Hybrid deep autoencoder with Curvature Gaussian for detection of various types of cells in bone marrow trephine biopsy images

Simultaneous Cell Detection and Classification in Bone Marrow Histology Images

“…In recent years, DL-based approaches have been shown to be very successful to analyze histological images [17]- [19], including nuclear/cell detection and classification [4], [10], [19]- [21]. Generally, DL is a hierarchical learning approach that learns high-level features from pixel intensities; these high-level feature representations are then used to differentiate objects for classification purposes.…”

“…In our previous works, the hybrid deep autoencoder (HDAE) network has been shown to be very efficient for nuclear/cell detection [21]. We have also previously designed a synchronized DL network to perform detection and classification simultaneously [26].…”

“…2 shows an example of such conventional DL framework, where two networks are used to detect and classify bone marrow hematopoietic stem cells, respectively. For instance, we can use the HDAE network [21] to detect cells and a CNN to classify different cell types (Fig. 2).…”

A New Generalized Deep Learning Framework Combining Sparse Autoencoder and Taguchi Method for Novel Data Classification and Processing

“…The cost function of training sparse autoencoder is also illustrated in (10). The training algorithm tries to reduce the cost function by finding the optimal parameters that essentially aims to reduce the value of .…”

“…Chaoqun Hong et al proposed a new pose retrieval technique which focuses on multimodal integration feature extraction and backpropagation deep neural network by using multilayered deep neural network with nonlinear mapping [9]. In [10] TzuHsi Song et al focused on bone marrow trepan biopsy images and proposed a hybrid deep autoencoder (HDA) network with Curvature Gaussian method for active and exact bone marrow hematopoietic stem cell detection via related highlevel feature correspondence. In [11] Yosuke Suzuki et al proposed a collaborative filtering based recommendation algorithm that employs the variation of similarities among users derived from different layers in stacked denoising autoencoders.…”