Dr. Alicia Browner participates in Artificial Intelligence Panel at SQA CVIC Meeting

Alicia Browner, Ph.D., Chief Technology Officer at Prelude Dynamics, presented today at the Society of Quality Assurance CVIC Meeting. Dr. Browner was selected to present as one of the panelist on the topic of artificial intelligence (AI) in health care and clinical trials.

The presentation focused on approaches to validation and how the FDA treats artificial intelligence technologies. The ultimate goal of AI is to get the computer to think and act like humans with the focus on thinking and acting rationally in order to be able to perform some of the functions we currently rely on humans to do. Currently, popular AI include Alexa, Siri and Sophia, all of which can do the simple task of looking for and retrieving information and sometime taking action such as playing a song. Yet, AI could do so much more.

Dr. Browner worked with neural networks and optimization in aerospace before founding Prelude Dynamics. She is currently developing an AI risk-based monitoring (RBM) tool to aid clinical study monitors in managing and monitoring the data. VISION™ Clinical Trial Optimization Platform already contains an innovative technology in its dictionary coding function. The function instantaneously matches the best options as the user types, allowing them to quickly select from a dictionary like WHO Drug, that contains over 300,000 entries. This IntelliMatch functionality is also used in search, reports, and exports as well as in the combo fields. Prelude is also planning to augment this technology with a learning algorithm to help offer the most prevalent options.

Dr. Browner said, “AI is going to change the landscape of clinical research. At Prelude we are strategically implementing our AI strategy and helping lead the industry into its implementation.”

Applications in artificial intelligence use methodologies to present knowledge, plan, process natural language, perceive things like speech, facial and object recognition, move (motion, as in robotics), and have social intelligence. One of the challenges, as an example, is that communication is known to be 80% or more non-verbal. So how do you teach a computer to understand that? The current strategy is machine learning.

Machine learning is the process of teaching a computer to learn by providing it with information and learning experiences. Initially, programmers teach the computer to learn through supervised learning. The end goal is for the computer to learn unsupervised. In order to accomplish this, technologies such as Baysian Networks, Neural Networks and Evolutionary Algorithms can be applied to assist decision making.

First CCDM designation received by Staff

Congratulations to Dr. Karina Loyo for becoming the first in our office to be a Certified Clinical Data Manager by the Society of Clinical Data Managers. Dr. Loyo passed the certification exam on August 22nd. This represents the first step towards SCDM industry partner certification for Prelude.

Prelude’s staff are integral in implementing best clinical data management practices to Client’s studies. Gaining recognition through SCDM’s CCDM certification provides an independent qualification and assurance to our clients about the quality of the staff they interact with and who manage their studies.

Richard Tieken, CEO, said, “We want every client to know they are in the best hands possible. Providing staff who are Certified Clinical Data Managers is just one way in which we continue to elevate the bar of excellence for our customer service, and show our commitment to the industry to always be at the leading edge of technology.” Prelude expects for other project managers to obtain their CCDM in the near future and is on track to achieve SCDM’s designation of CCDM® Industry Partner in the following months.

Prelude Commits to Excellence in Data Management

Clients trust Prelude’s expertise in helping them with their electronic data capture (EDC) and data management needs. This trust has been honed through years of dedication to providing the gold-standard in EDC functionality.  We have taken VISION™ beyond EDC to provide an all-in-one comprehensive platform that provides not only EDC, but also Clinical Trial Management System (CTMS) and Clinical Data Management System (CDMS) functionality.

Now, Prelude is taking our excellence in Data Management to the next level by committing to obtain recognition from the Society of Clinical Data Management as an industry partner in Data Management Excellence. Obtaining this designation is no easy feat. At least 25% of all personnel or a minimum of 10 employees engaged in data management must be Certified Clinical Data Managers (CCDM). The CCDM certification exam is comprehensive and rigorous. The passing rate is around 55% of those who take the exam. Currently, there are less than 20 industry partners.  Achieving it truly represents excellence in data management.

Prelude is committed to obtaining the SCDM CCDM Data Management Industry Partner status in the next 6 months.

Richard Tieken, CEO at Prelude said, “Showing our clients that they are dealing with only the best and most qualified company is as important to us as making sure our Client’s data is clean and safe. Becoming an industry partner is a natural step in letting our Clients know how committed we are to ensuring excellence in our software system and in their data.”

About half of Prelude’s staff is directly engaged in data management functions which include project management and building of studies. All project managers currently meet the qualifications to take the certification exam and most of them have already been approved to take it.

The Future of Visual Study Archives

Your EDC study has come to a close. All data has been finalized and the database has been locked, and now –it’s time to generate your study archive. How do you make sense of all that data visually? The last thing you or your data reviewers want to do is struggle to make sense of overwhelming data delivered in an endless and flat-lined, row-by-row spreadsheet.

In Today’s results-driven landscape, data aesthetics can often play a back-seat role. Large XML files storing computer-strung data points are comprehensive but not entirely human intuitive. Data reviewers can easily overlook relevant data while struggling to understand the visual context in which that data was entered. Flat XML files provide the full data in a raw form, but the visual picture of the study’s conductive methods is incomplete without…. well, visuals of the CRFs. What are the current solutions around this data presentation problem?

In recent years, visual PDF copies of case report forms have been the answer, but with PDFs there is still the problem of how to manage and bookmark every single screen of the study that was once viewed online, in a web-based EDC system. The PDFs are stagnant screenshots and require scrolling through a library of them, in a tiresome fashion. The ability to scroll through and navigate among every single PDF, paralleling the method in which the data was entered is just not feasible.

Prelude Dynamics has a better solution: an offline HTML archive. Dynamic and intuitive in structure it is easy to generate, access, and use. an HTML archive means no longer becoming entangled in a web of spreadsheet columns, unmanageable libraries of PDFs, and no more costly server fees to perpetually host the study for intermittent visual access. The HTML archive displays the entire study and its layout for you with a visual presentation of all data and function captured in one place.

The beauty of an offline HTML archive lies in its dynamic use and its cost reduction; all the appropriate links, highlights and audit trail per form are preserved and navigable exactly as they were presented within the EDC upon collection, but without the cost and security concerns of keeping a server running indefinitely. An offline HTML archive captures all the same data and their hierarchies in a read-only format just as it appeared within the EDC. This makes it infinitely easier for the auditor or data reviewer to navigate and visually scrutinize data results in question. Who entered what where and why are frequent data review questions easily answered with a visual archive.

With Prelude Dynamics’ VISION™ solutions, clients can receive the offline HTML archive in addition to a fully provided XML database (a SAS-approved import format and a native format.) The HTML archive really shines in providing a visual glimpse of studies past in read-only fashion, without altering the data format and results which were initially entered. HTML pages remain locked so they are not editable or alterable when viewing through any browser.

As clients grow the scope of their studies to get the most out of EDC, Prelude Dynamics provides greater solutions for data organization and presentation, in line with the expanding data submission requirements from regulatory agencies world-wide. Regulatory groups are putting increasing emphasis on the review process, and we believe that an offline HTML archive solution will surpass the PDF standard in greater visual accuracy, navigation, and preservation of data.

The future of visual study archives is offline HTML, an emerging standard that makes archives, data review and submission simpler and more fluid for all stakeholders and sponsors, and effectively expedites and refines the review of clinically studied products and their efficacy. HTML archive is an archive solution worth the investment.

Improving Measurement Accuracy with Electronic VAS Scales

Visual Analog Scales (VAS) are used because they provide greater accuracy and are more sensitive when measuring a variable that is on a continuum as compared to Likert scales. VAS scales are traditionally used to measure things such as pain and function. Researchers tend to prefer obtaining more accurate data, but have shied away from using VAS because measuring it accurately is time consuming and errors can be easily made. Measuring VAS on paper requires multiple people to measure the location of the mark, and photocopying could change the length of the line.

eVAS captures the same information as previously captured on paper. When the patient touches the screen or clicks on the line, a mark is recorded. The value of this mark is then automatically calculated by the system, eliminating the need of multiple measures and removing errors from the process. This makes eVAS more accurate and superior to Likert scales.

So what are we waiting for? The greatest concern is whether the line can be exactly 100 mm on every screen it is viewed on. Prelude Dynamics solves this issue by using one of two methods: either using a set length line that is not affected by the device’s screen size or by allowing users to calibrate their device using a 10 cm line, such as the length of a credit card. Either will ensure the length of the line.

Alternatively, does it really matter that the line is not exactly a certain length? The length was set when humans needed to be able to easily read the location of the mark on a line, so a scale of 100 mm was appropriate. With computers, the location can be calculated relative to the length of the line with accuracy to infinite decimal points on a by-pixel scale. According to an article in the International Journal of Medical Research, there is no significant difference between the two lengths they examined (100 mm and 75 mm) therefore they recommend the use of eVAS on iPads for patient clinical data collection.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4731681/

Several studies have been conducted using VISION™’s eVAS. The users report it is easy to use and improves accuracy of the results. Learn more about one such study by reviewing the VAS case study. Consider adding eVAS to your next study if it makes sense for what you are researching.

A Case for HTML Archives for FDA Submission

Currently FDA official submissions must be in XML or PDF format, with the preference being the submission of at least the XML format. The rationale for these is the need to have a document that will be readable and accessible in perpetuity. This is definitely an important consideration. However, these formats contain limitations when it comes to the review, oversight and approval process as clinical reviewers are more effective “seeing” a case as a set of forms the way the Investigator filled out those forms, while analysis staff are more effective with the XML raw information. How do you handle ancillary data, such as images and graphical information that were part of the process which cannot be imbedded in XML?

Below we will present a case for allowing HTML archives to be added as a part of or one of the allowable formats for FDA submission, fundamentally to support clinical case review. The raw XML would continue to be favored format for data analysis.

HTML, Hyper Text Markup Language, is the foundation of all computer languages used to create web-based interfaces. HTML is a specialized form of XML that when combined with images and style sheets (CSS) make a web page appear as it does. The Internet has been growing exponentially since its inception transforming business and everyday life; the Internet is here to stay and so is HTML. “Billions and billions” of HTML web pages exist in the world today. HTML can be compared to the alphabet. Like the alphabet, it provides us with basic element in which we can combine in order to communicate via web pages. HTML does for computer language what the alphabet does for our language. It is safe to say HTML will stand the test of time.

One might assume that HTML requires a web server to operate, but in fact, an HTML page is just a document that resides on a computer hard drive on a server, and can also be saved locally. All browsers support “Save As” and can save a web page for later viewing when the computer is disconnected from the Internet. All that’s needed is any brand of web-browser. As such, these HTML/XML pages are safe, secure, and as perpetual as any computer file on the hard drive.

Now that we have established that HTML is a viable solution in perpetuity, let’s examine the differences between XML, PDF and HTML archives. The main differences are listed in the table below.

[table id=archives /]

In conclusion, while the XML archive is a natural non-proprietary requirement for data submission because it does allow quick processing of the data in order to analyze the safety and efficacy of a drug, providing HTML data in addition is superior to PDF files in that it facilitates the case review within the same context in which the data was originally collected by Investigators. HTML will increase FDA reviewer’s confidence in the quality of the data and its outcomes.

Prelude Dynamics has initiated conversations with FDA CVM to discuss allowing HTML archives to be submitted. Among the discussion is also the possibility that Sponsors could chose to allow CVM to have access to the server in read-only mode where they could interact with the study data and even run reports, statistics and graphs within the system before final submission occurs in order to guide sponsors on any additional requirements or concerns they might have in order to help expedite the final review process, reduce the need for clarifications, and be quicker to market.

Contact

Jim Pedzinski – VP of Business Development
jpedzinski[at]preludedynamics.com – 512-476-5100 ext. 210