Basal Cell Carcinoma Diagnosis with Fusion of Deep Learning and Telangiectasia Features.

In recent years, deep learning (DL) has been used extensively and successfully to diagnose different cancers in dermoscopic images. However, most approaches lack clinical inputs supported by dermatologists that could aid in higher accuracy and explainability. To dermatologists, the presence of telangiectasia, or narrow blood vessels that typically appear serpiginous or arborizing, is a critical indicator of basal cell carcinoma (BCC). Exploiting the feature information present in telangiectasia through a combination of DL-based techniques could create a pathway for both, improving DL results as well as aiding dermatologists in BCC diagnosis. This study demonstrates a novel "fusion" technique for BCC vs non-BCC classification using ensemble learning on a combination of (a) handcrafted features from semantically segmented telangiectasia (U-Net-based) and (b) deep learning features generated from whole lesion images (EfficientNet-B5-based). This fusion method achieves a binary classification accuracy of 97.2%, with a 1.3% improvement over the corresponding DL-only model, on a holdout test set of 395 images. An increase of 3.7% in sensitivity, 1.5% in specificity, and 1.5% in precision along with an AUC of 0.99 was also achieved. Metric improvements were demonstrated in three stages: (1) the addition of handcrafted telangiectasia features to deep learning features, (2) including areas near telangiectasia (surround areas), (3) discarding the noisy lower-importance features through feature importance. Another novel approach to feature finding with weak annotations through the examination of the surrounding areas of telangiectasia is offered in this study. The experimental results show state-of-the-art accuracy and precision in the diagnosis of BCC, compared to three benchmark techniques. Further exploration of deep learning techniques for individual dermoscopy feature detection is warranted.

Hybrid Topological Data Analysis and Deep Learning for Basal Cell Carcinoma Diagnosis.

A critical clinical indicator for basal cell carcinoma (BCC) is the presence of telangiectasia (narrow, arborizing blood vessels) within the skin lesions. Many skin cancer imaging processes today exploit deep learning (DL) models for diagnosis, segmentation of features, and feature analysis. To extend automated diagnosis, recent computational intelligence research has also explored the field of Topological Data Analysis (TDA), a branch of mathematics that uses topology to extract meaningful information from highly complex data. This study combines TDA and DL with ensemble learning to create a hybrid TDA-DL BCC diagnostic model. Persistence homology (a TDA technique) is implemented to extract topological features from automatically segmented telangiectasia as well as skin lesions, and DL features are generated by fine-tuning a pre-trained EfficientNet-B5 model. The final hybrid TDA-DL model achieves state-of-the-art accuracy of 97.4% and an AUC of 0.995 on a holdout test of 395 skin lesions for BCC diagnosis. This study demonstrates that telangiectasia features improve BCC diagnosis, and TDA techniques hold the potential to improve DL performance.

Mast Cell Proteases Activate Astrocytes and Glia-Neurons and Release Interleukin-33 by Activating p38 and ERK1/2 MAPKs and NF-κB.

Inflammatory mediators released from activated microglia, astrocytes, neurons, and mast cells mediate neuroinflammation. Parkinson's disease (PD) is characterized by inflammation-dependent dopaminergic neurodegeneration in substantia nigra. 1-Methyl-4-phenylpyridinium (MPP+), a metabolite of parkinsonian neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), induces inflammatory mediators' release from brain cells and mast cells. Brain cells' interaction with mast cells is implicated in neuroinflammation. However, the exact mechanisms involved are not yet clearly understood. Mouse fetal brain-derived cultured primary astrocytes and glia-neurons were incubated with mouse mast cell protease-6 (MMCP-6) and MMCP-7, and mouse bone marrow-derived mast cells (BMMCs) were incubated with MPP+ and brain protein glia maturation factor (GMF). Interleukin-33 (IL-33) released from these cells was quantitated by enzyme-linked immunosorbent assay. Both MMCP-6 and MMCP-7 induced IL-33 release from astrocytes and glia-neurons. MPP+ and GMF were used as a positive control-induced IL-33 and reactive oxygen species expression in mast cells. Mast cell proteases and MPP+ activate p38 and extracellular signal-regulated kinases 1/2 (ERK1/2), mitogen-activated protein kinases (MAPKs), and transcription factor nuclear factor-kappa B (NF-κB) in astrocytes, glia-neurons, or mast cells. Addition of BMMCs from wt mice and transduction with adeno-GMF show higher chemokine (C-C motif) ligand 2 (CCL2) release. MPP+ activated glial cells and reduced microtubule-associated protein 2 (MAP-2) expression indicating neurodegeneration. IL-33 expression increased in the midbrain and striatum of PD brains as compared with age- and sex-matched control subjects. Glial cells and neurons interact with mast cells and accelerate neuroinflammation and these interactions can be explored as a new therapeutic target to treat PD.

Glia Maturation Factor Dependent Inhibition of Mitochondrial PGC-1α Triggers Oxidative Stress-Mediated Apoptosis in N27 Rat Dopaminergic Neuronal Cells.

Parkinson's disease (PD) is a progressive neurodegenerative disease affecting over five million individuals worldwide. The exact molecular events underlying PD pathogenesis are still not clearly known. Glia maturation factor (GMF), a neuroinflammatory protein in the brain plays an important role in the pathogenesis of PD. Mitochondrial dysfunctions and oxidative stress trigger apoptosis leading to dopaminergic neuronal degeneration in PD. Peroxisome proliferator-activated receptor-gamma coactivator-1 alpha (PGC-1α or PPARGC-α) acts as a transcriptional co-regulator of mitochondrial biogenesis and energy metabolism by controlling oxidative phosphorylation, antioxidant activity, and autophagy. In this study, we found that incubation of immortalized rat dopaminergic (N27) neurons with GMF influences the expression of peroxisome PGC-1α and increases oxidative stress, mitochondrial dysfunction, and apoptotic cell death. We show that incubation with GMF reduces the expression of PGC-1α with concomitant decreases in the mitochondrial complexes. Besides, there is increased oxidative stress and depolarization of mitochondrial membrane potential (MMP) in these cells. Further, GMF reduces tyrosine hydroxylase (TH) expression and shifts Bax/Bcl-2 expression resulting in release of cytochrome-c and increased activations of effector caspase expressions. Transmission electron microscopy analyses revealed alteration in the mitochondrial architecture. Our results show that GMF acts as an important upstream regulator of PGC-1α in promoting dopaminergic neuronal death through its effect on oxidative stress-mediated apoptosis. Our current data suggest that GMF is a critical risk factor for PD and suggest that it could be explored as a potential therapeutic target to inhibit PD progression.

Brain and Peripheral Atypical Inflammatory Mediators Potentiate Neuroinflammation and Neurodegeneration.

Neuroinflammatory response is primarily a protective mechanism in the brain. However, excessive and chronic inflammatory responses can lead to deleterious effects involving immune cells, brain cells and signaling molecules. Neuroinflammation induces and accelerates pathogenesis of Parkinson's disease (PD), Alzheimer's disease (AD) and Multiple sclerosis (MS). Neuroinflammatory pathways are indicated as novel therapeutic targets for these diseases. Mast cells are immune cells of hematopoietic origin that regulate inflammation and upon activation release many proinflammatory mediators in systemic and central nervous system (CNS) inflammatory conditions. In addition, inflammatory mediators released from activated glial cells induce neurodegeneration in the brain. Systemic inflammation-derived proinflammatory cytokines/chemokines and other factors cause a breach in the blood brain-barrier (BBB) thereby allowing for the entry of immune/inflammatory cells including mast cell progenitors, mast cells and proinflammatory cytokines and chemokines into the brain. These peripheral-derived factors and intrinsically generated cytokines/chemokines, α-synuclein, corticotropin-releasing hormone (CRH), substance P (SP), beta amyloid 1-42 (Aß1-42) peptide and amyloid precursor proteins can activate glial cells, T-cells and mast cells in the brain can induce additional release of inflammatory and neurotoxic molecules contributing to chronic neuroinflammation and neuronal death. The glia maturation factor (GMF), a proinflammatory protein discovered in our laboratory released from glia, activates mast cells to release inflammatory cytokines and chemokines. Chronic increase in the proinflammatory mediators induces neurotoxic Aß and plaque formation in AD brains and neurodegeneration in PD brains. Glial cells, mast cells and T-cells can reactivate each other in neuroinflammatory conditions in the brain and augment neuroinflammation. Further, inflammatory mediators from the brain can also enter into the peripheral system through defective BBB, recruit immune cells into the brain, and exacerbate neuroinflammation. We suggest that mast cell-associated inflammatory mediators from systemic inflammation and brain could augment neuroinflammation and neurodegeneration in the brain. This review article addresses the role of some atypical inflammatory mediators that are associated with mast cell inflammation and their activation of glial cells to induce neurodegeneration.