Hybrid Deep Learning Framework for Melanoma Diagnosis Using Dermoscopic Medical Images.

Background: Melanoma, or skin cancer, is a dangerous form of cancer that is the major cause of the demise of thousands of people around the world. Methods: In recent years, deep learning has become more popular for analyzing and detecting these medical issues. In this paper, a hybrid deep learning approach has been proposed based on U-Net for image segmentation, Inception-ResNet-v2 for feature extraction, and the Vision Transformer model with a self-attention mechanism for refining the features for early and accurate diagnosis and classification of skin cancer. Furthermore, in the proposed approach, hyperparameter tuning helps to obtain more accurate and optimized results for image classification. Results: Dermoscopic shots gathered by the worldwide skin imaging collaboration (ISIC2020) challenge dataset are used in the proposed research work and achieved 98.65% accuracy, 99.20% sensitivity, and 98.03% specificity, which outperforms the other existing approaches for skin cancer classification. Furthermore, the HAM10000 dataset is used for ablation studies to compare and validate the performance of the proposed approach. Conclusions: The achieved outcome suggests that the proposed approach would be able to serve as a valuable tool for assisting dermatologists in the early detection of melanoma.

Toward Efficient Bactericidal and Dye Degradation Performance of Strontium- and Starch-Doped Fe2O3 Nanostructures: In Silico Molecular Docking Studies.

In this study, various concentrations of strontium (Sr) into a fixed amount of starch (St) and Fe2O3 nanostructures (NSs) were synthesized with the co-precipitation approach to evaluate the antibacterial and photocatalytic properties of the concerned NSs. The study aimed to synthesize nanorods of Fe2O3 with co-precipitation to enhance the bactericidal behavior with dopant-dependent Fe2O3. Advanced techniques were utilized to investigate the structural characteristics, morphological properties, optical absorption and emission, and elemental composition properties of synthesized samples. Measurements via X-ray diffraction confirmed the rhombohedral structure for Fe2O3. Fourier-transform infrared analysis explored the vibrational and rotational modes of the O-H functional group and the C=C and Fe-O functional groups. The energy band gap of the synthesized samples was observed in the range of 2.78-3.15 eV, which indicates that the blue shift in the absorption spectra of Fe2O3 and Sr/St-Fe2O3 was identified with UV-vis spectroscopy. The emission spectra were obtained through photoluminescence spectroscopy, and the elements in the materials were determined using energy-dispersive X-ray spectroscopy analysis. High-resolution transmission electron microscopy micrographs showed NSs that exhibit nanorods (NRs), and upon doping, agglomeration of NRs and nanoparticles was observed. Efficient degradations of methylene blue increased the photocatalytic activity in the implantation of Sr/St on Fe2O3 NRs. The antibacterial potential for Escherichia coli and Staphylococcus aureus was measured against ciprofloxacin. E. coli bacteria exhibit inhibition zones of 3.55 and 4.60 mm at low and high doses, respectively. S. aureus shows the measurement of inhibition zones for low and high doses of prepared samples at 0.47 and 2.40 mm, respectively. The prepared nanocatalyst showed remarkable antibacterial action against E. coli bacteria rather than S. aureus at high and low doses compared to ciprofloxacin. The best-docked conformation of the dihydrofolate reductase enzyme against E. coli for Sr/St-Fe2O3 showed H-bonding interactions with Ile-94, Tyr-100, Tyr-111, Trp-30, ASP-27, Thr-113, and Ala-6.

Photocatalytic and antibacterial activity of graphene oxide/cellulose-doped TiO2 quantum dots: in silico molecular docking studies.

Graphene oxide (GO) and cellulose nanocrystal (CNC)-doped TiO2 quantum dots (QDs) were effectively synthesized by employing the co-precipitation method for the degradation of dyes and antimicrobial applications. A series of characterizations, i.e., XRD, FTIR, UV-visible spectroscopy, EDS, FE-SEM, and HR-TEM, was used to characterize the prepared samples. A reduction in PL intensity was observed, while the band gap energy (E g) decreased from 3.22 to 2.96 eV upon the incorporation of GO/CNC in TiO2. In the Raman spectra, the D and G bands were detected, indicating the presence of graphene oxide in the composites. Upon doping, the crystallinity of TiO2 increased. HR-TEM was employed to estimate the interlayer d-spacing of the nanocomposites, which matched well with the XRD data. The photocatalytic potential of the prepared samples was tested against methylene blue, methylene violet, and ciprofloxacin (MB:MV:CF) when exposed to visible light for a certain period. The antibacterial activity of GO/CNC/TiO2 QDs against Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) bacteria in vitro was tested to determine their potential for medicinal applications. The molecular docking investigations of CNC-TiO2 and GO/CNC-doped TiO2 against DNA gyrase and FabI from E. coli and S. aureus were found to be consistent with the results of the in vitro bactericidal activity test. We believe that the prepared nanocomposites will be highly efficient for wastewater treatment and antimicrobial activities.

Improved catalytic activity and bactericidal behavior of novel chitosan/V2O5 co-doped in tin-oxide quantum dots.

The novel V2O5/chitosan (CS) co-doped tin oxide (SnO2) quantum dots (QDs) were synthesized via co-precipitation technique. The optical, structural, morphological, and catalytic properties of the concerned specimens were examined by UV-Vis, PL, FTIR, X-ray diffraction, HR-TEM, and EDS. Structural analysis through XRD confirmed the tetragonal structure of SnO2; meanwhile, HR-TEM measurements unveiled quantum dot morphology. Rotational and vibrational modes related to functional groups of (O-H, C-H, Sn-O, and Sn-O-Sn) have been assessed with FTIR spectra. Through UV-Vis spectroscopy, a reduction in band-gap (4.39 eV to 3.98 eV) and redshift in co-doped spectra of SnO2 were identified. Both CS/SnO2 and V2O5-doped CS@SnO2 showed promising catalytic activity in all media. Meanwhile, CS/SnO2 showed higher activity for use in hospital and industrial dye degradation in comparison to dopant-free Ch/SnO2. For V2O5/CS@ SnO2 QDs, inhibition domains of G -ve were significantly confirmed as 1.40-4.15 mm and 1.85-5.45 mm; meanwhile, for G +ve were noticed as 2.05-4.15 mm and 2.40-5.35 mm at least and maximum concentrations, correspondingly. These findings demonstrate the efficient role of V2O5/CS@SnO2 QDs towards industrial dye degradation and antimicrobial activity.

Deep Learning Approach for Automatic Microaneurysms Detection.

In diabetic retinopathy (DR), the early signs that may lead the eyesight towards complete vision loss are considered as microaneurysms (MAs). The shape of these MAs is almost circular, and they have a darkish color and are tiny in size, which means they may be missed by manual analysis of ophthalmologists. In this case, accurate early detection of microaneurysms is helpful to cure DR before non-reversible blindness. In the proposed method, early detection of MAs is performed using a hybrid feature embedding approach of pre-trained CNN models, named as VGG-19 and Inception-v3. The performance of the proposed approach was evaluated using publicly available datasets, namely "E-Ophtha" and "DIARETDB1", and achieved 96% and 94% classification accuracy, respectively. Furthermore, the developed approach outperformed the state-of-the-art approaches in terms of sensitivity and specificity for microaneurysms detection.