CASE STUDY: Machine learning for activity prediction, as part of lead compound generation
The Challenge: The ability to quickly iterate multiple large feature sets with the flexibility to test models at scale is a challenge for any data scientist. Read More
BioSymetrics CSO Gabe Musso to Present on Artificial Intelligence and Machine Learning for Biomedical R&D Innovation at the 2019 BIO International Convention
PHILADELPHIA – (June 3, 2019) – Biomedical artificial intelligence (AI) company BioSymetrics will offer a presentation on the use of AI and machine learning for healthcare and biomedical R&D innovation at the 2019 BIO International Convention, which is being held June 3-6 at the Pennsylvania Convention Center in Philadelphia. The company’s Chief Scientific Office, Gabe Musso, will lead the presentation on Wednesday, June 5, at 11:15 am, in Theater 3. Read More
Our own Gabe Musso, CSO of BioSymetrics will be presenting how Augusta™ is being deployed to increase the speed and confidence of machine learning and AI in biomedical use cases.
Theater 3, Level 200
Wednesday From 11:15 AM To 11:30 AM
Looking for something different in the evening?
Why not get out and walk with us as we take a short tour of the historical Society Hill/Olde City section of Philadelphia including the Betsy Ross House, Liberty Bell, Independence Hall, and Head House Square. You can be sure we will end the short walk with a cheesesteak at iconic Jim’s on South Street. (Our CMO grew up in Philly and swears it is better than Pats or Geno’s.)
Challenge: Combine Disparate Data Sets in PreProcessing for ML
Summary: Compelling results show that combining data sources generally allowed better diagnostic performance than with any data set alone (Figures 1&2) Read More
TORONTO – (May 21, 2019) – Biomedical artificial intelligence (AI) company BioSymetrics will be attending Collision 2019, which is being held May 20-23 at the Entercare Centre in Toronto. BioSymetrics is pioneering proprietary data science and machine learning software in order to best optimize medical discovery and life sciences productivity.
The company’s Chief Scientific Officer, Gabe Musso, will be available for meetings at Collision during the HealthConf track on Thursday, May 23, at BioSymetrics’ booth (Booth #B101) and can discuss how BioSymetrics’ inaugural product, the Augusta™ software platform, provides a holistic solution to some of the current industry challenges, including the utilization of good data without bias to feed machine learning models, as well as solving the ability to interrogate multiple models. BioSymetrics simplifies biomedical data science projects by reducing the time spent on data pre-processing, thus ultimately helping to build more effective predictive models that enable solutions to reach the market more quickly.
NOTE: Booth #B101 will be located in a different area on Thursday; please ask for directions at the information desk.
Media & Analyst Briefings
To schedule a briefing with BioSymetrics at Collision, please contact BioSymetrics’ PR team at 774-551-6679 or email@example.com.
For additional information, visit http://biosymetrics.com.
Korie Grill, Account Executive
We were recently invited to join the MaRS innovation hub of Canadian science and tech companies. We are honored and humbled, as the MaRS community is certainly a prestigious group of organizations with a mission to create a better world. Excited to begin our partnership.
It’s very encouraging to see the FDA showing forward thinking when it comes to AI/ML-based diagnostics. Even the best model can be outdated quickly given the changing data landscape, making frameworks like Augusta that allow for model refinement and evolution to be crucial. We welcome this mindset as we work with our partners to build more interpretable and more adaptable diagnostic models.
In any data science pipeline there are a number of options that are selected for data processing (e.g. contrast settings for medical images, data imputation approaches, bandpass filter cut-offs for ecg signals). Typically, these options are selected manually based on previous experience of the Data Scientist or recommendations from previous, similar studies. In contingent AI, any “settable” parameter for data processing, data integration, or feature selection is permuted and the corresponding effects on the downstream predictive model measured. This process is similar to hyperparameter tuning in machine learning, however instead of optimizing only the machine learning model, the entire data science pipeline (including model selection) is subject to optimization.