Digital mammography. ROC studies of the effects of pixel size and unsharp-mask filtering on the detection of subtle microcalcifications

We investigated the spatial resolution requirement and the effect of unsharp-mask filtering on the detectability of subtle microcalcifications in digital mammography. Digital images were obtained by digitizing conventional screen-film mammograms with a 0.1 X 0.1 mm2 pixel size, processed with unsharp masking, and then reconstituted on film with a Fuji image processing/simulation system (Fuji Photo Film Co., Tokyo, Japan). Twenty normal cases and 12 cases with subtle microcalcifications were included. Observer performance experiments were conducted to assess the detectability of subtle microcalcifications in the conventional, the unprocessed digital, and the unsharp-masked mammograms. The observer response data were evaluated using receiver operating characteristic (ROC) and LROC (ROC with localization) analyses. Our results indicate that digital mammograms obtained with 0.1 X 0.1 mm2 pixels provide lower detectability than the conventional screen-film mammograms. The detectability of microcalcifications in the digital mammograms is improved by unsharp-mask filtering; the processed mammograms still provide lower accuracy than the conventional mammograms, however, chiefly because of increased false-positive detection rates for the processed images at each subjective confidence level. Viewing unprocessed digital and unsharp-masked images in pairs resulted in approximately the same detectability as that obtained with the unsharp-masked images alone. However, this result may be influenced by the fact that the same limited viewing time was necessarily divided between the two images.     

The neu-oncogene product in serum and tissue of patients with breast carcinoma

BACKGROUND:: A soluble 105 kD neu-related protein is detectable in conditionedmedium from breast cancer cells expressing the neu-oncogeneproduct and in serum of nude mice bearing tumors that overexpressneu-oncogene PATIENTS AND METHODS:: In 100 patients with primary (n - 33) relapse-free (n - 6) andmetastatic (n - 61) breast carcinoma the serum levels of thesoluble new-related protein were investigated by ELISA techniques.Median age was 57 years, range 26–89 years. RESULTS:: The neu-protein serum levels were below 40 HNU/ml (human neu-antigenunit) in 72 patients and 40 or more HNU/ml in 28 patients. In30 patients with primary breast carcinoma, tested before mastectomy,all serum- neu-protein samples were negative. However, 26 of61 metastazised patients (43%) were serum-neu-protein-positive.In disseminated disease (n – 61), serum-neu-protein-positivitywas more likely to be seen in patients with visceral metastases(18/33 – 54%), than in patients with nonvisceral metastases(8/28 – 28%). Furthermore, monitoring of the serum-neu-proteinlevels reflected clinical course. For 53 patients original paraffin-embeddedtumor material was available for studying immunohistochemicalneu-protein expression. In 39/53 (73%) patients immunohistochemicaland ELISA data showed corresponding results. In 27/30 (90%)patients, from whom sera and tissue could be obtained at thesame time at primary mastectomy, results of immunohistochemistryin primary tumor and serum ELISA were negative and mutuallyconfirmatory. However, the other three patients were positivefor immunohistochemical neu-protein expression in primary tumorbut negative for serum-neu-protein expression. CONCLUSIONS:: Our results suggest that patients with advanced breast cancerand an elevated serum-neu-protein level may have a poor clinicaloutcome. This test might be a useful tool for monitoring patientswith advanced breast carcinoma, but not those with early disease.Further prospective studies are warranted to elucidate the questionof whether this test can contribute to determining prognosisand treatment strategies. breast carcinoma, c-erb-B2, HER-2, neu, oncogene, pl85     

Digital radiography of subtle pulmonary abnormalities: an ROC study of the effect of pixel size on observer performance

Forty conventional radiographs with examples of mild interstitial infiltrates and subtle pneumothoraces and 40 normal studies of the chest were selected and digitized, with pixel sizes of 1.0, 0.5, 0.2, and 0.1 mm. Observer performance tests were carried out using receiver operating characteristic analysis. Conventional radiographs and digitized images were compared. The results indicate that, in such cases, diagnostic accuracy increases significantly as the pixel size is reduced, at least to the 0.1-mm level. We conclude that, for digital systems using screen-film or similar image receptors, use of a pixel size substantially larger than 0.1 mm may result in some loss of diagnostic accuracy.     

Data compression: effect on diagnostic accuracy in digital chest radiography

High-resolution digital images make up very large data sets that are relatively slow to transmit and expensive to store. Data compression techniques are being developed to address this problem, but significant image deterioration can occur at high compression ratios. In this study, the authors evaluated a form of adaptive block cosine transform coding, a new compression technique that allows considerable compression of digital radiographs with minimal degradation of image quality. To determine the effect of data compression on diagnostic accuracy, observer tests were performed with 60 digitized chest radiographs (2,048 x 2,048 matrix, 1,024 shades of gray) containing subtle examples of pneumothorax, interstitial infiltrate, nodules, and bone lesions. Radiographs with no compression, with 25:1 compression, and with 50:1 compression ratios were presented in randomized order to 12 radiologists. The results suggest that, with this compression scheme, compression ratios as high as 25:1 may be acceptable for primary diagnosis in chest radiology.     

Arteriovenous fistulas of the lung--their clinical aspects and radiologic images

Based on the data of 15 patients the typical radiological signs of pulmonary arteriovenous fistulas, their differential diagnostic implications, clinical signs and symptoms and management are reviewed. In addition, lung function parameters of 9 patients are presented. The critical role of pulmonary angiography and sequential dynamic CT for the diagnosis is stressed, particularly in those cases in which conventional chest radiographs and tomographies fail to show characteristics abnormalities. Since pulmonary arteriovenous fistulas tend to increase in size and cause complications, they should be subjected to intermediate treatment, even if they are of no haemodynamic relevance.     

Augmentation laser for proliferative diabetic retinopathy that fails to respond to initial panretinal photocoagulation.

PURPOSE: A study was performed to determine if diabetic subjects who fail to respond to initial panretinal photocoagulation with regression of retinopathy risk factors do better with supplemental panretinal photocoagulation. METHODS: Thirty-five patients with 3 or more retinopathy risk factors who failed to respond to panretinal photocoagulation with regression to less than 3 retinopathy risk factors by 3 weeks after initial panretinal photocoagulation were prospectively randomized to augmentation laser panretinal photocoagulation (MORE) or to no additional treatment (NOMORE). RESULTS: Six months after initial treatment, the MORE group (n = 16) had regressed a mean of -0.94 retinopathy risk factors (with 95% confidence interval [CI] -1.60 to -0.26), compared with -0.21 retinopathy risk factors (95% CI -0.69 to 0.27) in the NOMORE (n = 19) group (P = 0.055). However, by 1 year, there was no statistically significant difference in the amount of regression of retinopathy risk factors with a mean decrease of -1.12 (95% CI -2.0 to -0.24) versus -1.05 retinopathy risk factors (95% CI -1.80 to -0.28) in the 2 groups, respectively. Similarly, for visual acuity, there was no difference in outcome. For all study patients, the persistence of three or more retinopathy risk factors was associated with a poorer visual result than if there was regression to less than three retinopathy risk factors. CONCLUSION: This study shows that although augmentation panretinal photocoagulation achieved faster regression of retinopathy risk factors, by 1 year, there was no difference in either mean regression of retinopathy risk factors or visual acuity between eyes treated or not treated with augmentation panretinal photocoagulation. In addition, the study shows that the persistence of 3 or more retinopathy risk factors 1 year after treatment was associated with a poorer visual result. Because sample size limited the power of the study to find small differences between groups, and because in proliferative diabetic retinopathy small differences could be important clinically, the authors do not recommend changes in current clinical practice.     

Visualization of myocardial infarction 6 h after injection of 111In-antimyosin antibodies using a blood pool subtraction technique.

111In-antimyosin antibodies are capable of visualizing myocardial infarction (MI). Because of slow blood clearance, images are usually recorded 24 or 48 h postinjection. In this pilot study, a blood pool subtraction technique, which makes it possible to visualize MI 6 h postinjection, is validated. Twenty-five patients with proven MI (16 anterior, 9 inferior) were imaged a few minutes, 6 and 24 h after an injection of 111 MBq 111In-labelled antimyosin antibodies. Three planar views are obtained each time. Using software which performs the geometric registration, the grey level normalization and the subtraction of images, the blood pool image (obtained a few minutes postinjection) is subtracted from the 6 h image. The resulting image is the blood pool corrected 6 h image. The 24 h images and the blood pool corrected 6 h images were interpreted blindly and the number of correct, incorrect and impossible MI localizations was counted. The number of correct localizations is 19/25 for the standard 24 h images and 22/25 for the blood pool corrected 6 h images. Then, with this blood pool subtraction method, it is possible to visualize MI 6 h postinjection. This has to be taken into account when discussing the role of antimyosin scintigraphy in the management of patients with MI.     

Physiological neutrophilia of pregnancy is not associated with a rise in plasma granulocyte colony-stimulating factor (G-CSF)

A patient with neutropenia and life-threatening infections secondary to T-γ lymphoproliferative disease, who did not respond to treatment with recombinant human G-CSF (filgrastim), was treated with filgrastim plus cyclosporine A (CyA). The patient achieved a good response in the absolute neutrophil count and subsequently required a dose reduction in the filgrastim. The patient was eventually discontinued from the CyA but continues on filgrastim alone. While on therapy, the large granular lymphocytes disappeared from the circulation and the beta-TCR rearrangement, which was present prior to beginning therapy, became undetectable. The patient had no significant toxicity to the CyA or the filgrastim and he has not experienced any serious infections or required hospitalization. Filgrastim has proven to be relatively nontoxic and of some benefit to patients with this disease and should probably be utilized first when treatment is necessary. However, if improvement is not observed, these findings suggest that a trial of the combination of CyA plus filgrastim may be beneficial.