2 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 2 / 2
Yayın Automated cell nucleus detection for large-volume electron microscopy of neural tissue(IEEE, 2014-04-29) Tek, Faik Boray; Kroeger, Thorben; Hamprecht, Fred A.; Mikula, ShawnVolumetric electron microscopy techniques, such as serial block-face electron microscopy (SBEM), generate massive amounts of image data that are used for reconstructing neural circuits. Typically, this requires time-intensive manual annotation of cells and their connections. To facilitate this analysis, we study the problem of automated detection of cell nuclei in a new SBEM dataset that contains cerebral cortex, white matter, and striatum from an adult mouse brain. The dataset was manually annotated to identify the locations of all 3309 cell nuclei in the volume. We make both dataset and annotations available here. Using a hybrid approach that combines interactive learning, morphological processing, and object level feature classification, we demonstrate automated detection of cell nuclei at 92.4% recall and 95.1% precision. These algorithms are not RAM-limited and can scale to arbitrarily large datasets.Yayın Convolutional attention network for MRI-based Alzheimer's disease classification and its interpretability analysis(IEEE, 2021-09-17) Türkan, Yasemin; Tek, Faik BorayNeuroimaging techniques, such as Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET), help to identify Alzheimer's disease (AD). These techniques generate large-scale, high-dimensional, multimodal neuroimaging data, which is time-consuming and difficult to interpret and classify. Therefore, interest in deep learning approaches for the classification of 3D structural MRI brain scans has grown rapidly. In this research study, we improved the 3D VGG model proposed by Korolev et al. [2]. We increased the filters in the 3D convolutional layers and then added an attention mechanism for better classification. We compared the performance of the proposed approaches for the classification of Alzheimer's disease versus mild cognitive impairments and normal cohorts on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. We observed that both the accuracy and area under curve results improved with the proposed models. However, deep neural networks are black boxes that produce predictions that require further explanation for medical usage. We compared the 3D-data interpretation capabilities of the proposed models using four different interpretability methods: Occlusion, 3D Ultrametric Contour Map, 3D Gradient-Weighted Class Activation Mapping, and SHapley Additive explanations (SHAP). We observed that explanation results differed in different network models and data classes.
