Biomedical Data Types

Machine learning holds a lot of promise for a variety of sectors in the biomedical space; however, there have been barriers to entry. The challenges include inability and difficulties in using raw data from multiple sources, inconsistency in data standards and annotations, selection of tools for a particular problem, and dealing with data sets that are too big, too fast, or too complicated to handle with traditional systems.

The use of massive data machine learning in the health and medical field holds much promise, with the deployment of targeted analytics and predictive analytics increasingly leading to life-saving advancements. However, traditional machine learning technologies are simply incompatible with the raw data formats available in the health space, and few, if any, gold standard protocols exist for data standardization, normalization, and harmonization.

To address this challenge, BioSymetrics has created a customized and flexible health and medical processing and modeling environment, which provides data exploration and machine-learning algorithms, and generates real-time updates for massive health data sets. All with the goal of tackling problems that have been impossible to solve to date in ‘Precision Medicine’ and beyond.

AugustaTM provides rapid analytics with unlimited inputs and real-time predictive modeling for use cases such as:

  • Bioinformatics/health informatics, computational biology using:
    • MRI/fMRI and other imaging modalities
    • EEG
    • EKG/ECG
    • Genetics, Proteomics
    • Wearables data
    • EHR/EMR data
  • Real Time Integrative Analytics: combining genomics with clinical and patient monitoring data from your mobile device à towards precision medicine applications in real time
  • Small molecule activity prediction
  • Patient care quality and program analysis
  • Drug discovery and development analysis
  • Hospital telemetry and operational data
  • Health Insurance fraud detection

IoT in Biomedicine

Gartner reports that by 2020 the Internet of Things (IoT) will consist of more than 25 billion devices producing exabytes of data. These smart and connected devices unleash a treasure trove of data, yet business are having trouble extracting actionable insights from this ever-expanding data pool. No other industry is growing more rapidly and has a higher need for ultra-fast, real-time analytics for nearly unlimited data sets.

AugustaTM is uniquely suited for Internet of Things data due to its real time analytics capabilities, and SymetryMLTM’s its incremental and decremental learning and its ability to create thousands of models on the fly. Potential use cases are numerous, but include:

  • Mission critical trends of use, cost reductions, better plans for customers, improvements in margins, next best action
  • Patterns of use, real-time feedback on issues, changes in condition, use of systems
  • Smarter health
  • Wearables data analytics and integration with other data types