Full Transcript
Access the complete word-for-word transcript for easy reference and accessibility.
Hello. Today I’m going to walk through a day in the life of approving parameters for use in charts using the platform. This might sound small, but it’s a critical step to ensure that your data is trustworthy, traceable, and chart ready. The first thing that I’m going to do is navigate to the paper batch record dashboard, and I’ll apply a filter for the last day of parameters.
Here we’ll notice that each parameter has a tag with a confidence score. Green is high confidence; the system is very sure what it read. Medium is yellow; the system is mostly sure, but there might be some uncertainty. It looks like it got this correct. And then low confidence, which is red; the system isn’t confident in what it pulled in. Maybe the handwriting was messy or it was partially cropped.
So on mass, we can approve all of these parameters. And then I’ll go in and change this parameter, which is jotted down incorrectly. All of the actions that I’m doing are part of the audit log and ensure that we are 21 CFR compliant. For this final parameter here, we will be editing this JMZ 18 June Z1 in order to make sure that the correct information is captured.
We use a layout aware OCR in order to contextualize where the snippets are on the location of the document. You can see on the left we have all of our potential snippets, and using correlations between all the snippets, we know where the information is. So if our filter lot information moved step down or step up, we’re still able to contain and contextualize that information.
Once we’ve made the change, I can save it, and this will go in for use of training of the system. So as your operators and QC analysts write more information and the system ingests more of it, the system will get better at understanding what it’s reading. So now that all of my parameters have been approved, I will go to the genealogy to show where it’s been contextualized.
Now we will come into the genealogy to show that the snippets have been moved into the correct node. This genealogy contains both paper records and electronic records, but for today we’re only going to focus on the paper records within the system. We use a pre-processing step in order to contextualize the snippets with the node so that there’s no need of moving information from location to location.
So you can see that we got our site information, now our batch information, our thaw start time; these are all the parameters that we’ve approved previously. Now that all of our parameters have been approved, such as our hold time, we’ll be able to come into our charting module and show a created chart. I will pull up a paper chart, and then we’ll go to a dashboard where we have all of our parameters.
So this is a chart generated using paper records. We can see all of our parameters on the left; the pHs were on the bottom of the page; conductivity and hold time we approved. And we can also do some editing of the chart as well in the form of modifying limits, displays, groups, or rules. But for the purpose of this, we’re just going to focus on the chart.
We can see that our CFL 7788 has made it to the chart. And the next time a batch record is put into the system and the snippets are approved, they will naturally find their way onto this chart as well. This workflow turns messy batch records into clean, trusted parameters which are ready for analysis and decision making. And the best part? The more you use it, the smarter it gets.
Thank you guys for watching, and stay tuned for more walkthroughs on how Mareana makes batch data easier to use every day.
Would you like me to summarize the key features of the platform mentioned in the video?
/ Any questions? We'd love to help
Frequently Asked Questions.
Mareana has helped numerous firms in the Pharmaceutical, Chemical, Medical Device, and Industrial Manufacturing industries.
This video demonstrates how users approve extracted parameters from paper batch records so they can be confidently used in charts, analytics, and decision-making.
They indicate how confident the system is in what it extracted:
-
Green: High confidence
-
Yellow: Medium confidence with minor uncertainty
-
Red: Low confidence, usually due to handwriting or image quality
Yes. All actions are captured in an audit log to support 21 CFR Part 11 compliance, ensuring full traceability and accountability.
The platform uses layout-aware OCR and contextual correlations between snippets, allowing it to correctly identify parameters even if their position changes on the page.
The system supports both, but this video focuses specifically on how paper records are processed and contextualized.
Newly ingested and approved parameters automatically populate existing charts, keeping analyses up to date without rework.
It’s built for operators, QC analysts, and data teams who need reliable batch data for compliance, trending, and decision-making.
It turns messy paper batch records into clean, chart-ready data—saving time, reducing errors, and improving confidence in your insights.
Every action taken—approvals, edits, and corrections—is automatically captured in an audit log. This ensures full traceability of who changed what and when, supporting audit readiness and regulatory compliance.
Yes. Approved parameters are contextualized through genealogy, allowing auditors to trace charted data back to its original paper source and associated batch information.
Genealogy shows how approved parameters are linked to batches, sites, and records, making it easier to understand data context during audits or root cause discussions.
Because parameters are reviewed, approved, logged, and charted as part of daily operations, teams don’t need to reconstruct data history during an audit—it’s already documented.
Contact Us
Would you like to learn more about our pharmaceutical manufacturing solutions? Fill out the form below and one of our experts will reach out to you promptly.