Cognition Network Technology prototype of a CAD system for mammography to assist radiologists by finding similar cases in a reference database.

PURPOSE: We present a new approach for computer-aided detection and diagnosis in mammography based on Cognition Network Technology (CNT). Originally designed for image processing, CNT has been extended to also perform context- and knowledge-driven analysis of tabular data. For the first time using this technology, an application was created and evaluated for fully automatic searching of patient cases from a reference database of verified findings. The application aims to support radiologists in providing cases of similarity and relevance to a given query case. It adopts an extensible and knowledge-driven concept as a similarity measure. METHODS: As a preprocessing step, all input images from more than 400 patients were fully automatically segmented and the resulting objects classified--this includes the complete breast shape, the position of the mammilla, the pectoral muscle, and various potential candidate objects for suspicious mass lesions. For the similarity search, collections of object properties and metadata from many patients were combined into a single table analysis project. Extended CNT allows for a convenient implementation of knowledge-based structures, for example, by meaningfully linking detected objects in different breast views that might represent identical lesions. Objects from alternative segmentation methods are also be considered, so as to collectively become a sufficient set of base-objects for identifying suspicious mass lesions. RESULTS: For 80% of 112 patient cases with suspicious lesions, the system correctly identified at least one corresponding mass lesion as an object of interest. In this database, consisting of 1,024 images from a total of 303 patients, an average of 0.66 false-positive objects per image were detected. An additional testing database contained 480 images from 120 patients, 15 of whom were annotated with suspicious mass lesions. Here, 47% (7 out of 15) of these were detected automatically with 1.13 false-positive objects per image. A diagnosis is predicted for each patient case by applying a majority vote from the reference findings of the ten most similar cases. Two separate evaluation scenarios suggest a fraction of correct predictions of respectively 79 and 76%. CONCLUSION: Cognition Network Technology was extended to process table data, making it possible to access and relate records from different images and non-image sources, such as demographic patient data or parameters from clinical examinations. A prototypal application enables efficient searching of a patient and image database for similar patient cases. Using concepts of knowledge-driven configuration and flexible extension, the application illustrates a path to a new generation of future CAD systems.

Magnetic resonance-guided, vacuum-assisted breast biopsy: results from a European multicenter study of 538 lesions.

BACKGROUND: The objective of this study was to determine the accuracy, reproducibility, and clinical value of magnetic resonance (MR)-guided, vacuum-assisted breast biopsy (MR-VAB) in a prospective, multicenter study. METHODS: In 5 European centers, MR-VAB was performed or attempted on 538 suspicious lesions that were visible or could targeted only by MR imaging (MRI). Verification of malignant or borderline lesions included reexcision of the biopsy cavity. Benign biopsy results were verified by retrospective correlation of histology with preinterventional and postinterventional MRI studies. Follow-up of 24-48 months (median, 32 months) was available for 491 of 538 patients. RESULTS: MR-VAB was unsuccessful or was not completed in 21 of 538 patients, for which an immediate repeat biopsy was recommended. Five hundred seventeen of 538 performed VAB procedures (96%) were successful. Histology yielded 138 (27%) malignancies, 17 (3%) atypical ductal hyperplasias, and 362 (70%) benign entities. No false-negative diagnoses occurred among the 517 successful MR-VAB procedures. The positive predictive value of VAB depended on patient preselection, which differed according to the indication for the initial MRI study. CONCLUSIONS: The results of this study indicated that MR-VAB offers excellent accuracy. Small lesion size did not prove to be a limitation.

Diagnosis of osteoporosis: visual assessment on conventional versus digital radiographs.

In many radiological departments conventional radiography has been replaced by digital radiography. Therefore, the purpose of this study was to analyze the visual detection of osteopenia/osteoporosis with both digital and conventional radiographs. In 286 patients we retrospectively evaluated radiographs of the lumbar spine in two planes. One hundred twenty-eight patients had conventional and 158 patients had digital radiographs. Patients with pre-existing vertebral fractures were excluded. Four experienced musculoskeletal radiologists blinded to the values of DXA and to the patients' ages assessed independently from each other whether the bone density of the lumbar spines was normal or decreased. The results of dual X-ray absorptiometry served as the standard of reference. The threshold value for the diagnosis of osteopenia was a T-score less than -1 SD according to the WHO classification of osteoporosis. Sensitivity/specificity was 86%/36% for conventional and 72%/47% for digital radiographs. The overall diagnostic accuracy was 68% for conventional and 64% for digital radiographs. Eighty percent of the patients with osteopenia and 96% of the patients with osteoporosis were correctly assessed as true positive on conventional radiographs and 65% (osteopenia) and 82% (osteoporosis) on digital radiographs. Interobserver agreement was markedly lower for digital (35%) than for conventional radiographs (73%). However, the differences were not statistically significant. There is no major difference in diagnostic accuracy in the assessment of osteopenia/osteoporosis using digital and conventional radiographs, respectively. However, the high interobserver variance on digital radiographs indicates that visual assessment of osteoporosis/osteopenia is problematic, which may be due to image processing and postprocessing algorithms that manipulate the visual aspect of bone density.