Lesion localization and surgical resection for non-palpable breast cancer / 中华肿瘤杂志

<p><b>OBJECTIVE</b>To investigate the methods of lesion localization and surgical treatment for non-palpable breast cancer, presented with only small calcification lesion on the images.</p><p><b>METHODS</b>From November 2003 to August 2007, 61 patients with non-palpable lesion were finally pathologically diagnosed as early breast cancer (T1-2N0M0), based on the small calcification lesions shown by full field digital mammography (FFDM) through molybdenum target, and the rich blood supply shown by type-B ultrasonic examination. Accurate lesion-localization prior to surgical resection was conducted, and sample re-examination by FFDM was done after resection. Patients with single lesion underwent breast-conserving surgery, precise excision with the aid of image-guided wire localization, and stage I breast reconstruction was performed simultaneously using wide-based gland-tissue flap. Patients with multiple lesions received modified radical mastectomy.</p><p><b>RESULTS</b>Among the 50 patients treated with breast-conserving surgery, the accuracy of localization for lesions was 100% (50/50), and all lesions were excised completely with a negative margin proven by FFDM re-examination and pathological examination. The superior rate of mammaplasty was 86.0% (43/50) according to JCRT criteria, with a compliance difference of 1.5 cm. Modified radical mastectomy was performed in 11 patients. The follow-up period in this series was from 6 to 58 months with a mean follow-up time of 39 months. Distant metastases were detected in only one patient and local recurrence was not observed yet.</p><p><b>CONCLUSION</b>Lesion localization by FFDM in patients with non-palpable breast cancer is accurate and practical. In patients with single lesion, breast-conserving resection followed by synchronous stage I breast reconstruction with wide-based gland-tissue flap is appropriate.</p>

Evaluation of gastric carcinoma with CT three-dimensional imaging compared with conventional barium study / 中华放射学杂志

0.05), but the lesion detection sensitivity of SSD and Raysum display were lower than that of UGI(? 2=4.17,7.11, and 5.14,4.17, P0.05). Excess fluid remained in the stomach and patient respiratory movement during breath holds were the reasons causing severe artifacts (6.1%) that influenced the diagnostic evaluation. Conclusion The performance of CTVG was equivalent to UGI in the detection of advanced gastric carcinoma and superior to UGI in the Borrmann′s classification. CTVG has potential in the detection of early gastric carcinoma. The value of SSD and Raysum display was limited in the evaluation of gastric carcinoma when used alone clinically.

Evaluation of image quality of CT virtual endoscopy / 中华放射学杂志

Objective To analyze and evaluate the image quality and the impact factors of CT virtual endoscopy (CTVE) in the application of various body parts. Methods CTVE images from 143 patients were classified into 2 types according to the chosen threshold: low attenuation lumen (including natural air filled and artificially air filled lumina) and high attenuation lumen. A 4 point scale was used to evaluate the differences of diagnostic confidence, artifacts, and overall image quality rating between both kinds of lumina. Results There was no statistically significant differences in diagnostic confidence of CTVE images among 3 kinds of lumina, but the image quality of CTVE of natural air filled lumen was better than that of artificially air filled lumen ( P 0.05). There was a significant difference in artifact rating of CTVE images among three kinds of lumina with the most artifacts seen on images of high attenuation lumen. Common artifacts observed on the CTVE images were attributed to patient motion artifacts such as respiration, inappropriate choice of spiral CT scanning parameters, and improper modification of thresholds. Conclusion The threshold setting was one of the most important factors impacting the image quality of CTVE images, and most artifacts seen on the CTVE images were easily recognized on the basis of their characteristic appearance, and did not influence the diagnostic information heavily.