children). The percentage of adults who referred pain of minor intensity was greater in the treated patients than in the controls: 51.4% vs 14.3% (p&lt;0.001) at the third hour, 71.4% vs 2.9% (p&lt;0.001) at the sixth hour and 88.6% vs 14.3% (p&lt;0.001) at the ninth hour. In the pediatric groups, 80.0% of the treated subjects referred moderate pain at the third hour, compared to 40% of the control group. At the ninth hour all children (100%) referred moderate pain compared to 20% of the control group. No treated patient required analgesic therapy during the post-operative observation period. The application of EMLA in the compartment after tonsillectomy determined an extremely significant reduction of postoperative pain in terms of intensity and duration.

A quantitative measure for the uniformity of fixation density: The Voronoi method

In order to characterize the uniformity of fixation density, we propose aquantitative measure based on Voronoi diagrams, in which cells are defined around fixation locations. We examined how normalized cell size distributions are related to homogeneous and inhomogeneous fixation densities. Two possible measures for use with the Voronoi method are discussed. Both show good correlation with subjective visual evaluations of the uniformity of fixation densities. Not only are these measures objective and quantitative, they also have a simple intuitive meaning: They may be thought of as reflecting the clustering of fixations.

Effects of staining and bleaching on color change of dental composite resins

Discoloration of resin-based composites by colored solutions is a common problem. The use of bleaching agents for discolored natural teeth is becoming increasingly popular. It is not clear if bleaching agents can remove the stain from composite resins. The purpose of this study was to investigate the effects of 2 staining solutions and 3 bleaching systems on the color changes of 2 dental composite resins. Forty-five disk-shaped specimens (9 x 2.5 mm) of each of 2 composite resins, Filtek Supreme (FS) and Esthet X (EX), were prepared. The specimens were then divided into 3 groups of 15 specimens each and immersed in 2 staining solutions (coffee or red wine) or distilled water (control) for 3 hours daily over a 40-day test period. The 3 groups were then divided into 3 subgroups (n = 5), and 3 bleaching agents (Crest Night Effects, Colgate Simply White Night, or Opalescence Quick) were applied to the surface of the specimens over a 14-day period. Color of the specimens was measured with a spectrophotometer using CIELAB color space relative to CIE standard illuminant D55 at baseline, after staining, and after bleaching. The color differences (deltaE(ab)*) between the 3 measurements were calculated. The value deltaE(ab)* = 3.3 was used as an acceptable value in subjective visual evaluations. Analysis of variance and nonparametric analysis (Kruskal-Wallis test and Mann-Whitney test) were used to analyze the data. After staining, FS had more color change than EX and was more affected by the wine solution. After bleaching, the color of both EX and FS specimens returned to the baseline. The color differences between bleaching and baseline were less than value deltaE(ab)* = 3.3 for all groups. The nanocomposite (FS) changed color more than the microhybrid composite (EX) as a result of staining in coffee or red wine solutions. After bleaching, discoloration was removed completely from the composite resins tested.

「錯視現象」が頭部ＣＴ画像の診断能に及ぼす影響―明るさ知覚に関する錯視現象の画像診断への関与―

Even if the visual impression of the photographic density of the brain in head CT images is shown as physically the same, it is known that optical illusions of lightness perception (assimilation, contrast, picture frame effect, etc.) occur and that practical density can be observed psychologically differently, according to differences in the color of the skull and background, and differences in cases (differences in picture pattern). Therefore, in this study, in order to clarify the influence of optical illusion on detectability in diagnosis, the author attempted to compare detectability in four sample cases, consisting of acute cerebral infarction (1), acute epidural hematoma (1), and chronic subdural hematoma (2), using visual subjective evaluation. In the case of acute cerebral infarction, there was no significant difference in detectability between the original image and the virtual images. Further, it clarified that the original head CT image (acute epidural hematoma) with the high-density hematoma recognized at the marginal limited part of the brain was inferior to virtual images in detectability, while it clarified that the original head CT image (chronic subdural hematoma) with the low-density hematoma was superior to virtual images in detectability, because of visual psychological emphasis on the difference of the film contrast between the hematoma and white skull.

Use of bisphosphonates can dramatically improve pain in advanced hormone-refractory prostate cancer patients

Approximately 85% of patients who die from prostate cancer present the spread of bone metastases. Even though the radiological appearance of such metastases is osteoblastic, it is now known that these lesions coexist in their microenvironment with blastic and lytic lesions. The process always begins with bone lysis by osteoclast proliferation, paralleling nearby bone deposition. The treatment options are palliative and have poor clinical response with short-lived improvement. We have studied the clinical effect of bisphosphonates (clodronate) in the treatment of skeletal complications from prostate cancer. In an open prospective study, 58 patients with hormone-refractory prostate cancer with bone metastases were assessed from November 2000 to September 2003. The mean age was 70.3 y (range: 51-87 y). Bone scintigraphy, plain X-ray, assaying of prostate-specific antigen (PSA) and biochemical tests were requested before and following treatment. Patients were previously and subsequently assessed using the visual pain scale (0-10) and Karnofsky's index after the first and second intravenous (i.v.) infusions (administration of i.v. clodronate every 28 days) and every 4-6 months thereafter. Student's t-test was used for statistical analysis. A total of 53 patients (91.4%) showed improvement after the first and/or second cycle, which persisted for at least 4 months (average 6.3 months). The averages on the visual pain scale improved from 7.4 (range: 2-8) to 2.4 (0-7) and on Karnofsky's index from 43 (32-58) to 73 (50-82). The radiological appearance of the metastases improved in 27 patients (46.5%) and there were few relapses (six patients; 10.3%). Clodronate was effective in the treatment of skeletal complications from prostate cancer. There was an objective response in 91.4% of treated patients, with a marked improvement in the subjective visual pain scale evaluation as well as on Karnofsky's index, with low side effects.

Visual impression of photographic density of lung, thorax, and mediastinum in chest CT images: verification of optical illusion by visual subjective evaluation of medical images

Even if the brain in head CT images is shown physically at the same photographic density, optical illusion (assimilation, contrast, etc.) occurs and practical density can be observed psychologically differently. Therefore, in this study, in order to clarify the visual characteristics of lightness perception in chest CT images, I attempted to compare it with the visual impression of the photographic density of lung, thorax, and mediastinum, using visual subjective evaluation. The results were as follows. (1) In the case of the preferential window setting of lung, both thorax and mediastinum, which surround the lung, are recognized as a wide white frame with contrast, and the photographic density of the lung psychologically seems blackish as a result of the "picture frame effect". (2) In the case of the preferential window setting of thorax and mediastinum, the visual impression of the photographic density varies among observers. (3) In the case of the virtual double window setting of lung, thorax, and mediastinum, under the influence of both high-density and low-density areas in the radiological anatomy of thorax and mediastinum, the photographic density of lung psychologically appears whitish as a result of the "grouping theories" of lightness computation. Further, under the influence of original gray lung, the photographic densities of thorax and mediastinum psychologically appear to be whitish.

Evaluation of myocardial perfusion with grey-scale ultra-harmonic and multiple-frame triggering. The need for quantification

Background and Objective: Contrast echocardiography has been recently introduced as a new technique for evaluating myocardial perfusion in a qualitative basis. The objective of this study was to test whether a visual subjective evaluation of myocardial perfusion by myocardial contrast echocardiography adequately matches the data obtained with an off-line quantification of myocardial perfusion. Methods: Sixty-one myocardial segments were evaluated by myocardial contrast echocardiography with Ultra-harmonic and Multiframe Triggering in 11 patients 3-7 days after an anterior myocardial infarction, using SH-U 563A (Levovistâ, Schering AG, Berlin, Germany) as contrast agent. Myocardial perfusion was classified as grade 1 (absent), 2 (patchy or incomplete) and 3 (complete) in each segment. The quantitative analysis was performed off-line by a different investigator blinded to the qualitative evaluation, using a commercially available software. The quantitative data on grey-scale obtained were compared between grade 1, 2 and 3 segments. Results: Of the 61 segments, 45 (73.8%) were classified as grade 3, whereas the remaining 16 (26.2%) were considered to be abnormally perfused (grade 2: n=12, 19.6%; grade 1: n=4, 6.6%). Segments with grade 1 perfusion had a significantly higher grey-scale value (123.6±41.3 vs. 70.1±34.3, p=0.004). However, there were no significant differences between segments with perfusion grade 2 and 3 (76.8±33.2 vs. 68.3±34.8, p=0.452). Conclusion: Qualitative assessment of myocardial perfusion by Ultra-harmonic and Multiframe Triggering is of limited value, since only myocardial segments with absent perfusion may be reliably identified. This findings support the need of quantification in the evaluation of myocardial perfusion by contrast echocardiography.

Visual impression of photographic density of the brain in head CT images: verification of optical illusion by visual subjective evaluation of medical images

The purpose of this study was to clarify the psychological influence of optical illusion on the photographic density of the brain in head CT images. In this study, I attempted to compare the visual impression of photographic density of the brain using visual subjective evaluation according to observers and cases. The results showed that the visual impression of photographic density of the brain at cerebellar CT depended upon the mode of appearance of the optical illusion (assimilation, contrast, picture frame effect, etc.). In addition, that impression varied according to differences in cases (difference of picture pattern) and observers. Further, it was clarified that the optical illusion had been observed according to individual differences among observers. Moreover, cases could be divided into two main classes. Some cases were recognized as a kind of outline figure (the white skull was recognized as an induction color, and the brain as a non-induction color) with assimilation, and consequently the brain seemed whitish. In other cases (the white skull was not recognized as an induction color) in which there was no assimilation, the brain did not seem whitish. This was considered to have occurred because the degree of influence of the skull on the visual impression of the photographic density of the brain had changed according to differences in cases.

A measure of quality for evaluating methods of segmentation and edge detection

A new measure of quality is proposed for evaluating the performance of available methods of image segmentation and edge detection. The technique is intended for the evaluation of low error results and features an objective assessment of discrepancy with respect to the theoretical edge, in tandem with subjective visual evaluation using both the neighbourhood and error-interaction criteria. The proposed mathematical model is extremely simple, even from the perspective of computational execution. A training of the measure has been put in practice, which uses visual evaluation of a set of error patterns by a team of observers. Encouraging results were obtained for a selection of test images, especially in relation to other recently proposed and/or currently employed quality measures.

Usefulness of Combined Fat- and Fluid-Suppressed SPIR-FLAIR Images in Optic Neuritis: Comparison with Fat-Suppressed SPIR or STIR Images or STIR images

PURPOSE: To compare the usefulness of combined fat- and fluid-suppressed selective partial inversion recovery-fluid attenuated inversion recovery(SPIR-FLAIR) images in the detection of high signal intensity of the optic nerve in optic neuritis with that of fat-suppressed selective partial inversion recovery(SPIR) or short inversion time inversion recovery(STIR) images. MATERIALS AND METHODS: Two radiologists independently analyzed randomly mixed MR images of 16 lesions in 14 patients (M:F=7:7; mean age, 40years) in whom optic neuritis had been clinically diagnosed. All subjects underwent both SPIR-FLAIR and fat-suppressed SPIR or STIR imaging, in a blind fashion. In order to evaluate the optic nerve, coronal images perpendicular to its long axis were obtained. The detection rate of high signal intensity of the optic nerve, the radiologists' preferred imaging sequences, and intersubject consistency of detection were evaluated. 'High signal intensity' was defined as the subjective visual evaluation of increased signal intensity compared with that of the contralateral optic nerve or that of white matter. RESULTS: The mean detection rate of high signal intensity of the optic nerve was 90% for combined fat- and fluid-suppressed SPIR-FLAIR images, and 59% for fat-suppressed SPIR or STIR images. In all cases in which the signal intensity observed on SPIR-FLAIR images was normal, that on fat-suppressed SPIR or STIR images was also normal. The radiologists preferred the contrast properties of SPIR-FLAIR to those of fat-suppressed SPIR or STIR images. CONCLUSION: In the diagnosis of optic neuritis using MRI, combined fat- and fluid-suppressed SPIR-FLAIR images were more useful for the detection of high signal intensity of the optic nerve than fat-suppressed SPIR or STIR images. For the evaluation of optic neuritis, combined fat- and fluid-suppressed SPIR-FLAIR imaging is superior to fat-suppressed SPIR or STIR imaging.