[en] To evaluate the clinical significance of the screening mammography in the detection of the breast diseases, especially breast carcinoma. We analyzed 1,800 cases of mammography retrospectively. The mammography was done as a part of routine check in Health Counselling Center, Korea University Medical Center, during 9 months from November 1993 to July 1994. The age range was from 23 years to 76 years, mean 49.8 years, and the largest age group was 6th decade (31.4%). According to the mammographic findings, we divided the subjects into three groups; normal group, abnormal group in need of follow-up study, abnormal group requiring biopsy. On mammography, the normal group consisted of 1,534 cases (85%), and the abnormal group consisted of 266 cases (15%). The abnormal findings were benign-looking calcification (n = 140), fibroadeno ma (n = 29), fibrocystic changes (n = 27), cyst (n = 23), malignant lesion (n = 15) lipoma (n = 7), and others. In four of 15 cases, which were suspected to be malignant on mammograms, breast carcinoma was confirmed pathologically. In four cases of breast carcinoma, one was under 40 and the other 3 were over 50 years of age. All of the breast cancers were under 3 cm in size, and the mammographic findings of breast cancer included spiculated margin (n = 3), parenchymal distortion (n = 3), malignant calcification (n = 2) and enlarged axillary node (n = 1). Screening mammogram is helpful for early detection of non-palpable breast cancer, especially for women over 50 years of age

[en] To determine the anatomic variations that can lead to optic nerve damage during the sugical treatment of posterior paranasal sinus lesions two hundred optic nerves of 100 persons were examined using ostiomeatal unit CT(OMU CT). The anatomical features of this nerve and posterior paranasal sinuses were classified into four types:the optic nerve adjacent to the sphenoid sinus without indentation of the sinus wall (type 1);the optic nerve adjacent to the sphenoid sinus, causing indentation of the sinus wall (type 2);the optic nerve passing through the sphenoid sinus (type 3);and the optic nerve adjacent to the sphenoid sinus and posterior ethmoid sinus (type 4). Bony dehiscence around the optic nerve and pneumatization of the anterior clinoid process were also evaluated. The anatomical classification of the optic nerve and posterior paranasal sinuses was as follows:type 1, 1326(66%); type 2, 60(30%); type 3, 6(3%), and type 4, 2(1%). Bony dehiscence around the optic nerve had developed in 58 cases (29%) and pneumatization of the anterior clinoid process in 13(6.5%). These conditions were most common in type 3 optic nerve, and second most common in type 2. The 2 and 3 optic nerves, bony dehiscence around the optic nerve and pneumatization of the anterior clinoid process are the anatomic variations that can lead to optic nerve damage during the surgical treatment of posterior paranasal sinus lesions. To prevent optic nerve damage, these factors should be carefully evaluated by OMU CT

[en] To evaluate the significance of focal bulging of hepatic contour on the differentiation of subcapsular hemangioma from hepatocellular carcinoma. Sonographic findings of 19 subcapsular hemangiomas and 19 HCCs were retrospectively evaluated for the presence of focal bulging of hepatic contour. The hemangioma group consisted of ten male and nine female patients aged 26∼60 (mean, 45.8) years. The HCC group consisted of eleven male and eight female patients aged 43∼72 (mean, 56.1) years. Eighteen hemangiomas were located in the right lobe and one in the left lobe. Fourteen HCCs were located in the right lobe and five in the left lobe. The diameter of hemangiomas was ranged 2.0∼9.8 (mean, 4.96) cm and the diameter of HCCs was ranged 2.2∼8.5 (mean, 4.97) cm. All (100%) hemangiomas showed no bulging of hepatic contour and 89.5% (17 / 19) of HCCs showed focal bulging of hepatic contour. US demonstration of focal bulging of hepatic contour is useful in the differential diagnosis of subcapsular hemangioma from hepatocellular carcinoma

[en] To evaluate contrast enhancement patterns of hepatomas (HCCs) on arterial dominant, portal dominant, and delayed phase of CT scan by using double spiral CT. Using double-spiral CT, three-phase dynamic liver scan was performed on 27 patients with 45 nodular HCC lesions. Non-ionic contrast medium (100-120ml) was injected intravenously with an automatic injector at the rate of 2-4 ml/sec. CT scans were obtained at 25-30 sec (arterial dominant phase). 60 sec (portal dominant phase), and 5 min (delayed phase) after administration of the contrast medium. The tumor were divided into two groups according to the diameter(<3 cm and ≥3 cm). In each group, contrast enhancement patterns of HCCs were classified as follows: high, central high, peripheral high, iso, low, or mixed attenuation. The incidence and contrast enhancement patterns of thin peripheral rims(psevdocapsules) were also analyzed. There were 23 lesions with a diameter less than 3 cm, while 22 lesions were 3 cm or larger. On the arterial dominant phase, tumors smaller than 3 cm showed high(74%), iso(17%), mixed(4%), and peripheral high(4%) attenuation, while the attenuation of tumors larger than 3 cm was high(73%), mixed(9%), iso(17%), and low(9%). On the portal dominant phase, tumors smaller than 3 cm were most commonly iso(43%) or high(35%), while tumors larger than 3 cm were low(65%) or iso(18%). On the delayed phase, tumors were most commonly low in attenuation regardless of size. A thin peripheral rim was observed in 22% of tumors smaller than 3 cm and in 64% of tumors larger than 3 cm. The rim showed iso(53%) or low(42%) attenuation on the arterial dominant phase, high(47%) or iso(47%) attenuation on the portal dominant phase, and high(100%) attenuation on the delayed phase. For the detection and diagnosis of HCCS, an understanding of their contrast enhancement pattern is useful

[en] To evaluate the utility of three phases of spiral CT in the diagnosis of stomach cancer. Between August 1994 and March 1995, thirty eight patients with stomach cancer, demonstrated on spiral CT, underwent surgery. Twenty-eight cases were advanced and ten were early. There were 27 men, and 11 women, and their average age was 52.8 years old (33-77). After ingestion of 600-700ml of water, 120-140ml of nonionic contrast material was injected intravenously. Spiral CT scanning was performed in 10mm slice thickness and of 10mm/sec table speed. Three phase image were obtained at 25sec (arterial phase), 60-65sec (venous phase) and 4min (equilibrium phase) after the start of bolus injection. On each phase, CT findings were compared with pathologic results, and tumor detectibility, depth of tumor invasion and lymph node metastasis was analysed. Thirty of the 38 carcinomas (79%) were detected on the arterial phase, 33 (81%) on the venous phase and 30 (79%) on the equilibrium phase. Depth of tumor invasion was measured accurately in 27 of 38 cases (71%) : T1-4/10 (40%), T2-8/11 (73%), T3-13/15 (87%), T4-2/2 (100%). We overstaged one case of T1 as T2 and two cases of T2 as T3, and understaged one case of T2 as T1 and two cases of T3 as T2. Among the 16 enlarged lymph nodes larger than 8mm, 13 cases were positive on pathologic examination and the sensitivity was 65%. With three-phase spiral CT scanning, we obtained 71% accuracy of depth of tumor invasion. The venous phase is most useful for tumor detection and for determining depth of tumor invasion and lymph node metastasis

[en] Solvent extraction of La, Ce, Pr, Nd and Sm from chloride solutions was studied using PC88A and Cyanex 572 as the extractants. Our results indicated that the acidity of the aqueous solution and saponification degree had some effect on the percentage extraction of metal ions. The distribution coefficient of the metal ions increased with increasing aqueous solution pH in both of PC88A and Cyanex 572. But the pH50 value was lowered by about 0.05~0.34 in PC88A. In extraction with saponified Cyanex 572 with NaOH, the distribution coefficients of the metal ions were independent of the degree of saponification, and did little to improve the separation behavior of these metal ions. With Cyanex 572 as the extractant, the separation factor of the binary La/Ce and Nd/Sm was enhanced about 5.0, but Ce/ Pr and Pr/Nd showed no difference in separation factor as compared with PC88A. The group separation factor of La/(Pr, Nd, Sm) and (Pr, Nd)/Sm was increased by 3.0~8.0, but for (La, Ce)/(Pr, Nd) and Ce/ (Pr, Nd) it was only increased by 0.2~0.7 in the Cyanex 572 system.

[en] This study designed a counter current extraction process for comparing the separation efficiencies that are obtained using Cyanex 572 and PC88A for the elements in a Pr and Nd mixed solution. This process involves multiple extraction and scrubbing processes using a mixer settler. The process parameters are calculated using an equation proposed by Xu Guangxian. The separation factor for the extraction and scrubbing process was 0.09 and 0.08 higher when using Cyanex 572 compared to the experimental values obtained using PC88A. The optimized extraction factors such as the stage numbers, feeding solution flow rate, organic phase flow rate, and scrubbing solution flow rate were calculated using the optimal extraction ratio equation. In order to derive optimum extraction conditions, the target purity and recovery ratio of Nd was set to 99.9% and 99% respectively. When using Cyanex 572 for the counter current extraction process, the total number of extraction and scrubbing stages decreased by 11 in comparison with PC88A. A comparison between the extraction, scrubbing stages, and other parameters indicate that Cyanex 572 is preferable over PC88A from the viewpoint of economic efficiency.

[en] To compare the detectability of tumor vascularity using triphasic spiral CT, angiography, CT during hepatic arteriography (CTHA) and single-level dynamic CTHA(SLD-CTHA). Seventy-nine pathologically confirmed nodular hepatocellular carcinomas(HCCs) in 77 patients were included in this study. Sixty patients were male and 17 were female, and their ages ranged from 31 to 77 (average, 57.4) years. HCCs were classified into three groups according to the size: less than 2 cm (n=20), 2-4 cm (n=32), and more than 4 cm (n=27) in diameter. If a portion of tumor demonstrated greater enhancement than surrounding liver parenchyma, vascularity was deemed to be present. Detectability by each imaging technique was compared according to size and overall. Hypervascularity was frequently detected by SLD-CTHA [90.9% (40/44)], followed by CTHA [88.0%(66/75)], angiography [80.3%(61/76)], triphasic spiral CT [72.4%(42/58)]. In the less than 2 cm group, detectability rates for triphasic spiral CT, angiography, CTHA and SLD-CTHA were 53.3%(8/15), 55.6%(10/18), 76.5%(13/17) and 87.5%(6/7), respectively, while the 2-4 cm group demonstrated corresponding figures of 71.4%(15/21), 78.1%(25/32), 84.4%(27/32) and 86.4%(19/22). In the more than 4 cm group, the rate for triphasic spiral CT was 86.4%(19/22), while for angiography, CTHA and dynamic CTHA, it was 100%. In the detection of hypervascularity of HCC, SLD-CTHA showed the highest rate, followed by CTHA, angiography, and triphasic spiral CT. In HCCs less than 4 cm in diameter, the corresponding ordering was SLD-CTHA, CTHA, angiography and triphasic spiral CT, but in HCCs of more than 4 cm, angiography, CTHA and SLD-CTHA detected hypervascularity equally well. Lesion size most affected the findings of angiography

[en] The aim of this study was to compare the accuracy of the conventional method and Bayesian analysis in estimating the probability of malignancy in solitary pulmonary nodules. On average, those using the conventional method correctly interpreted 34.7(78.9%) of 44 cases of malignant nodules and 27.7 (71%) of 39 benign nodules, while those using Bayesian analysis correctly classified 32.3 cases of malignant nodules (73.4%) and 25.3 cases of benign nodules (64.9%). Between the two teams, there was no statistically significant difference in the accuracy of qualitative assessment (p >0.5). In ROC analysis conventional interpretation and Bayesian analysis showed an accuracy of Az = 80.8 and Az = 76.7, respectively. Among 59 patients known to have smoked, the conventional method showed an accuracy of Az = 79.0 without this knowledge and A z = 72.5, respectively. Information relating to smoking history thus did not significantly improve the accuracy of prediction ( p > .05). For estimating the probability of malignancy in solitary pulmonary nodules, the accuracy of the conventional method of interpretation is not significantly difference from that of Bayesian analysis : information relating to smoking history significantly improve the accuracy of neither method. (author). 38 refs., 4 tabs., 2 figs