A5037135208- Biomedical Text Mining and Ontologies
- Topic Modeling
- Scientific Computing and Data Management
- Machine Learning in Healthcare
- Distributed and Parallel Computing Systems
- Liver Disease Diagnosis and Treatment
- Hepatocellular Carcinoma Treatment and Prognosis
- Dementia and Cognitive Impairment Research
- Genomics and Rare Diseases
- Research Data Management Practices
- Cancer Genomics and Diagnostics
- Bioinformatics and Genomic Networks
- Advanced Text Analysis Techniques
- Radiomics and Machine Learning in Medical Imaging
- Chronic Disease Management Strategies
- Genomics and Phylogenetic Studies
- Semantic Web and Ontologies
- Liver Disease and Transplantation
- Orthopedic Infections and Treatments
- Quality and Safety in Healthcare
- Data Quality and Management
- Advanced Graph Neural Networks
- Ethics in Clinical Research
- Artificial Intelligence in Healthcare and Education
- Algorithms and Data Compression
Medical University of Vienna
2023
Humboldt-Universität zu Berlin
2012-2021
San Francisco VA Health Care System
2020
Kaiser Permanente
2020
Charité - Universitätsmedizin Berlin
2019
Humboldt State University
2009
Research results are primarily published in scientific literature and curation efforts cannot keep up with the rapid growth of literature. The plethora knowledge remains hidden large text repositories like MEDLINE. Consequently, life scientists have to spend a great amount time searching for specific information. enormous ambiguity among most names biomedical objects such as genes, chemicals diseases often produces too unspecific search results. We present GeneView, semantic engine...
PURPOSE Precision oncology depends on the availability of up-to-date, comprehensive, and accurate information about associations between genetic variants therapeutic options. Recently, a number knowledge bases (KBs) have been developed that gather such basis expert curation scientific literature. We performed quantitative qualitative comparison Clinical Interpretations Variants in Cancer, OncoKB, Cancer Gene Census, Database Curated Mutations, CGI Biomarkers (the cancer genome interpreter...
One of the greatest strengths artificial intelligence (AI) and machine learning (ML) approaches in health care is that their performance can be continually improved based on updates from automated data. However, ML models are currently essentially regulated under provisions were developed for an earlier age slowly updated medical devices-requiring major documentation reshape revalidation with every update model generated by algorithm. This creates minor problems will retrained only...
Clinically significant posthepatectomy liver failure (PHLF B+C) remains the main cause of mortality after major hepatic resection. This study aimed to establish an APRI+ALBI, aspartate aminotransferase platelet ratio (APRI) combined with albumin-bilirubin grade (ALBI), based multivariable model (MVM) predict PHLF and compare its performance indocyanine green clearance (ICG-R15 or ICG-PDR) albumin-ICG evaluation (ALICE).
With the increasing popularity of scientific workflows, public repositories are gaining importance as a means to share, find, and reuse such workflows. As sizes these grow, methods compare workflows stored in them become necessity, for instance, allow duplicate detection or similarity search. Scientific complex objects, their comparison entails number distinct steps from comparing atomic elements whole. Various studies have implemented workflow came up with often contradicting conclusions...
Abstract Objective We present the Berlin-Tübingen-Oncology corpus (BRONCO), a large and freely available of shuffled sentences from German oncological discharge summaries annotated with diagnosis, treatments, medications, further attributes including negation speculation. The aim BRONCO is to foster reproducible openly research on Information Extraction medical texts. Materials Methods consists 200 manually deidentified cancer patients. Annotation followed structured quality-controlled...
Until recently, genomics has concentrated on comparing sequences between species. However, due to the sharply falling cost of sequencing technology, studies populations individuals same species are now feasible and promise advances in areas such as personalized medicine treatment genetic diseases. A core operation is read mapping, i.e., finding all parts a set genomes which within edit distance k given query sequence ( -approximate search). To achieve sufficient speed, current algorithms...
Abstract Background Aspartate aminotransferase/platelet ratio index (APRI) and albumin–bilirubin grade (ALBI) are validated prognostic indices implicated as predictors of postoperative liver dysfunction after hepatic resection. The aim this study was to evaluate the relevance combined APRI/ALBI score for clinically meaningful outcomes. Methods Patients undergoing hepatectomy were included from American College Surgeons National Surgical Quality Improvement Program database. association...
Abstract Vast amounts of medical information are still recorded as unstructured text. The knowledge contained in this textual data has a great potential to improve clinical routine care, support research, and advance personalization medicine. To access knowledge, the underlying be semantically integrated – an essential prerequisite which is extraction from documents. A body work, good selection openly available tools for semantic integration domain exist, yet almost exclusively English...
Scientific workflows have become a valuable tool for large-scale data processing and analysis. This has led to the creation of specialized online repositories facilitate workflow sharing reuse. Over time, these grown sizes that call advanced methods support discovery, in particular effective similarity search. Here, we present novel intuitive measure is based on layer decomposition. Layer decomposition accounts directed dataflow underlying scientific workflows, property which not been...
Diagnosis and treatment decisions in cancer increasingly depend on a detailed analysis of the mutational status patient's genome. This relies previously published information regarding association variations to disease progression possible interventions. Clinicians large degree use biomedical search engines obtain such information; however, vast majority scientific publications focus basic science have no direct clinical impact. We develop Variant-Information Search Tool (VIST), engine...
The decreasing cost of obtaining high-quality calls genomic variants and the increasing availability clinically relevant data on such are important drivers for personalized oncology. To allow rational genome-based decisions in diagnosis treatment, clinicians need intuitive access to up-to-date comprehensive variant information, encompassing, instance, prevalence populations diseases, functional impact at molecular level, associations druggable targets, or results from clinical trials. In...
<ns4:p><ns4:bold>Background:</ns4:bold> Timely diagnosis of dementia is a policy priority in the United Kingdom (UK). Primary care physicians receive incentives to diagnose dementia; however, 33% patients are still not receiving diagnosis. We explored automating early detection using data from patients’ electronic health records (EHRs). investigated: a) how machine-learning model could accurately identify before physician; b) if models be tuned for subtype; and c) what best clinical features...
Abstract Tables are a common way to present information in an intuitive and concise manner. They used extensively media such as scientific articles or web pages. Automatically analyzing the content of tables bears special challenges. One most basic tasks is determination orientation table: In column tables, columns represent one entity with different attribute values rows; row vice versa, matrix give on pairs entities. this paper, we address problem classifying given table into three layouts...
Text Mining has established itself as a valuable tool for knowledge extraction in many commercial and scientific areas. Accordingly, large number of different methods have been developed focusing on broad range tasks. We report novel system architecture that is fundamentally service-based, i.e., it models implements text mining routines independent, yet federated services. The several layers: (1) Base services perform various fundamental They all implement fixed interface but keep their...
Tables are a popular and efficient means of presenting structured information. They used extensively in various kinds documents including web pages. display information as two-dimensional matrix, the semantics which is conveyed by mixture structure (rows, columns), headers, caption, content. Recent research has started to consider tables first class objects, not just an addendum texts, yielding interesting results for problems like table matching, completion, or value imputation. All these...