Ephir - Health IT

Cloud Research Lab

Using the Amazon Cloud, we created a centralized research facility for scientists to perform statistical epidemiological research.

Over 100 scientists from around the world simultaneously conduct epidemiological research using our Research Lab.
Data of over 100M patients are available within the Research Lab.
The Lab allows the automation of treatment-outcome association analysis using 20 statistical methods
15 scientific papers have been published based on the conducted research
The Lab allows the massively parallel launch of hundreds of experiments
It is currently the largest SAS/R infrastructure on the cloud
It provides HIPAA-compliant data security to safeguard sensitive Patient Health Information (PHI)

A real time data warehouse that consolidates information from a variety of clinical sources to present a unified view of clinical studies.

Consolidated and formatted data such as patient demographics, ICD-9 encounters, lab results, pharmacy information, genomic data are integrated into a centralized data warehouse
The CDISC standards-based warehouse is structured in ODM format.
It allows to perform data aggregation and conversion to generate SDTM format of all collected information
It supports the automation of manual processes of data conversion and the load to creation of submission material to FDA
It provides multiple predefined visualizations
It supports decision making through reporting and browsing: Patient Profiling, Summary Listings, cohort visualizations
It allows to create tables and listings for reporting to FDA
It has SAS Integration
It is Part 11 compliant and fully documented for easy validation at the client site

Data Visualization application provides real-time access to statistical results

Centralized repository of gene information resource portal to facilitate research

The solution provides mechanism to collect gene sequences and related information from private and public sources
It creates comprehensive web services architecture access central gene database
It automates and loads and distributes semi-manually gene sequences from corporate and open source resources
It builds mechanisms for information extraction and search allowing third party developers to build interface software to other applications, e.g. Sequence Exploration, Data Mining, etc.
The system performs gene annotation to mark up descriptive notations such as names, functions and biological features
It is used at to five large pharmaceutical companies

Centralized repository of gene information resource portal to facilitate search and research

We created a corporate workflow system to submit bulletin recommendations to doctors generated by prospective analytics (not in the chart)
This provides a web-based mechanism to report PHI over Intranet and non-PHI information using extranet process

We developed applications that replaced a large set of paper-based medical SOPs for doctors and nurses with an electronic version
The application distributed to over 80% of the hospitals in the US

We created an automated solution to collect data from 40 large health organization. Organized data into a consistent architecture
This provides a web based system management and reporting