A5056635080- AI in cancer detection
- Molecular Biology Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Neurological diseases and metabolism
- Retinal Imaging and Analysis
- Prenatal Screening and Diagnostics
- Reproductive Biology and Fertility
- MicroRNA in disease regulation
- Bladder and Urothelial Cancer Treatments
- Parkinson's Disease Mechanisms and Treatments
- Prostate Cancer Diagnosis and Treatment
- Extracellular vesicles in disease
- Neuroinflammation and Neurodegeneration Mechanisms
- Digital Imaging for Blood Diseases
- Cancer Immunotherapy and Biomarkers
- Colorectal Cancer Screening and Detection
- Inflammation biomarkers and pathways
- Single-cell and spatial transcriptomics
- Retinal and Optic Conditions
- Retinal Diseases and Treatments
- Assisted Reproductive Technology and Twin Pregnancy
- Cancer-related molecular mechanisms research
- Cancer Genomics and Diagnostics
- Nuclear Receptors and Signaling
- Neurological Disease Mechanisms and Treatments
Cornell University
2019-2025
Weill Cornell Medicine
2020-2025
Lander Institute
2022-2024
Memorial Sloan Kettering Cancer Center
2017-2021
The hemizygous R47H variant of triggering receptor expressed on myeloid cells 2 (TREM2), a microglia-specific gene in the brain, increases risk for late-onset Alzheimer’s disease (AD). Using transcriptomic analysis single nuclei from brain tissues patients with AD carrying mutation or common (CV)–TREM2, we found that R47H-associated microglial subpopulations had enhanced inflammatory signatures reminiscent previously identified disease-associated microglia (DAM) and hyperactivation AKT, one...
Background A definitive diagnosis of prostate cancer requires a biopsy to obtain tissue for pathologic analysis, but this is an invasive procedure and associated with complications. Purpose To develop artificial intelligence (AI)‐based model (named AI‐biopsy) the early using magnetic resonance (MR) images labeled histopathology information. Study Type Retrospective. Population Magnetic imaging (MRI) data sets from 400 patients suspected histological (228 acquired in‐house 172 external...
Abstract Parkinson’s disease (PD) is a serious neurodegenerative disorder marked by significant clinical and progression heterogeneity. This study aimed at addressing heterogeneity of PD through integrative analysis various data modalities. We analyzed (≥5 years) individuals with de novo using machine learning deep learning, to characterize individuals’ phenotypic trajectories for subtyping. discovered three pace subtypes exhibiting distinct patterns: the Inching Pace subtype (PD-I) mild...
Abstract Increased deposition of extracellular matrix (ECM) is a known inhibitor axonal regrowth and remyelination. Recent in vitr o studies have demonstrated that oligodendrocyte differentiation impacted by the physical properties ECM. However, characterization mechanical healthy injured CNS myelin challenging, has largely relied on non-invasive, low-resolution methods. To address this, we employed atomic force microscopy to perform micro-indentation measurements demyelinated tissue at...
Assessing fertilized human embryos is crucial for in vitro fertilization, a task being revolutionized by artificial intelligence. Existing models used embryo quality assessment and ploidy detection could be significantly improved effectively utilizing time-lapse imaging to identify critical developmental time points maximizing prediction accuracy. Addressing this, we develop compare various status across distinct development stages. We present BELA, state-of-the-art model that surpasses...
Rheumatoid arthritis (RA) is a complex immune-mediated inflammatory disorder in which patients suffer from inflammatory-erosive arthritis. Recent advances on histopathology heterogeneity of RA synovial tissue revealed three distinct phenotypes based cellular composition (pauci-immune, diffuse and lymphoid), suggesting that etiologies warrant specific targeted therapy motivates need for cost effective phenotyping tools preclinical clinical settings. To this end, we developed an automated...
Building accurate prediction models and identifying predictive biomarkers for treatment response in Muscle-Invasive Bladder Cancer (MIBC) are essential improving patient survival but remain challenging due to tumor heterogeneity, despite numerous related studies. To address this unmet need, we developed an interpretable Graph-based Multimodal Late Fusion (GMLF) deep learning framework. Integrating histopathology cell type data from standard H&E images with gene expression profiles derived...
Abstract Parkinson’s disease (PD) is a complex neurodegenerative disorder with diverse clinical manifestations. To better understand this disease, research has been done to categorize, or subtype, patients, using an array of criteria derived from assessments and biospecimen analyses. In study, data the BioFIND cohort, we aimed at identifying subtypes moderate-to-advanced PD via comprehensively considering motor non-motor A total 103 patients were included for analysis. Through use...
533 Background: Accurate prediction of response to neoadjuvant chemotherapy (NAC) is essential for optimizing treatment outcomes in muscle-invasive bladder cancer (MIBC). We present a cutting-edge multimodal deep learning model aimed at predicting NAC from accessible H & E images and molecular data. Methods: designed Graph-based Multi-modal Late Fusion (GMLF) that integrates three types data the S1314-COXEN clinical trial: 1) Neural embeddings 182 whole slide (WSIs), generated via...
Skeletal muscle consists of multinucleated cells in which the myonuclei are evenly spaced throughout cell. In Drosophila, this pattern is established embryonic myotubes, where move via microtubules (MTs) and MT-associated protein Ensconsin (Ens)/MAP7, to achieve their distribution. Ens regulates multiple aspects MT biology, but little known about how itself regulated. We find that physically interacts colocalizes with Bsg25D, Drosophila homologue centrosomal Ninein. Bsg25D loss enhances...
Abstract Cellulose nanocrystals (CNC) are linear organic nanomaterials derived from an abundant naturally occurring biopolymer resource. Strategic modification of the primary and secondary hydroxyl groups on CNC introduces amine iodine group substitution, respectively. The (0.285 mmol per gram functionalized (fCNC)) further reacted with radiometal loaded-chelates or fluorescent dyes as tracers to evaluate pharmacokinetic profile fCNC in vivo. In this way, these nanoscale macromolecules can...
Immune checkpoint inhibitors (ICIs) are standard-of-care as first-line (1L) therapy for advanced non-small cell lung cancer (aNSCLC) without actionable oncogenic driver mutations. While clinical trials demonstrated benefits of ICIs over chemotherapy, variation in outcomes across patients has been observed and trial populations may not be representative practice. Predictive models can help understand heterogeneity treatment effects, identify predictors meaningful outcomes, inform decisions....
Abstract Assessing fertilized human embryos is crucial for in vitro-fertilization (IVF), a task being revolutionized by artificial intelligence and deep learning. Existing models used embryo quality assessment chromosomal abnormality (ploidy) detection could be significantly improved effectively utilizing time-lapse imaging to identify critical developmental time points maximizing prediction accuracy. Addressing this, we developed compared various ploidy status across distinct development...
γ-Secretase is a multisubunit complex that catalyzes intramembranous cleavage of transmembrane proteins. The lipid environment forms membrane microdomains serve as spatio-temporal platforms for proteins to function properly. Despite substantial advances in the regulation γ-secretase, effect local microenvironment on γ-secretase poorly understood. Here, we characterized and quantified partitioning its substrates, amyloid precursor protein (APP) Notch, into bilayers using solid-supported model...
Abstract Age-related macular degeneration (AMD) is the leading cause of vision loss. Some patients experience loss over a delayed timeframe, others at rapid pace. Physicians analyze time-of-visit fundus photographs to predict patient risk developing late-AMD, most severe form AMD. Our study hypothesizes that 1) incorporating historical data improves predictive strength late-AMD and 2) state-of-the-art deep-learning techniques extract more image features than clinicians do. We incorporate...
Abstract Parkinson’s disease (PD) is a progressive neurodegenerative disorder marked by significant clinical and progression heterogeneity resulting from complex pathophysiological mechanisms. This study aimed at addressing of PD through the integrative analysis broad spectrum data sources. We analyzed spanning over 5 years individuals with de novo PD, using machine learning deep learning, to characterize individuals’ phenotypic trajectories for subtyping. discovered three pace subtypes...
ABSTRACT The hemizygous R47H variant of TREM2 , a microglia-specific gene in the brain, increases risk for late-onset Alzheimer’s disease (AD). In this study, we identified subpopulation microglia with disease-enhancing proinflammatory signatures associated mutation human AD brains and tauopathy mouse brains. Using transcriptomic analysis at single-nuclei level from patients or common (CV )-TREM2 matched sex, pathology APOE status, found that was cell type- sex-specific transcriptional...