Data Science, Machine Learning
and Reporting in Life Sciences Industry
Strict standards and guidelines apply in the life science industry. Regulatory authorities, such as the European Medicines Agency (EMA) and the U.S. Food and Drug Administration (FDA), require detailed procedures, regulations and documentation that must be adhered to.
Using our experiences, we support our clients in setting up their data science, machine learning and reporting knowhow and solutions, especially in validated system environments.
Typical applications starting with data:
- Integration and harmonising of data sources
- Data preparation and automation for analysis and reporting
- Manual and automated data-entry with control process (double blind)
- Data Science und Machine Learning as a process
- Streaming Analytics
Specific applications focussing on reporting:
- Monthly Trend Reports and Trending (MTR)
- Annual Product Reviews (APR)
- Product Quality Reports (PQR)
- Impurity Profile Reports (IPR)
- Continues Process Verification
- Automated Validation Protocols & Reports
- Stability Study Reports and Trend Analysis
- Automated Release Reports
- Shelf Life and Stability Analysis
- Method Validation & Trending
Applications in different departments:
- Research & Development
- From generic statistics to sample size calculations up to Design of Experiments (DoE)
- Production
- Process analysis, process optimization, process monitoring
- Cross-site process monitoring and reporting
- Quality
- Reporting
- Predictive Maintenance
- Six sigma for the enterprise
Data Science, Machine Learning and Reporting as a process in Life Sciences
Data Science and Machine Learning in Life Science must have a process approach, that includes more than the classic CRISP-Modell:
- the access to whatever data source that is available
- The possibility to transform, blend and enrich the data the way that it becomes analysable
- a data validation step
- the usage of whatever algorithms that makes sense (=Science) including open source software like R or Python
- an audit trail that automatically protocols the whole process
- an electronic approving process that makes it paperless
- a user-, data source-, method/algorithm-version management
- the possibility to deploy results/code to other system and/or interact with them
- a business rule engine
- that is documented systematically
- ONE platform for managing centrally all these functions
With our approach of “data science as a process” and the platform we use, we replace the existing applications with a workflow approach that includes the access to the data sources, the data manipulation, the use of the right method, the analysis, the report and the validation or approval process. The solutions include an audit trail functionality, from the data access on everything automatically “documented”, changes are easy to be implemented and versioned on every stage. That makes it easier and less expensive to validate.
Here you can find a use case of digitalizing CPV and Validated Reporting at an international key-player in pharma/biotech.
Validated Reporting (APR/PQR)
Generating Annual Product Reviews (APR in the USA) or Product Quality Reviews (PQR in Europe) is a very time-consuming and cost-intensive process. Using Statistica, you can automate the APR and PQR creation process to the greatest possible extent.
Shelf-Life Estimation
Companies in the pharmaceutical industry use statistical methods to calculate the runtime of a product or the repeat testing period of an active ingredient during long-term storage. This is because there are strict requirements from the European approval authorities. The Statistica add-in Shelf-Life Estimation enables you to perform automated calculations according to Q1E specifications.
Trending
In order to consistently monitor processes and gain an improved understanding of them, you can use Statistica in an optimal way. Statistica naturally complies with the strict GxP standards.
Pharmaceutical and laboratory solutions
All solutions enable the standardization of statistical data analysis and analysis as well as the automatic generation of reports in the form of tables and / or graphics and thus contribute to process optimization.
Our applications include the trend towards product stability, monitoring of the production environment, process validation and ongoing review, regular product reviews, solution profiling, and assessment and validation of uncertainties of analytical methods.
Contact
Search