Zusammenfassung
The data collection contains 43 optical coherence tomography (OCT) volumes from both healthy and pathological middle ears of 29 subjects recorded with a non-invasive endoscopic OCT device applicable in vivo. Due to the shadowing of preceding structures interpreting such OCT volumes needs expertise and is time-consuming. Nowadays, deep neural networks have emerged to facilitate this process regarding segmentation, classification and registration. Hence, the dataset offers semantic segmentations of five crucial anatomical structures (tympanic membrane, malleus, incus, stapes and promontory), and sparse landmarks delineating the salient features of the structures additionally. The complete dataset provides the possibility to develop and train own networks and algorithms for the evaluation of middle ear OCT volumes.