In this study, we proposed a novel convolutional neural network (CNN) architecture for classification of benign and malignant breast cancer (BC) in histological images. To improve the delivery and use of feature information, we chose the DenseNet as the basic building block and interleaved it with the squeeze-and-excitation (SENet) module. We conducted extensive experiments with the proposed framework by using the public domain BreakHis dataset and demonstrated that the proposed framework can produce significantly improved accuracy in BC classification, compared with the state-of-the-art CNN methods reported in the literature.

Intervertebral disc degeneration (IDD) has been considered as the primary pathological mechanism that underlies low back pain. Understanding the molecular mechanisms underlying human IDD is imperative for making strategies to treat IDD-related diseases. Herein, we report the molecular programs, lineage progression patterns, and paths of cellular communications during the progression of IDD using single-cell RNA sequencing (scRNA-seq) on nucleus pulposus (NP) cells from patients with different grades of IDD undergoing discectomy. New subtypes of cells and cell-type-specific gene signatures of the metabolic homeostatic NP cells (Met NPC), adhesive NP cells (Adh NPC), inflammatory response NP cells (IR NPC), endoplasmic reticulum stress NP cells (ERS NPC), fibrocartilaginous NP cells (Fc NPC), and CD70 and CD82⁺ progenitor NP cells (Pro NPC) were identified. In the late stage of IDD, the IR NPC and Fc NPC account for a large proportion of NPC. Importantly, immune cells including macrophages, T cells, myeloid progenitors, and neutrophils were also identified, and further analysis showed that significant intercellular interaction between macrophages and Pro NPC occurred via MIF (macrophage migration inhibitory factor) and NF-kB signaling pathways during the progression of IDD. In addition, dynamic polarization of macrophage M1 and M2 cell subtypes was found in the progression of IDD, and gene set functional enrichment analysis suggested a significant role of the macrophage polarization in regulating cell metabolism, especially the Pro NPC. Finally, we found that the NP cells in the late degenerative stage were mainly composed of the cell types related to inflammatory and endoplasmic reticulum (ER) response, and fibrocartilaginous activity. Our results provided new insights into the identification of NP cell populations at single-cell resolution and at the relatively whole-transcriptome scale, accompanied by cellular communications between immune cells and NP cells, and discriminative markers in relation to specific cell subsets. These new findings present clues for effective and functional manipulation of human IDD-related bioremediation and healthcare.

The diversity of base classifiers and integration of multiple classifiers are two key issues in the field of ensemble learning. This paper puts forward a hybrid ensemble algorithm combining AdaBoost and genetic algorithm(GA) for cancer classification with gene expression data. The decision group is designed to increase the diversity of base classifier pool, and the GA is used to assign weight to each base classifier, thus to improve the classification performance by avoiding local extrema. The decision groups composed by using base classifiers, including K-nearest neighbor (KNN), Naïve Bayes (NB) and Decision Tree (C4.5). Experimental results show that the proposed algorithm is superior to those existing ensemble learning methods, such as Bagging, Random Forest (RF), Rotation Forest (RoF), AdaBoost, AdaBoost-BPNN, AdaBoost-SVM and AdaBoost-RF, especially it has better performance on small samples and unbalanced gene expression data processing.