BioSymetrics Launches Augusta™ Architect

By Augusta™

Biomedical Machine Learning Just Got a Whole Lot Easier

Newest component to Augusta AI platform, the first biomedical machine learning framework, enables users to reduce bias, increase speed to market and improve patient outcomes

BOSTON and NEW YORK – October 29, 2019 – BioSymetrics Inc., a biomedical artificial intelligence (AI) company, today announced the release and general availability of the newest workflow component of its Augusta software platform, Augusta Architect. Augusta is the first biomedical-specific machine learning (ML) framework on the market.

The addition of Augusta Architect brings BioSymetrics’ advanced capabilities directly into the hands of the end user. It enables data scientists in hospitals, healthcare systems and pharmaceutical companies to perform fast, effective data pre-processing and ML through the Augusta platform. With the addition of Augusta Architect, they can optimize processing parameters, reducing bias and improving results for drug discovery/development, diagnostics, and precision medicine. Augusta Architect has a simple syntax that allows users to quickly process and integrate multiple, diverse datatypes and run or compare multiple ML algorithms.

“Augusta Architect is unique in its focus on efficient pre-processing, enabling more transparent model building in biomedicine,” said Dr. Eric Schadt, founder and CEO of Sema4. “We’re excited to be users of the platform.”

Augusta, which launched in 2017, is designed to increase precision and shorten timelines for R&D innovations and discovery. The platform operates across three core modules – Augusta Pre-Processing, Augusta ML and Augusta Architect – and the addition of Augusta Architect allows simpler workflow generation integrating both the pre-processing and ML components. This reduces the time required for users to program the flow of data from ingestion through output.

“Augusta Architect uniquely allows optimization of the entire data science workflow, from pre-processing to machine learning,” said Gabriel Musso, Chief Scientific Officer for BioSymetrics. “Data scientists can devote more of their time to building effective models, and have more confidence in their quest to bring new drugs to market, make faster and more accurate diagnoses, and improve patient outcomes.”

Matt Hickey, founding director of the innovative UK-based health technology company Intacare, said, “We are committed to the effective research and development of precision datasets for predictive analytics to better showcase cancer outcomes for our clients, who demand efficiency as they seek to bring cost-efficient healthcare to their patients. Augusta has supported us in a variety of projects in different settings. In health insurance, BioSymetrics saved us months of man hours and approximately $6.4 million (£5-million) annually through a new cancer service cost-code conformity model. In healthcare research, BioSymetrics has helped us identify correlations between multiple patient-reported outcome measurement (PROM) tools and clinical parameters associated with patients being treated with brachytherapy for prostate cancer, ultimately resulting in the development of a short-form digital PROM with high diagnostic and predictive qualities. Typically, both projects would have required us to perform intensive manual data pre-processing. However, the use of Augusta enabled the company to drive faster, more accurate and meaningful innovation.”

For more information on the Augusta platform, please visit www.biosymetrics.com/augusta. To request a demo of Augusta, including the new Augusta Architect workflow component, please visit www.biosymetrics.com/demorequest.

 

About BioSymetrics

BioSymetrics Inc. is based in New York, Boston, and Toronto. Serving hospitals and health systems and biopharmaceutical, precision medicine, and technology companies, BioSymetrics is empowering healthcare and R&D innovation with leading data science and analytics expertise, focused on making end-to-end machine learning accessible to scientists and life sciences organizations. For more information, visit www.biosymetrics.com and Twitter @biosymetrics.

 

Contact:

Brian Lowe, Assistant Vice President, Elevate Communications, Boston MA | Tel 508-523-4901

Stacy Grisinger, Vice President, Elevate Communications, Boston, MA | Tel 617-861-3654

 

# # #

Case Study: BioSymetrics Identifies $5Million in Savings for Healthcare Network

By Case Study, Value Based Care

USE CASE: Value Based Care

CLIENT: Major UK based Healthcare network in partnership with Intacare.

OVERVIEW: The annual cost of radiotherapy is escalating year-on-year with little visibility of root cause and control.  Maintaining cost efficient healthcare for patients required an investigation of current code/claim and cost data.

GOAL: Identify and quantify potential cost savings of revising existing reimbursement mechanisms.

“Processing the data manually would have required many months of man hours.”

PROBLEM: Data was incongruous.  Each healthcare provider used different systems, taxonomies, codes and cost basis for mapping radiotherapy procedures when submitting claims.

  • 75,000+ claims
  • 1725+ unique narratives
  • 1000’s of individual codes and duplicate codes
  • Data types: text, numeric

SOLUTION: Intacare used Augusta Pre-Processing workflows to quickly normalize procedure and cost data collected from multiple sources. The team also created an automatable workflow to streamline future analysis and report generation.


RESULTS:

  • Identified inefficiencies across 24,281 claims
  • 39% of total claims
  • Projected cost savings: $5Million

CONCLUSION:

Pre-processing of data using Augusta workflows reduced the time to manage the data from multiple weeks to just hours.

Moreover, the workflows are now available within Augusta as standard packages, easily replicated for future projects or if additional data needs to be interrogated.

Download the Intacare case study in PDF format.

Complete the form below.

Identifying and Addressing the Challenges in the Diagnosis of Sepsis

By Blog Post

Sepsis is the leading cause of death in the Intensive Care Unit, and it’s responsible for 1 in 3 hospital deaths. Each hour without treatment increases a patient’s risk of death by 4-8%. Thus, early detection of sepsis is crucial for improving survival.

With the inclusion of advanced data preprocessing and machine learning, our research has been able to better predict which patients will get sepsis during their hospital stay.

In our study, we sought to develop a robust sepsis prediction model using physiological data (vital signs and lab results) from the 2019 PhysioNet Challenge. In the first phase of our analysis, we trained a recurrent neural network using long short term memory (LSTM). While the LSTM parameters themselves can be optimized in well-understood ways to produce a more accurate classifier, the impact of pre-processing parameters on sepsis prediction performance remain largely unknown.

Now entering into our second phase, we are applying Augusta™ on a patient data set consisting of ~40,000 patients with the intent of more systematically considering the impact of upstream decisions made in processing the data before training a model.

Initial results look promising.