CORE SESSION STREAM 1: DATA/IT
To fully benefit from the vast quantity of data available, we need to overcome the many challenges that restrict discoverability, integration and access to access to large, rich and varied data sets. These sessions will focus on the above for healthcare organizations, research institutes, tech companies, pharma and biotech.
Directors, Heads, VPs and scientists/developers from pharma/biotech, research institutes, healthcare in the following and related areas: Data and technology strategy, regulatory leads, IT, research IT, bioinformatics, machine learning, predictive analytics, bio-computing, data scientists and software engineers.
AI, machine learning, discoverability, integration, unstructured data, data matching, clean-up, data access, data sharing, blockchain, public datasets.
TABLE 1: Solving Disparate Data Discoverability and Integration and Challenges
The benefit of combining biomedical data from multiple sources for efforts in the context of precision medicine is undisputed. The quality and usefulness of data integration depends upon the adoption of standards and suitable mechanisms for researchers to submit and annotate data - so that different types of data are conveniently linked and easily searchable for future querying and analysis.
TABLE 2: Automated Data Matching and Clean Up for Better Insights from Unstructured Data
An extremely large percentage of data remains unstructured with vital information lost inside hand-written notes and medical images. The collection and interpretation of these types of unstructured data is a high-time consuming, and variable in approach. Challenges occur as much of the data is incomplete, inconsistent and it is unclear what information is actually relevant. Much of these processes can be made easier, faster and more accurate with the use of computational methods, automation and the application of machine learning.
TABLE 3: Addressing Constraints of Access and Usability for Private and Public Datasets
Maximising the usability for datasets is a vital component of the data sharing process. Many companies are buying access to public and private data with each set of data carrying individual constraints and pitfalls. A standard of quality and practicality has to be set to ensure the value of each dataset and make best use of rich data.
TABLE 4: Blockchain-Based Data Sharing and Access Frameworks
Blockchain and decentralisation of genomic data storage will allow for the security of personal data and its management by the owner themselves. All exchanges of such data between genome owner and researchers can be traced and may even become incentivised, increasing data sharing and ultimately improving research. However, ensuring this privacy and control over genomic data in practice is one of the many challenges that companies face when considering blockchain technologies.