Although automated Skin Lesion Classification (SLC) is a crucial integral step in computer-aided diagnosis, it remains a challenging task due to inconsistency in texture, color, indistinguishable boundaries, and shapes. In this article, we propose an automatic and robust framework for the dermoscopic SLC named Dermoscopic Expert (DermoExpert). The DermoExpert consists of a preprocessing, a hybrid Convolutional Neural Network (hybrid-CNN), and transfer learning. The proposed hybrid-CNN classifier consists of three distinct feature extractors, with the same input images, which are fused to achieve better-depth feature maps of the corresponding lesion. Those distinct and fused feature maps are classified using the different fully connected layers, which are then ensembled to get a final prediction probability. In the preprocessing, we use lesion segmentation, augmentation, and class rebalancing. For boosting the lesion recognition, we have also employed geometric and intensity-based augmentation as well as the class rebalancing by penalizing the loss of the majority class and adding extra images to the minority classes. Additionally, we leverage the knowledge from a pre-trained model, also known as transfer learning, to build a generic classifier, although small datasets are being used. In the end, we design and implement a web application by deploying the weights of our DermoExpert for automatic lesion recognition. We evaluate our DermoExpert on the ISIC-2016, ISIC-2017, and ISIC-2018 datasets, where our DermoExpert has achieved the area under the receiver operating characteristic curve (AUC) of 0.96, 0.95, and 0.97, respectively. The experimental results outperform the recent state-of-the-art by a margin of 10.0 % and 2.0 % respectively for ISIC-2016 and ISIC-2017 datasets in terms of AUC. The DermoExpert also outperforms, in concerning a balanced accuracy, by a margin of 3.0 % for ISIC-2018 dataset. Since our framework can provide better-classification on three different test datasets, even with limited training data, it can lead to better-recognition of melanoma to aid dermatologists. Our source code, and segmented masks, for ISIC-2018 dataset, will be made publicly available for the research community for further improvements.
Previous studies showed that prolonged exposure to fluoride (F-) and aluminum (Al3+) ions is associated with numerous diseases including neurological disorders. They don't have any known biological function. But they can bind with proteins that interact with ions similar to them. Such unwanted interactions affect the normal biological function of the target proteins, as well as their downstream protein-protein interactions. Several studies show the detrimental effects posed by them including Alzheimer's disease. However, their target proteins have never been reported. Here, we have screened for the human protein targets subjected to F- and Al3+ interactions by using data-driven prediction tools. We have identified 20 different proteins that directly bind with them (10 interact with fluoride and 10 with aluminum). In addition, protein-protein interaction has been explored to find the proteins that indirectly interact with F- and Al3+. We have found 86 indirect targets for F- and 90 for Al3+. Furthermore, 19 common protein targets have been identified, including proteins (9 out of 19) associated with neurodegenerative disorders. However, wet lab experiments are beyond our scopes to validate the binding networks. Additional studies must be warranted.
A large number of studies in the past months have proposed deep learning-based Artificial Intelligence (AI) tools for automated detection of COVID-19 using publicly available datasets of Chest X-rays (CXRs) or CT scans for training and evaluation. Most of these studies report high accuracy when classifying COVID-19 patients from normal or other commonly occurring pneumonia cases. However, these results are often obtained on cross-validation studies without an independent test set coming from a separate dataset and have biases such as the two classes to be predicted come from two completely different datasets. In this work, we investigate potential overfitting and biases in such studies by designing different experimental setups within the available public data constraints and highlight the challenges and limitations of developing deep learning models with such datasets. We propose a deep learning architecture for COVID-19 classification that combines two very popular classification networks, ResNet and Xception, and use it to carry out the experiments to investigate challenges and limitations. The results show that the deep learning models can overestimate their performance due to biases in the experimental design and overfitting to the training dataset. We compare the proposed architecture to state-of-the-art methods utilizing an independent test set for evaluation, where some of the identified bias and overfitting issues are reduced. Although our proposed deep learning architecture gives the best performance with our best possible setup, we highlight the challenges in comparing and interpreting various deep learning algorithms’ results. While the deep learning-based methods using chest imaging data show promise in being helpful for clinical management and triage of COVID-19 patients, our experiments suggest that a larger, more comprehensive database with less bias is necessary for developing tools applicable in real clinical settings.
With a variety of accessible Single Nucleotide Polymorphisms (SNPs) data on human p53 gene, this investigation is intended to deal with detrimental SNPs in p53 gene by executing diverse valid computational tools, including Filter, SIFT, PredictSNP, Fathmm, UTRScan, ConSurf, Phyre, Tm-Adjust, I-Mutant, Task Seek after practical and basic appraisal, dissolvable openness, atomic progression, and analysing the energy minimization. Of 581 p53 SNPs, 420 SNPs are found to be missense or non-synonymous and 435 SNPs are in the 3 prime UTR and 112 SNPs are of every 5 prime UTR from which 16 non synonymous SNPs (nsSNPs) as non-tolerable while PredictSNP package predicted 14 (taking consideration SNP colored green by two or more than 2 analyses is neutral). By concentrating on six bioinformatics tools of various dimensions a combined output is generated where 14 nsSNPs are prone to exert a deleterious effect. By using diverse SNP analysing tools we have found 5 missense SNPs in the 3 crucial amino acids position in the DNA binding domain. The underlying discoveries are fortified by I-Mutant and Project HOPE. The ExPASy-PROSITE tools characterized whether the mutations located in the functional part of the protein or not. This study provides a decisive outcome concluding the accessible SNPs information by recognizing the five harming nsSNPs: rs28934573 (S241F), rs11540652 (R248Q), rs121913342 (R248W), rs121913343 (R273C) and rs28934576 (R273H). The findings of this investigation recognize the detrimental nsSNPs which enhance the danger of various kinds of oncogenesis in patients of different populations’ in genome-wide studies (GWS).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.