Classification of skin cancer from dermoscopic images using deep neural network architectures.

A powerful medical decision support system for classifying skin lesions from dermoscopic images is an important tool to prognosis of skin cancer. In the recent years, Deep Convolutional Neural Network (DCNN) have made a significant advancement in detecting skin cancer types from dermoscopic images, in-spite of its fine grained variability in its appearance. The main objective of this research work is to develop a DCNN based model to automatically classify skin cancer types into melanoma and non-melanoma with high accuracy. The datasets used in this work were obtained from the popular challenges ISIC-2019 and ISIC-2020, which have different image resolutions and class imbalance problems. To address these two problems and to achieve high performance in classification we have used EfficientNet architecture based on transfer learning techniques, which learns more complex and fine grained patterns from lesion images by automatically scaling depth, width and resolution of the network. We have augmented our dataset to overcome the class imbalance problem and also used metadata information to improve the classification results. Further to improve the efficiency of the EfficientNet we have used ranger optimizer which considerably reduces the hyper parameter tuning, which is required to achieve state-of-the-art results. We have conducted several experiments using different transferring models and our results proved that EfficientNet variants outperformed in the skin lesion classification tasks when compared with other architectures. The performance of the proposed system was evaluated using Area under the ROC curve (AUC - ROC) and obtained the score of 0.9681 by optimal fine tuning of EfficientNet-B6 with ranger optimizer.

Carbon Dots Fabrication: Ocular Imaging and Therapeutic Potential.

Vision loss is a major complication in common ocular infections and diseases such as bacterial keratitis, age-related macular degeneration (AMD) and diabetic retinopathy (DR). The prevalence of such ophthalmic diseases represents an urgent need to develop safe, effective, and long-term treatments. Current therapies are riddled with drawbacks and limitations which calls for the exploration of alternative drug delivery mechanisms. Toxicity of the inorganic metals and metal oxides used for drug delivery raise safety concerns that are alleviated with the alternate use of, a natural and organic polymer which is both biocompatible and environmentally friendly. Carbon dots (CDs) represent a great potential in novel biomedical applications due to their tunable fluorescence, biocompatibility, and ability to be conjugated with diverse therapeutic materials. There is a growing interest on the exploitation of these properties for drug delivery with enhanced bio-imaging. However, there are limited reports of CD applications for ophthalmic indications. In this review, we focus on the CD potential and the development of translational therapies for ophthalmic diseases. The current review presents better understanding of fabrication of CDs and how it may be useful in delivering anti-bacterial agents, anti-VEGF molecules as well as imaging for ophthalmic applications.

Self-Assembly of Block Heterochiral Peptides into Helical Tapes.

Patterned substitution of d-amino acids into the primary sequences of self-assembling peptides influences molecular-level packing and supramolecular morphology. We report that block heterochiral analogs of the model amphipathic peptide KFE8 (Ac-FKFEFKFE-NH2), composed of two FKFE repeat motifs with opposite chirality, assemble into helical tapes with dimensions greatly exceeding those of their fibrillar homochiral counterparts. At sufficient concentrations, these tapes form hydrogels with reduced storage moduli but retain the shear-thinning behavior and consistent mechanical recovery of the homochiral analogs. Varying the identity of charged residues (FRFEFRFE and FRFDFRFD) produced similarly sized nonhelical tapes, while a peptide with nonenantiomeric l- and d-blocks (FKFEFRFD) formed helical tapes closely resembling those of the heterochiral KFE8 analogs. A proposed energy-minimized model suggests that a kink at the interface between l- and d-blocks leads to the assembly of flat monolayers with nonidentical surfaces that display alternating stacks of hydrophobic and charged groups.

Suppression of Cocaine-Evoked Hyperactivity by Self-Adjuvanting and Multivalent Peptide Nanofiber Vaccines.

The development of anti-cocaine vaccines that counteract the rewarding effects of the drug are currently being investigated as adjunct therapies for prevention of relapse in abstinent users. However, cocaine is weakly immunogenic and requires conjugation to carrier proteins and coadministration with strong adjuvants, which carry the risk of local reactogenicity and systemic toxicity. Here we report synthetic and multivalent self-assembling peptide nanofibers as adjuvant-free carriers for cocaine vaccines. A novel cocaine hapten modified at the P3 site was conjugated to the N-terminus of an amphipathic self-assembling domain KFE8. In aqueous buffers the cocaine-KFE8 conjugate assembled into ß-sheet rich nanofibers, which raised anti-cocaine antibodies without the need for added adjuvants in mice. Vaccinated mice were treated with cocaine and a significant negative correlation was observed between antibody levels and cocaine-evoked hyperactivity. These totally synthetic and multivalent nanofibers with well-defined chemical composition represent the first generation of adjuvant-free cocaine vaccines.

Enhancing the Magnitude of Antibody Responses through Biomaterial Stereochemistry.

d-Amino acid analogs of peptides and proteins are attractive for applications in biotechnology and medicine due to their reduced proteolytic sensitivity. Here, we report that self-assembling peptide nanofibers composed of d-amino acids act as immune adjuvants, and investigate their ability to induce antibody responses in comparison to their l-amino acid counterparts. The model antigenic peptide OVA (chicken egg ovalbumin aa 323-339) from chicken egg ovalbumin, known to elicit antibody responses in mice, was linked to an l- or d-amino acid self-assembling peptide domain to generate enantiomeric nanofibers displaying the same epitope. The chiral nature of the fusion peptides was confirmed by circular dichrosim spectroscopy and transmission electron microscopy studies indicated that OVA-bearing enantiomers self-assembled into nanofibers with similar morphologies. In mice, d-amino acid peptide nanofibers displaying OVA elicited stronger antibody responses, equivalent levels of CD4+ T cell responses, and long-term antigen-presentation in vivo compared to l-amino acid nanofibers. Our findings indicate that self-assembling peptides composed of d-amino acids are strong immune adjuvants and that biomaterial stereochemistry can be used as a design tool to program adaptive immune responses for vaccine development.