cTAKES is a widely used, open source and free tool for clinical natural language processing. Unlike general purpose NLP tools, cTAKES is specialized for clinical texts, incorporating UMLS resources for finding medical concepts and packaged with machine learning models trained on gold standard clinical texts. CHIP faculty Guergana Savova's many federally funded projects have contributed to the development of cTAKES and the gold standard resources it relies on. These projects include THYME (temporal information extraction module), DeepPhe (cancer deep phenotype extraction), and SHARP (UMLS relation extraction, coreference resolution, and assertion status modules). The original paper describing cTAKES is one of the most highly cited articles in JAMIA since its publication.
C3-PRO / Apple ResearchKit / Android ResearchStack
The Consent, Contact, and Community framework for Patient Reported Outcomes – “C3-PRO” in short – is an open source toolchain that connects, in a standards-compliant fashion, any ResearchKit app to the widely-used clinical research infrastructure Informatics for Integrating Biology and the Bedside (i2b2). C3-PRO leverages the emerging health data standard Fast Healthcare Interoperability Resources ( FHIR) for data representation and over-the-wire format, allowing to easily substitute individual components of the research toolchain with custom components. The toolchain is currently used in the C Tracker and Feverprints ResearchKit studies while an expansion to Android’s ResearchStack and additional modules are in the works.
Clinical Decision Support Hooks provides a way to integrate external advice at the point of care. SMART CDS Hooks enhances the Electronic Health Record system to notify external service when specific activities occur within a user session. For example, services can register to be notified when a new patient record is opened, or when a medication is prescribed. Services can then return “cards” displaying text, actionable suggestions, or links to launch a SMART app from within the workflow. CDS Hooks is being developed through an open collaboration among EHR vendors, decision support vendors, and clinical providers.
Flu Near You
Thousands of individuals submit weekly health reports to benefit their community and track the flu. It's the power of the crowd captured for public health surveillance.
Genomic Information Commons
Inspired by a common vision of accelerated genomic discovery, collaboration, and improved clinical outcomes, leaders at Boston Children’s Hospital (BCH), Cincinnati Children’s Hospital Medical Center (CCHMC), the Children’s Hospital of Philadelphia (CHOP), Washington University at St. Louis (WUSTL), and the University of Pittsburgh Medical Center (UPMC) have come together to create the Genomic Information Commons (GIC). The GIC is an NCATS-funded, multi-institutional, federated genomic data commons. The mission of the GIC is to elucidate the role of genetic variation in human disease by instrumenting the healthcare system to continuously update a widely accessible reference database of highly phenotyped genomic sequences, representative of a broad range of demographics and conditions. The GIC has established globally scalable technologies, policies, and procedures for sharing genomic data, phenotypic data, and biospecimen metadata on broadly consented cohorts, across sites of care.
Health Data Fusion
CHIP researchers and trainees use extraordinary massive datasets in their work. In partnership with Aetna and other payors, we analyze big data on 50 million lives to glean insights into the inner workings of the healthcare system and to discover relationships among diseases, their consequences, and their heritability, not visible at lower scales. The SCILHS network includes electronic health record and patient-reported outcome data on more than 11 million patients across 13 health care systems. The i2b2 system at Boston Children’s Hospital enables datasets on millions of encounters. Through academic relationships with Twitter and Google, CHIP researcher analyze billions of Tweets and searches.
HealthMap brings together disparate data sources to achieve a unified and comprehensive view of the current global state of infectious diseases and their effect on human and animal health. This freely available Web site integrates outbreak data of varying reliability, ranging from news sources (such as Google News) to curated personal accounts (such as ProMED) to validated official alerts (such as World Health Organization). Through an automated text processing system, the data is aggregated by disease and displayed by location for user-friendly access to the original alert. HealthMap provides a jumping-off point for real-time information on emerging infectious diseases and has particular interest for public health officials and international travelers.
Healthmap Flu Trends
The HealthMap team at CHIP, in collaboration with Google, was granted access to the Google Search API in order to model search data. Key initial projects are Healthmap Flu Trends. The project leverages the data from multiple data sources including: FluNearYou.org, athenahealth.com, Google Trends, and Google Flu API Trends, to produce estimates of flu. Next steps are development of an ensemble approach to improve accuracy. CHIP researchers are providing HealthMap Flu Trends to the CDC as a national surveillance data source.
CHIP is a major contributor to the origin and development of the i2b2 and i2b2/TranSMART ecosystems. i2B2 is a scalable computational framework and analytic platflorm that is essentially a sidecar to the electronic health record designed to instrument the health system for research. i2b2/tranSMART is a knowledge module for i2b3 enabling scientists to develop and refine research hypotheses by investigating correlations between genetic and phenotypic data, and assessing their analytical results in the context of published literature and other work. Together, the platforms are deployed across over 300 medical centers.
The POPP system (Prediction of Patient Placement) aims to improve the flow of patients through the Emergency Department (ED) and hospital, by providing decision makers with real-time predictions of future patient disposition. Popp bridges predictive analytics to the point of care. We apply computer models on live data extracted from the Electronic Health Records to forecast not only what patients currently need, but what will they need in the near future - facilitating a smarter and more efficient use of resources. Building upon our predictive models we developed a Dashboard
Precision Link drives coherence between the BCH research and clinical missions by establishing the technology, policies and workflows to engage our patients in research, and treating each visit as an opportunity to build a learning health system culture here at Children’s. Precision Link strives to build a knowledge base of both omics and phenotypic data on the patient's walking through our doors--for use in population research as well as in care. It includes the Biobank for Health Discovery, a protocol for capturing the diversity of our patient population by consenting and storing biospecimens under a broad consent for discovery and care.
SMART Health IT
SMART Health IT is an open, standards-based technology platform that enables innovators to create apps that seamlessly and securely run across the healthcare system. Using an electronic health record (EHR) system or data warehouse that supports the SMART standard, patients, doctors, and healthcare practitioners can draw on this library of apps to improve clinical care, research, and public health. The SMART platform is composed of FHIR and OAuth based standards, open source tools for developers building apps and a publicly accessible app gallery. To date, dozens of clinical applications have been built on this platform, and the five largest EHR vendors have joined forces with the SMART team and HL7 to build SMART interfaces into their products.
SMART on FHIR Genomics
SMART on FHIR Genomics brings precision medicine to the point of care, extending the SMART platform, and bringing genomic data to medical decision making. The specification enables an ecosystem of clinical genomic apps. The SMART on FHIR EMR portion extends FHIR, and includes 1) a set of constraining profiles that lock down optionality and align vocabularies with Meaningful Use requirements, 2) a security layer that provides narrowly-scoped authorization to specific portions of a patient's record, 3) a single-sign-on layer, and 4) a user interface integration layer that allows apps to launch in the context of an existing EHR or patient portal session, conveying the current patient, encounter, and other details of the host environment. See the original paper.