Federated Learning for Healthcare Domain - Pipeline, Applications and Challenges

Federated learning is the process of developing machine learning models over datasets distributed across data centers such as hospitals, clinical research labs, and mobile devices while preventing data leakage. This survey examines previous research and studies on federated learning in the healthcare sector across a range of use cases and applications. Our survey shows what challenges, methods, and applications a practitioner should be aware of in the topic of federated learning. This paper aims to lay out existing research and list the possibilities of federated learning for healthcare industries.© 2022 Copyright held by the owner/author(s).

Virtual Screening of the Flavonoids Compounds with the SARS-CoV-2 3C-like Protease as the Lead Compounds for the COVID-19

As of 1st of September 2020, the COVID-19 pandemic has reached an unprecedented level of more than 25 million cases with more than 850,000 deaths. Moreover, all the drug candidates are still undergoing testing in clinical trials. In this regard, a breakthrough in drug design is neces-sary. One strategy to devise lead compounds is leveraging natural products as a lead source. Sever-al companies and research institutes are currently developing anti-SARS-CoV-2 lead from natural products. Flavonoids are well known as a class of antiviral compounds library. The objective of this research is to employ virtual screening methods for obtaining the best lead compounds from the library of flavonoid compounds. This research employed virtual screening methods that com-prised of downloading the protein and lead compound structures, QSAR analysis prediction, itera-tions of molecular docking simulation, and ADME-TOX simulation for toxicity prediction. The QSAR analysis found that the tested compounds have broad-spectrum antiviral activity, and some of them exhibit specific binding to the 3C-like Protease of the Coronavirus. Moreover, juglanin was found as the compound with the fittest binding with the Protease enzyme of SARS-CoV-2. Al-though most of the tested compounds are deemed toxic by the ADME-Tox test, further research should be conducted to comprehend the most feasible strategy to deliver the drug to the infected lung cells. The juglanin compound is selected as the fittest candidate as the SARS-CoV-2 lead compound in the tested flavonoid samples. However, further research should be conducted to observe the lead delivery method to the cell.Copyright © 2021 Bentham Science Publishers.

Computational tools for aptamer identification and optimization

Aptamers are single-stranded DNA or RNA oligonucleotides that can selectively bind to a specific target. They are generally obtained by SELEX, but the procedure is challenging and time-consuming. Moreover, the identified aptamers tend to be insufficient in stability, specificity, and affinity. Thus, only a handful of aptamers have entered the practical use stage. Recently, computational approaches have demonstrated a significant capacity to assist in the discovery of high-performance aptamers. This review discusses the advances achieved in several aspects of computational tools in this field, as well as the new progress in machine learning and deep learning, which are used in aptamer identification and optimization. To illustrate these computationally aided processes, aptamer selections against SARS-CoV-2 are discussed in detail as a case study. We hope that this review will aid and motivate researchers to develop and utilize more computational techniques to discover ideal aptamers effectively. Copyright © 2022 Elsevier B.V.

Content Based Medical Image Retrieval for Accurate Disease Diagnosis

Introduction: Content Based Image Retrieval (CBIR) system is an innovative technology to retrieve images from various media types. One of the CBIR applications is Content Based Medical Image Retrieval (CBMIR). The image retrieval system retrieves the most similar images from the historical cases, and such systems can only support the physician's decision to diagnose a disease. To extract the useful features from the query image for linking similar types of images is the major challenge in the CBIR domain. The Convolution Neural Network (CNN) can overcome the drawbacks of traditional algorithms, dependent on the low-level feature extraction technique. Objective: The objective of the study is to develop a CNN model with a minimum number of convolution layers and to get the maximum possible accuracy for the CBMIR system. The minimum number of convolution layers reduces the number of mathematical operations and the time for the model's training. It also reduces the number of training parameters, like weights and bias. Thus, it reduces the memory requirement for the model storage. This work mainly focused on developing an optimized CNN model for the CBMIR system. Such systems can only support the physicians' decision to diagnose a disease from the images and retrieve the relevant cases to help the doctor decide the precise treatment. Methods: The deep learning-based model is proposed in this paper. The experiment is done with several convolution layers and various optimizers to get the maximum accuracy with a minimum number of convolution layers. Thus, the ten-layer CNN model is developed from scratch and used to derive the training and testing images' features and classify the test image. Once the image class is identified, the most relevant images are determined based on the Euclidean distance between the query features and database features of the identified class. Based on this distance, the most relevant images are displayed from the respective class of images. The general dataset CIFAR10, which has 60,000 images of 10 different classes, and the medical dataset IRMA, which has 2508 images of 9 various classes, have been used to analyze the proposed method. The proposed model is also applied for the medical x-ray image dataset of chest disease and compared with the other pre-trained models. Results: The accuracy and the average precision rate are the measurement parameters utilized to compare the proposed model with different machine learning techniques. The accuracy of the proposed model for the CIFAR10 dataset is 93.9%, which is better than the state-of-the-art methods. After the success for the general dataset, the model is also tested for the medical dataset. For the x-ray images of the IRMA dataset, it is 86.53%, which is better than the different pre-trained model results. The model is also tested for the other x-ray dataset, which is utilized to identify chest-related disease. The average precision rate for such a dataset is 97.25%. Also, the proposed model fulfills the major challenge of the semantic gap. The semantic gap of the proposed model for the chest disease dataset is 2.75%, and for the IRMA dataset, it is 13.47%. Also, only ten convolution layers are utilized in the proposed model, which is very small in number compared to the other pre-trained models. Conclusion: The proposed technique shows remarkable improvement in performance metrics over CNN-based state-of-the-art methods. It also offers a significant improvement in performance metrics over different pre-trained models for the two different medical x-ray image datasets.

A computational approach for designing novel SARS-CoV-2 Mproinhibitors: combined QSAR, molecular docking, and molecular dynamics simulation techniques

The normal expression of the main protease (Mpro) plays a vital role in the life cycle of coronavirus. Highly active inhibitors could inhibit the normal circulation of the main protease to achieve therapeutic effects as anti-coronavirus agents. In the present research, 48 peptide compounds with SARS-CoV Mproinhibition selected from the literature were used to establish robust Topomer CoMFA (q2= 0.743,r2= 0.938, andrpred2= 0.700) and HQSAR (q2= 0.774,r2= 0.955, andrpred2= 0.723) models. Structural modification information was used for designing new Mproinhibitors. The high contribution-value descriptor generated by Topomer CoMFA was used to screen for the fragments that possess significant inhibitory activities from the ZINC drug database, and 24 new compounds with predicted high inhibitory activity at nanomolar concentration were designed by combining the high contribution value fragments. The molecular docking results further justified that these potential inhibitors could form hydrogen bonds with the residues of CYS145, GLN189, GLU166, HIS163, and GLY143 of target Mpro, which well explains their strong inhibitory effects. The molecular dynamics simulation results indicated that four highly active compounds could stably bond with SARS-CoV-2 Mproand might be promising anti-SARS-CoV-2 Mprocandidates. Finally, all the newly designed compounds showed premium ADMET properties as per the predictions by the server in the public domain. This research work not only provides robust QSAR models as valuable screening tools for future anti-coronavirus drug development but also renders the newly designed SARS-CoV-2 Mproinhibitors with activity at nanomolar concentration, which can be used for further characterization to obtain novel anti-coronavirus drugs for both SARS-CoV and SARS-CoV-2.