A5074553816- Dementia and Cognitive Impairment Research
- Health, Environment, Cognitive Aging
- Machine Learning in Healthcare
- Bioinformatics and Genomic Networks
- Biomedical Text Mining and Ontologies
- Alzheimer's disease research and treatments
- Health Systems, Economic Evaluations, Quality of Life
- Functional Brain Connectivity Studies
- Gaussian Processes and Bayesian Inference
- Genomics and Rare Diseases
- Birth, Development, and Health
- Epigenetics and DNA Methylation
- Ethics and Social Impacts of AI
- Nutritional Studies and Diet
- Autism Spectrum Disorder Research
- Cardiovascular Health and Risk Factors
- Parkinson's Disease Mechanisms and Treatments
- Topic Modeling
- Pregnancy and preeclampsia studies
- Cardiovascular Issues in Pregnancy
- Model Reduction and Neural Networks
- Body Contouring and Surgery
- Digital Mental Health Interventions
- Chronic Disease Management Strategies
- Frailty in Older Adults
Massachusetts General Hospital
2023-2025
Harvard University
2023-2025
Fraunhofer Institute for Algorithms and Scientific Computing
2019-2024
University of Bonn
2019-2024
Bonn Aachen International Center for Information Technology
2019-2024
Fraunhofer Society
2022
An increasing interest in a healthy lifestyle raises questions about optimal body weight. Evidently, it should be clearly discriminated between the standardised "normal" weight and individually To this end, basic principle of personalised medicine "one size does not fit all" has to applied. Contextually, but e.g. borderline mass index might for one person apparently suboptimal another strongly depending on individual genetic predisposition, geographic origin, cultural nutritional habits...
Abstract The anticipation of progression Alzheimer’s disease (AD) is crucial for evaluations secondary prevention measures thought to modify the trajectory. However, it difficult forecast natural AD, notably because several functions decline at different ages and rates in patients. We evaluate here AD Course Map, a statistical model predicting neuropsychological assessments imaging biomarkers patient from current medical radiological data early stages. tested method on more than 96,000...
Abstract Non-communicable chronic diseases (NCDs) have become a major global health concern. They constitute the leading cause of disabilities, increased morbidity, mortality, and socio-economic disasters worldwide. Medical condition-specific digital biomarker (DB) panels emerged as valuable tools to manage NCDs. DBs refer measurable quantifiable physiological, behavioral, environmental parameters collected for an individual through innovative technologies, including wearables, smart...
Artificial intelligence (AI) approaches pose a great opportunity for individualized, pre-symptomatic disease diagnosis which plays key role in the context of personalized, predictive, and finally preventive medicine (PPPM). However, to translate PPPM into clinical practice, it is utmost importance that AI-based models are carefully validated. The validation process comprises several steps, one testing model on patient-level data from an independent cohort study. recruitment criteria can bias...
Cardiovascular disease remains the leading cause of burden globally with far-reaching consequences including enormous socio-economic to healthcare and society at large. health is decisive for reproductive function, healthy pregnancy postpartum. During pregnancy, maternal cardiovascular system exposed highly increased haemodynamic stress that significantly impacts status mother offspring. Resulting from sub-optimal conditions overlooked in pre-pregnancy time, progressive abnormalities can be...
Parkinson's disease (PD) is a highly heterogeneous both with respect to arising symptoms and its progression over time. This hampers the design of modifying trials for PD as treatments which would potentially show efficacy in specific patient subgroups could be considered ineffective trial cohort. Establishing clusters patients based on their patterns help disentangle exhibited heterogeneity, highlight clinical differences among subgroups, identify biological pathways molecular players...
Abstract Introduction Numerous studies have collected Alzheimer's disease (AD) cohort data sets. To achieve reproducible, robust results in data‐driven approaches, an evaluation of the present landscape is vital. Methods Previous efforts relied exclusively on metadata and literature. Here, we evaluate by directly investigating nine patient‐level sets generated major clinical studies. Results The investigated cohorts differ key characteristics, such as demographics distributions AD...
Abstract Introduction Given study‐specific inclusion and exclusion criteria, Alzheimer's disease (AD) cohort studies effectively sample from different statistical distributions. This heterogeneity can propagate into cohort‐specific signals subsequently bias data‐driven investigations of progression patterns. Methods We built multi‐state models for six independent AD datasets to statistically compare patterns across them. Additionally, we propose a novel method clustering cohorts with regard...
The majority of biomedical knowledge is stored in structured databases or as unstructured text scientific publications. This vast amount information has led to numerous machine learning-based biological applications using either through natural language processing (NLP) data graph embedding models. However, representations based on a single modality are inherently limited.To generate better knowledge, we propose STonKGs, Sophisticated Transformer trained and Knowledge Graphs (KGs)....
Verbal communication is one of the most sophisticated human motor skills reflecting both-the mental and physical health an individual. Voice parameters quality changes are usually secondary towards functional and/or structural laryngological alterations under specific systemic processes, syndrome pathologies. These include but not restricted to dry mouth Sicca syndromes, body dehydration, hormonal linked pubertal, menopausal, andropausal status, respiratory disorders, gastrointestinal...
Background: Accessible datasets are of fundamental importance to the advancement Alzheimer’s disease (AD) research. The AddNeuroMed consortium conducted a longitudinal observational cohort study with aim discover AD biomarkers. During this study, broad selection data modalities was measured including clinical assessments, magnetic resonance imaging, genotyping, transcriptomic profiling, and blood plasma proteomics. Some collected were shared third-party researchers. However, incomplete,...
Abstract Background Currently, Alzheimer’s disease (AD) cohort datasets are difficult to find and lack across-cohort interoperability, the actual content of publicly available often only becomes clear third-party researchers once data access has been granted. These aspects severely hinder advancement AD research through emerging data-driven approaches such as machine learning artificial intelligence bias current findings towards few commonly used, well-explored cohorts. To achieve robust...
Individual organizations, such as hospitals, pharmaceutical companies, and health insurance providers, are currently limited in their ability to collect data that fully representative of a disease population. This can, turn, negatively impact the generalization statistical models scientific insights. However, sharing across different organizations is highly restricted by legal regulations. While federated access concepts exist, they technically organizationally difficult realize. An...
Abstract Dementia probably due to Alzheimer’s disease is a progressive condition that manifests in cognitive decline and impairs patients’ daily life. Affected patients show great heterogeneity their symptomatic progression, which hampers the identification of efficacious treatments clinical trials. Using artificial intelligence approaches enable enrichment trials serves promising avenue identify treatments. In this work, we used deep learning method cluster multivariate trajectories 283...
Abstract Motivation The importance of clinical data in understanding the pathophysiology complex disorders has prompted launch multiple initiatives designed to generate patient-level from various modalities. While these studies can reveal important findings relevant disease, each study captures different yet complementary aspects and modalities which, when combined, a more comprehensive picture disease etiology. However, achieving this requires global integration across studies, which proves...
Abstract Background The integration of heterogeneous, multiscale, and multimodal knowledge data has become a common prerequisite for joint analysis to unravel the mechanisms aetiologies complex diseases. Because its unique ability capture this variety, Biological Expression Language (BEL) is well suited be further used as platform semantic harmonization in networks systems biology. Results We have developed numerous independent packages capable downloading, structuring, serializing various...
Abstract INTRODUCTION While the influence of cross‐sectional β‐amyloid (Aβ) on longitudinal changes in cognition is well established, change‐on‐change between Aβ and less explored. METHODS A series bivariate latent change score models (LCSM) examined relationship 11 C‐Pittsburgh Compound‐B (PiB) positron emission tomography (PET) Preclinical Alzheimer's Cognitive Composite‐5 (PACC‐5) while adjusting for covariates, including medial temporal lobe (MTL) tau‐PET burden. We selected 352...
As machine learning and artificial intelligence increasingly attain a larger number of applications in the biomedical domain, at their core, utility depends on data used to train them. Due complexity high dimensionality data, there is need for approaches that combine prior knowledge around known biological interactions with patient data. Here, we present CLinical Embedding Patients (CLEP), novel approach generates new representations by leveraging both patient-level First, given dataset...
Abstract Dementia probably due to Alzheimer’s disease (AD) is a progressive condition that manifests in cognitive decline and impairs patients’ daily life. Affected patients show great heterogeneity their symptomatic progression, which hampers the identification of efficacious treatments clinical trials. Using artificial intelligence approaches enable enrichment trials serves promising avenue identify treatments. In this work, we used deep learning method cluster multivariate trajectories...
Abstract INTRODUCTION Currently, AD cohort datasets are difficult to find, lack across-cohort interoperability, and the content of shared often only becomes clear third-party researchers once data access has been granted. METHODS We accessed systematically investigated 20 major on data-level. A medical professional a specialist manually curated semantically harmonized acquired datasets. developed platform that facilitates exploration. RESULTS present ADataViewer, an interactive exploration...
Abstract Background Previous models of Alzheimer’s disease (AD) progression were primarily hypothetical or based on data originating from single cohort studies. However, datasets are subject to specific inclusion and exclusion criteria that influence the signals observed in their collected data. Furthermore, each study measures only a subset AD-relevant variables. To gain comprehensive understanding AD progression, heterogeneity robustness estimated patterns must be understood, complementary...