[Preoperative marking and biopsy of nonpalpable breast lesions with a guidance system for the open Magnetom]. / Präoperative Markierung und Biopsie nicht palpabler Mammaläsionen mit einer Zieleinrichtung am Magnetom Open.

PURPOSE: MRI-guided interventional maneuvers in the breast, when closed MRI scanners are employed are relatively complex and time consuming. The purpose of our investigations was to develop a special device for the localization and biopsy of breast lesions using an open low-field MRI Scanner (Magnetom Open, Siemens, Erlangen, Germany) permitting free access to the patient. MATERIALS AND METHODS: Due to its particular material and construction characteristics, a newly developed device for localization and biopsy of breast lesions is appropriate to remain within the magnetic field during the examination without degrading image quality. We performed 125 tests in a phantom and 31 investigations in patients including 26 wire-localizations and 5 biopsies. RESULTS: The interventional maneuvers in the phantom could be performed very precisely. In all 125 phantom studies, the needle (14 G coaxial biopsy cannula, Bard) was positioned close to the simulated lesion. In 25 out of 26 patients the needle was positioned within a distance less than 5 mm from the lesion. When using the 14 G coaxial biopsy cannula (Bard, Karlsruhe, Germany) the needle tip was found adjacent next to the lesion in all 12 cases (7 wire-localizations, 5 biopsies). CONCLUSIONS: Our results indicate that our device allows very precise preoperative localization of breast lesions within the Magnetom Open. Using MRI-compatible, large-core needles, biopsy under direct MRI control resulted in good results.

[Computer-assisted analysis of mammograms in routine clinical diagnosis]. / Computerassistierte Analyse von Mammographien in der klinischen Routinediagnostik.

PURPOSE: Computer-aided diagnosis in mammography is a topic many study groups have been concerned with since the first presentation of a system for computer-aided interpretation in 1967. Currently, there is only one system available for clinical use in mammaography, the CAD-System Image Checker (R2 Technology). The purpose of our prospective study was to evaluate whether the integration of the CAD-system into the routine of a radiological breast diagnosis unit is feasible. RESULTS: After the installation of the CAD-system, 300 patients with 1110 mammograms were included for evaluation in the present study. In 54 of these cases histological examination was indicated due to suspect criteria on conventional mammography. In 39 of 54 cases (72.2%) malignancy could be proven histologically. The CAD-system marked 82.1% of the histologically verified carcinomas correctly. 94.3% of all 1797 marks made by the CAD-system indicated normal or benign structures. Routinely performed CAD analysis prolonged patients waiting time by about 15 min because the marks of the CAD system had to be interpreted in addition to the routine diagnostic investigations. CONCLUSION: Our experience with the use of the CAD-system in daily routine showed that CAD analysis can easily be integrated into a preexisting mammography unit. However, the diagnostic benefit is not yet clearly established. Since the rate of false negative marks by the CAD-system Image Checker is still high, the results of CAD analysis must be checked and corrected by an observer well experienced in mammography reading.

[MR mammography. Preoperative marking of non-palpable breast lesions with the Magnetom open at 0.2 T]. / MR-Mammographie. Präoperative Markierung nichtpalpabler Mammaläsionen am Magnetom Open bei 0.2 T.

PURPOSE: To answer the following questions: whether reliable detection of lesions is possible in low-field-MRI-system (Magnetom Open 0.2 T) equipped with currently available hard- and software components in comparison to high field system (Magnetom Impact 1.0 T). Furthermore, whether localization of lesions suspect in MR-mammography can be realized in MR system of low field (Magnetom Open 0.2 T). PATIENTS AND METHODS: In 11 patients, suspect lesions were diagnosed in diagnostic MR-mammography acquired with high field system (Magnetom Impact 1.0 T) and were compared to low field MR-mammographies of 0.2 T (Magnetom Open 0.2 T). In six of the 11 patients a suspect lesion was localized using wire marking. RESULTS: All lesions considered suspect in diagnostic MR-mammography (Magnetom Impact 1.0 T) were also clearly identified in the 0.2 T system (Magnetom Open). In six cases wire marking was performed without any complications and with an accuracy of < or = 0.5 cm distance to the lesion. CONCLUSION: Although studies in the 0.2 T system clearly showed inferior SNR (34.6 vs. 83.1) and CNR (14.6 vs. 43.5) compared to studies with the high field system, all lesions considered suspect in diagnostic MR-mammography were reliably identified also in 0.2 T studies. Due to its open construction permitting permanent access to the breast and due to sufficient image quality, the Magnetom Open is suitable for interventions on the breast.