We propose a new method that employs transfer learning techniques to effectively correct sampling selection errors introduced by sparse annotations during supervised learning for automated tumor segmentation. The practicality of current learning-based automated tissue classification approaches is severely impeded by their dependency on manually segmented training databases that need to be recreated for each scenario of application, site, or acquisition setup. The comprehensive annotation of reference datasets can be highly labor-intensive, complex, and error-prone. The proposed method derives high-quality classifiers for the different tissue classes from sparse and unambiguous annotations and employs domain adaptation techniques for effectively correcting sampling selection errors introduced by the sparse sampling. The new approach is validated on labeled, multi-modal MR images of 19 patients with malignant gliomas and by comparative analysis on the BraTS 2013 challenge data sets. Compared to training on fully labeled data, we reduced the time for labeling and training by a factor greater than 70 and 180 respectively without sacrificing accuracy. This dramatically eases the establishment and constant extension of large annotated databases in various scenarios and imaging setups and thus represents an important step towards practical applicability of learning-based approaches in tissue classification.

Pathology of the external and middle ear is the most frequent reason to prescribe antibiotics and perform surgery in children and young adults. In the majority of cases imaging studies are not performed; the need for imaging comes when complications are suspected or when treatment is not effective. This paper discusses indications for temporal bone imaging studies and presents the most frequent pathological conditions, together with differential diagnosis, clinical symptoms and methods of treatment. This pictorial review describes major external and middle ear diseases, with special regard to clinical findings and appropriate reporting. Apprehension of the complete clinical picture, together with imaging clues, narrows differential diagnosis and helps avoid mistakes. Evaluation of temporal bone pathologies requires close cooperation between the clinician and radiologist.

Myositis ossificans is localized inflammatory process affecting skeletal muscles. Very rarely it can affect one of the neck muscles and present as a neck tumor, it can be misdiagnosed as the clinical, radiological and histological examinations can mimic a sarcoma. We report a 29 year old female patient with neck tumor suspected to be a sarcoma who underwent full diagnostics imaging and open bipsy with histopatological examination, afterwards surgical excision was performed. The aim of this study was to present the differential diagnosis based on diagnostics imaging between MO and malignant tumors, such as parosteal sarcoma, synovial sarcoma and malignant fibrous histiocytoma.