A5086791063- Machine Learning in Healthcare
- Sepsis Diagnosis and Treatment
- Scientific Computing and Data Management
- Artificial Intelligence in Healthcare and Education
- Research Data Management Practices
- Artificial Intelligence in Healthcare
- Radiomics and Machine Learning in Medical Imaging
- Healthcare Technology and Patient Monitoring
- Electronic Health Records Systems
- COVID-19 diagnosis using AI
- Ethics in Clinical Research
- Lung Cancer Diagnosis and Treatment
- Respiratory Support and Mechanisms
- Biomedical Text Mining and Ontologies
- Cardiac, Anesthesia and Surgical Outcomes
- Computational Physics and Python Applications
- Emergency and Acute Care Studies
- Biomedical and Engineering Education
- Time Series Analysis and Forecasting
- Intensive Care Unit Cognitive Disorders
- Data Visualization and Analytics
- Meta-analysis and systematic reviews
- Genetics, Bioinformatics, and Biomedical Research
- AI in cancer detection
- Frailty in Older Adults
Massachusetts Institute of Technology
2016-2025
Memorial Sloan Kettering Cancer Center
2024
Moscow Institute of Thermal Technology
2017-2024
University College London
2012-2022
Aarhus University Hospital
2022
Institut Pasteur
2022
Software (Spain)
2022
English Heritage
2022
Harvard–MIT Division of Health Sciences and Technology
2015-2021
Beth Israel Deaconess Medical Center
2018
MIMIC-III ('Medical Information Mart for Intensive Care') is a large, single-center database comprising information relating to patients admitted critical care units at large tertiary hospital. Data includes vital signs, medications, laboratory measurements, observations and notes charted by providers, fluid balance, procedure codes, diagnostic imaging reports, hospital length of stay, survival data, more. The supports applications including academic industrial research, quality improvement...
Critical care patients are monitored closely through the course of their illness. As a result this monitoring, large amounts data routinely collected for these patients. Philips Healthcare has developed telehealth system, eICU Program, which leverages to support management critically ill Here we describe Collaborative Research Database, multi-center intensive unit (ICU)database with high granularity over 200,000 admissions ICUs by Programs across United States. The database is deidentified,...
Abstract Chest radiography is an extremely powerful imaging modality, allowing for a detailed inspection of patient’s chest, but requires specialized training proper interpretation. With the advent high performance general purpose computer vision algorithms, accurate automated analysis chest radiographs becoming increasingly interest to researchers. Here we describe MIMIC-CXR, large dataset 227,835 studies 65,379 patients presenting Beth Israel Deaconess Medical Center Emergency Department...
Digital data collection during routine clinical practice is now ubiquitous within hospitals. The contains valuable information on the care of patients and their response to treatments, offering exciting opportunities for research. Typically, are stored archival systems that not intended support These often inaccessible researchers structured optimal storage, rather than interpretability analysis. Here we present MIMIC-IV, a publicly available database sourced from electronic health record...
Lack of reproducibility in medical studies is a barrier to the generation robust knowledge base support clinical decision-making. In this paper we outline Medical Information Mart for Intensive Care (MIMIC) Code Repository, centralized code generating reproducible on an openly available critical care dataset.Code provided load data into relational structure, create extractions data, and reproduce entire analysis plans including research studies.Concepts extracted include severity illness...
Chest radiography is an extremely powerful imaging modality, allowing for a detailed inspection of patient's thorax, but requiring specialized training proper interpretation. With the advent high performance general purpose computer vision algorithms, accurate automated analysis chest radiographs becoming increasingly interest to researchers. However, key challenge in development these techniques lack sufficient data. Here we describe MIMIC-CXR-JPG v2.0.0, large dataset 377,110 x-rays...
Abstract Introduction The neutrophil-to-lymphocyte ratio (NLR) is a biological marker that has been shown to be associated with outcomes in patients number of different malignancies. objective this study was assess the relationship between NLR and mortality population adult critically ill patients. Methods We performed an observational cohort unselected intensive care unit (ICU) based on records large clinical database. computed individual patient categorized by quartile ratio. association...
Objectives: To evaluate the relative validity of criteria for identification sepsis in an ICU database. Design: Retrospective cohort study adult admissions from 2008 to 2012. Setting: Tertiary teaching hospital Boston, MA. Patients: Initial admission all patients noncardiac surgical ICUs. Interventions: Comparison five different algorithms retrospectively identifying sepsis, including Sepsis-3 criteria. Measurements and Main Results: 11,791 23,620 (49.9%) met study. Within this subgroup,...
In quantitative research, understanding basic parameters of the study population is key for interpretation results. As a result, it typical first table ("Table 1") research paper to include summary statistics data. Our objectives are 2-fold. First, we seek provide simple, reproducible method providing papers in Python programming language. Second, use package improve quality reported papers.The tableone developed following good practice guidelines scientific computing and all code made...
IMPORTANCE Time-limited trials of intensive care are commonly used in patients perceived to have a poor prognosis.The optimal duration such is unknown.Factors as cancer diagnosis associated with clinician pessimism and may affect the decision limit independent patient's severity illness.OBJECTIVE To identify for short-term mortality critically ill cancer.DESIGN, SETTING, AND PARTICIPANTS Decision analysis using state-transition microsimulation model was performed simulate hospital course...
The digitization of health records and growing availability tumour DNA sequencing provide an opportunity to study the determinants cancer outcomes with unprecedented richness. Patient data are often stored in unstructured text siloed datasets. Here we combine natural language processing annotations1,2 structured medication, patient-reported demographic, registry genomic from 24,950 patients at Memorial Sloan Kettering Cancer Center generate a clinicogenomic, harmonized oncologic real-world...
We present the INSPIRE dataset, a publicly available research dataset in perioperative medicine, which includes approximately 130,000 surgical operations at an academic institution South Korea over ten-year period between 2011 and 2020. This comprehensive patient characteristics such as age, sex, American Society of Anesthesiologists physical status classification, diagnosis, procedure code, department, type anaesthesia. The also vital signs operating theatre, general wards, intensive care...
Related ArticleThis is a corrected version. See correction statement in: http://medinform.jmir.org/2015/1/e6/
The ability of caregivers and investigators to share patient data is fundamental many areas clinical practice biomedical research. Prior sharing, it often necessary remove identifiers such as names, contact details, dates in order protect privacy. Deidentification, the process removing identifiers, challenging, however. High-quality annotated for developing models scarce; target are highly heterogenous (for example, there uncountable variations names); anything less than perfect sensitivity...
Abstract Background Valvular heart disease is a highly prevalent condition that contributes significantly to cardiovascular morbidity and mortality. Echocardiography the gold-standard for valvular evaluation; however, diagnosis may be delayed due limited resources or expertise. Deep learning analysis of electrocardiography (ECG) has demonstrated ability identify subtle structural functional abnormalities within system. Purpose The aim this study was evaluate accuracy deep algorithm from...
Fundamental quality, safety, and cost problems have not been resolved by the increasing digitization of health care. This has progressed alongside presence a persistent divide between clinicians, domain experts, technical such as data scientists. The disconnect clinicians scientists translates into waste research care resources, slow uptake innovations, poorer outcomes than are desirable achievable. can be narrowed creating culture collaboration these two disciplines, exemplified events...