Life sciences organizations face a difficult balancing act: protecting patient privacy, meeting evolving global clinical trial transparency requirements, and maintaining sufficient data utility to support meaningful research. At the same time, clinical datasets and study documents are growing in volume and complexity. At the same time, manual de-identification can be slow, inconsistent, and vulnerable to costly errors. Teams also need defensible audit trails, clear role-based controls, and efficient workflows that enable them to produce submission-ready outputs without unnecessary delays.
Blur™ is a standalone SaaS platform designed to address these challenges. It successfully handles this by bringing together data de-identification, document anonymization, re-identification risk assessment, and workflow governance into a single configurable environment. It supports global regulatory expectations, including Health Canada, US FDA, EMA, EU CTR, and GDPR, while helping organizations maintain data usability and strengthen audit readiness.
With the release of Blur™ 2.10, d-wise by Instem is introducing a broad set of enhancements across CSR anonymization, data de-identification, risk simulation, administration, performance, and security. The update focuses on speeding up day-to-day work, improving control over complex document tasks, and helping teams make more informed anonymization decisions.
Accelerating Repetitive CSR Anonymization Tasks with Blur™
Clinical Study Reports frequently contain repeated content in the same position across many pages. Previously, users might have needed to create similar annotations one page at a time. This process can become very time-consuming in long, highly structured documents.
Blur™ 2.10 introduces a new “Repeat on Pages” capability in batch annotation mode. Users can now draw an annotation once and apply it across a specified list of pages or page ranges. These repeated annotations can also be grouped together. This enables consistent updates across the group when required.
The same page-range approach extends to out-of-scope content. Users can mark multiple pages or ranges as out of scope without relying on document bookmarks or sections. Additional configuration options provide control over whether manually drawn out-of-scope areas remain as drawn or expand to the full page. The configurations also provide whether page numbers are included in placeholder text, and how consecutive out-of-scope sections are represented in downloaded documents.
Together, these improvements help reduce repetitive effort. They also give users greater flexibility when working with CSRs whose bookmark structures are not perfectly consistent.
Greater Control Over Annotations and Document Output
Blur™ 2.10 also strengthens annotation management. When an annotation belongs to a connected group, users can choose whether to delete only the selected item or the entire group. The existing “Apply Action to All” capability now supports deletion as well as accept, reject, and defer actions. ON top of that, it supports a new confirmation step helps prevent accidental document-wide changes.
Output quality has also been enhanced. Blur’s “Hiding in Plain Sight” functionality now provides improved matching of original font type, color, weight, and angle when replacement text is generated. A new auto-size option for proposed downloads can instruct a PDF viewer to size custom replacement text automatically within the annotation area. Blur™ also improves handling of non-standard and partially embedded fonts. This helps reduce the risk of character loss or visually inconsistent replacement text.
For overlapping annotations, new project options allow organizations to define whether eclipsed or overlapped annotations should be removed from final downloads, proposed downloads, both, or neither. Teams can also configure the overlap percentage that triggers removal, offering more precise control over the final document.
More Precise Data De-identification and Risk analysis
Blur™ 2.10 adds flexibility to numeric banding. This is a very common method for reducing disclosure risk while retaining analytical usefulness. Administrators and project managers can now choose how decimal values are handled when integral numeric bands are applied. This either preserves the existing “Round Up” behavior or uses “Round Nearest”. This gives teams more control over how values are grouped and can help avoid unintentionally placing a value into a higher band.
Risk assessment workflows are also more portable. Primary and outlier simulation results can now be exported to Excel. This allows users to sort, filter, compare, and further examine results outside the application. This makes it easier to evaluate combinations of anonymization methods. It is also easier to identify an approach that appropriately balances privacy risk with data utility.
In CSR workflows, approximate “or fuzzy” string matching has been expanded and made more configurable. Rare disease terms can now use the selected matching strategy. Also, project settings allow teams to define the maximum number of character differences and minimum term length used for approximate matching. These controls help organizations tune detection to their documents and terminology while balancing sensitivity and false positives.
Blur™: A faster, More Manageable Administrative Experience
Large libraries of numeric bands and term lists are essential for repeatable, standardized de-identification. However, they can also create administrative overhead. Blur™ 2.10 improves loading performance for both numeric bands and term lists. A loading indicator provides clearer feedback while term lists are being retrieved. Term lists can also be presented with the newest items first to make recent additions easier to find.
Project discovery is also simpler. Keyword searches can now include users assigned to project roles. This helps administrators and team members locate projects by the people involved, not just by project names, study IDs, statuses, descriptions, or package names.
These changes may appear small individually, but collectively they reduce friction in frequently used workflows. They also help users spend more time on review and decision-making rather than navigation and waiting.
Reliability Improvements for Real-World Document Variation
Blur™ 2.10 also addresses several issues encountered in real-world study documents. It now supports uploading and processing PDFs with named destinations in their bookmarks. It also improves recognition of ages written in “NN-year old” format, consistently detects dates labeled with variants of “data cut off,” and ensures manually created out-of-scope sections can retain independent text. These fixes can reduce preprocessing and rework when source documents contain less-standard structures or phrasing.
Helping Transparency Teams Move With Greater Confidence
Blur™ 2.10 is designed to make Clinical Trial Transparency work more efficiently without sacrificing precision or control. Faster batch annotation, more flexible out-of-scope handling, improved document output, configurable “fuzzy” matching, exportable risk results, and stronger administration and security controls all contribute to a more streamlined end-to-end workflow.
For pharmaceutical sponsors, biotechnology companies, CROs, and transparency teams, the value is practical. Blur™ reduces repetitive tasks, improves decision-making, delivers more consistent outputs, and provides greater confidence that patient privacy is protected while useful clinical information remains available to regulators and researchers.
Reach out to learn how Blur can help your organization de-identify clinical trial data and documents, assess re-identification risk, and prepare audit-ready outputs for global transparency requirements. Don’t forget to follow us on LinkedIn!
