A5088473143- Radiomics and Machine Learning in Medical Imaging
- COVID-19 Clinical Research Studies
- RNA modifications and cancer
- RNA Research and Splicing
- Cutaneous Melanoma Detection and Management
- Lung Cancer Diagnosis and Treatment
- RNA and protein synthesis mechanisms
- AI in cancer detection
- COVID-19 diagnosis using AI
- Genomics and Phylogenetic Studies
- Chromosomal and Genetic Variations
- SARS-CoV-2 and COVID-19 Research
- COVID-19 epidemiological studies
- Advanced Glycation End Products research
- Plant Disease Resistance and Genetics
- Urinary Bladder and Prostate Research
- Poxvirus research and outbreaks
- Long-Term Effects of COVID-19
- Cancer Genomics and Diagnostics
- Bacillus and Francisella bacterial research
- Diet and metabolism studies
- Advanced Radiotherapy Techniques
- Cancer-related cognitive impairment studies
- Urinary and Genital Oncology Studies
- Cancer-related gene regulation
Stanford University
2021-2024
Middle East Technical University
2023
Hacettepe University
2023
National Hospital for Neurology and Neurosurgery
2021-2022
University College London
2018-2022
The Francis Crick Institute
2018-2022
University College Dublin
2016
Abstract Undetected infection and delayed isolation of infected individuals are key factors driving the monkeypox virus (now termed mpox or MPXV) outbreak. To enable earlier detection MPXV infection, we developed an image-based deep convolutional neural network (named MPXV-CNN) for identification characteristic skin lesions caused by MPXV. We assembled a dataset 139,198 lesion images, split into training/validation testing cohorts, comprising non-MPXV images ( n = 138,522) from eight...
The immune response to SARS-CoV-2 infection requires antibody recognition of the spike protein. In a study designed examine molecular features anti-spike and anti-nucleocapsid antibodies, patient plasma proteins binding pre-fusion stabilised complete nucleocapsid were isolated analysed by matrix-assisted laser desorption ionisation-time flight (MALDI-ToF) mass spectrometry. Amongst immunoglobulins, high affinity for human serum albumin was evident in preparations. Careful comparison revealed...
Magnetic resonance images (MRI) of the brain exhibit high dimensionality that pose significant challenges for computational analysis. While models proposed MRIs analyses yield encouraging results, complexity neuroimaging data hinders generalizability and clinical application. We introduce DUNE, a neuroimaging-oriented encoder designed to extract deep-features from multisequence MRIs, thereby enabling their processing by basic machine learning algorithms. A UNet-based autoencoder was trained...
Abstract Crosslinking and Immunoprecipitation (CLIP) is a powerful technique to obtain transcriptome-wide maps of in vivo protein-RNA interactions, which are important understand the post-transcriptional mechanisms mediated by RNA binding proteins (RBPs). Many variant CLIP protocols have been developed improve efficiency convenience cDNA library preparation. Here we describe an improved individual nucleotide resolution protocol (iiCLIP), can be completed within 4 days from UV crosslinking...
The involvement of immunoglobulin (Ig) G3 in the humoral immune response to SARS-CoV-2 infection has been implicated pathogenesis acute respiratory distress syndrome (ARDS) COVID-19. exact molecular mechanism is unknown, but it thought involve this IgG subtype's differential ability fix, complement and stimulate cytokine release. We examined binding convalescent patient antibodies immobilized nucleocapsids spike proteins by matrix-assisted laser desorption/ionization-time flight (MALDI-ToF)...
Abstract Genomic methods have been valuable for identifying RNA-binding proteins (RBPs) and the genes, pathways, processes they regulate. Nevertheless, standard motif descriptions cannot be used to predict all RNA targets or test quantitative models cellular interactions regulation. We present a complete thermodynamic model binding S. cerevisiae Pumilio protein PUF4 derived from direct data 6180 RNAs measured using on massively parallel array (RNA-MaP) platform. The is highly similar that of...
The relationship between DNA sequence, biochemical function, and molecular evolution is relatively well-described for protein-coding regions of genomes, but far less clear in noncoding regions, particularly, eukaryote genomes. In part, this because we lack a complete description the essential elements genome. To contribute to challenge, used saturating transposon mutagenesis interrogate Schizosaccharomyces pombe We generated 31 million insertions, theoretical coverage 2.4 insertions per...
Through technological innovations, patient cohorts can be examined from multiple views with high-dimensional, multiscale biomedical data to classify clinical phenotypes and predict outcomes. Here, we aim present our approach for analyzing multimodal using unsupervised supervised sparse linear methods in a COVID-19 cohort. This prospective cohort study of 149 adult patients was conducted tertiary care academic center. First, used canonical correlation analysis (CCA) identify quantify...
Abstract With an estimated 3 billion people globally lacking access to dermatological care, technological solutions leveraging artificial intelligence (AI) have been proposed improve 1 . Diagnostic AI algorithms, however, require high-quality datasets allow development and testing, particularly those that enable evaluation of both unimodal multimodal approaches. Currently, the majority dermatology algorithms are built tested on proprietary, siloed data, often from a single site with only...
ABSTRACT Background Non‐small‐cell lung cancer (NSCLC) remains a global health challenge, driving morbidity and mortality. The emerging field of radiogenomics utilizes statistical methods to correlate radiographic tumor features with genomic characteristics from biopsy samples. Radiomic techniques automate the precise extraction imaging regions in scans, which are subjected machine learning (ML) predict attributes. Methods In retrospective study two NSCLC patient cohorts separated by 5...
In this work, we investigate chemo- thermotherapy, a recently clinically-approved post-surgery treatment of non muscle invasive urothelial bladder carcinoma. We developed mathematical model and numerically simulated the physical processes related to treatment. The is based on conductive Maxwell's equations used simulate therapy administration Convection-Diffusion equation for incompressible fluid study heat propagation through tissue. parameters correspond data provided by thermotherapy...
In this study, we develop a 3D beta variational autoencoder (beta-VAE) to advance lung cancer imaging analysis, countering the constraints of conventional radiomics methods. The extracts information from public computed tomography (CT) datasets without additional labels. It reconstructs nodule images with high quality (structural similarity: 0.774, peak signal-to-noise ratio: 26.1, and mean-squared error: 0.0008). model effectively encodes lesion sizes in its latent embeddings, significant...
Current methods for microplastic identification in water samples are costly and require expert analysis. Here, we propose a deep learning segmentation model to automatically identify microplastics microscopic images. We labeled images of from the Moore Institute Plastic Pollution Research employ Generative Adversarial Network (GAN) supplement generate diverse training data. To verify validity generated data, conducted reader study where an was able discern real at rate 68 percent. Our...
Beta variational auto-encoder (beta-VAE) is a deep learning technique that can learn features from medical imaging without any additional labels. In this study, we propose 3D as an information extractor be trained on public lung CT datasets. We showed the beta-VAE reconstruct images. The reconstructed images achieved high similarity with original (structural similarity: 0.774, peak signal-to-noise ratio: 26.1, and mean squared error: 0.0008). size of lesions encoded in latent embeddings....
Abstract Since the immune response to SARS-CoV2 infection requires antibody recognition of Spike protein, we used MagMix, a semi-automated magnetic rack reproducibly isolate patient plasma proteins bound pre-fusion stabilised and nucleocapsid conjugated beads. Once eluted, MALDI-ToF mass spectrometry identified range immunoglobulins, but also in protein beads found high affinity for human serum albumin. Careful comparison revealed preferential capture advanced glycation end product (AGE)...
Abstract The involvement of IgG3 in the humoral immune response to SARS-CoV2 infection has been implicated pathogenesis ARDS COVID-19. exact molecular mechanism is unknown but may be due differential ability Fc region fix complement and stimulate cytokine release. We examined convalescent patients’ antibodies binding immobilised nucleocapsid spike protein by MALDI-ToF mass spectrometry. was a major immunoglobulin found all samples. Differential analysis spectral signatures for versus...
Abstract Genomic methods have been valuable for identifying RNA-binding proteins (RBPs) and the genes, pathways, processes they regulate. Nevertheless, standard motif descriptions cannot be used to predict all RNA targets or test quantitative models cellular interactions regulation. We present a complete thermodynamic model binding S. cerevisiae Pumilio protein PUF4 derived from direct data 6180 RNAs measured using on massively parallel array (RNA-MaP) platform. The is highly similar that of...
<title>Abstract</title> Through technological innovations, patient cohorts can be examined from multiple views with high-dimensional, multiscale biomedical data to classify clinical phenotypes and predict outcomes. Here, we aim present our approach for analyzing multimodal using unsupervised supervised sparse linear methods in a COVID-19 cohort. This prospective cohort study of 149 adult patients was conducted tertiary care academic center. First, used canonical correlation analysis (CCA)...
Abstract Background Non-protein-coding regions of eukaryotic genomes remain poorly understood. Diversity studies, comparative genomics and biochemical outputs genomic sites can be indicators functional elements, but none produce fine-scale genome-wide descriptions all elements. Results Towards the generation a comprehensive description elements in haploid Schizosaccharomyces pombe genome, we generated transposon mutagenesis libraries to density one insertion per 13 nucleotides genome. We...