A contract research organisation in Europe approached us to assist in the interpretation and classification of clinical trial results. The company presents findings of carefully controlled clinical research for pharmaceutical corporations. The accurate representation of outcomes, both in terms of transparency and veracity, is absolutely crucial to avoid mistakes, which, of course, would impact directly on patient safety and our client’s reputation.
Clinical trials of new drugs are costly in terms of time and money.
They are operationally complex because regulatory control is, quite correctly, very strict, and the various groups of parties involved must each have their interests satisfied. Patients must be safeguarded, the pharmaceutical companies must make a profit, doctors must fully understand results, regulatory bodies must be kept informed and satisfied, so the list of important stakeholders is long and diverse.
Contract Research Organisations enrol patients into clinical trials. People who fit the criteria of the condition being treated by the drugs under examination are sourced from different locations, in different population cohorts & demographic sectors.
Along with the patients who are using the drugs under examination, there are also control groups to consider during patient outcome tracking. These results are documented via case report forms, which can be both electronic and/or printed on paper. The forms contain, amongst many outcome criteria, formally categorised descriptions of the most important aspects of the clinical trial: the adverse effects and serious adverse effects of the drugs under examination. Obviously, these important issues need to be fully understood to ensure that a drug is safe to pass the trial.
The report forms, one per patient, are usually text-based descriptions, using regular expressions of medical terminology. Depending on the size, complexity and geographical spread of the patient sample, the reports need to be collated and studied by various medical professionals. Crucially, the frequency of any (serious) adverse effects, and the conditions that tend to cause them, is a finding that needs to be the most obvious 'flag' for the report reader to recognise immediately.
Ensuring that findings are presented optimally and accurately is a time consuming, manual and therefore error-prone process.
to identify terms for the adverse effects found within the case report forms, by parsing the content to create unique dictionaries for the clinical trial. A further benefit of creating such a lexicon means that the solution can also be used in subsequent trials, as it becomes ever more populated with keywords after each case is examined. Crucially, this examination no longer needs to be performed manually, rather it’s done by the system parsing the report’s text.
to discover key topics and key adverse effects across all descriptions of the adverse outcomes that were present. Our solution then classified the major topics in question, which improved the quantification of frequency of symptoms and side-effects. We then included the creation of an API that interfaced with the client’s in house reporting system. Field report forms were automatically synchronised with the client's central server and time-stamped.
even if the wording used isn’t exactly the same, but describes very similar concepts. Consequently, a graphical user interface was developed to demonstrate the clustering and frequency of adverse effects within the reports.
Furthermore, the LSA technique uses word colocation to create associations and hence tuples (finite ordered lists) of words that appear together often, can become associated.
LSA can be used to process various types of text lists, free-form notes, emails etc, provided that the body of text contains multiple terms. This flexibility allows the AI to process various writing styles used by doctors and researchers who carry out the field work.
Processing data from many studies and creating conceptual connections between clinical trials is an exciting capability. This can help all parties to step back and see a bigger picture, perhaps also recognising hitherto undiscovered problematic or positive relationships between many different drugs.
Our client reported very positive feedback from Fountech Solutions’ answer, in terms of processing speed, transparency and presentation of results. They were happy to confirm that their ability to improve their offerings to their own clients in turn, with such insightful and unambiguous findings, was an affirmative step forward for their organisation.
The vCAIO will conduct a situational assessment to identify the needs of, and opportunities for, the organisation. They will provide a report outlining a gap analysis and the baseline from where to start defining the AI strategy going forward
Create a vision of where the organisation should be going
Set standards for strategy implementation through establishing the correct management structures and considering ethical issues
Make things happen by operating at the right level, knowing what to outsource and what to achieve in-house for optimum delivery
Build capacity in the organisation by recruiting the right talent and identifying appropriate vendors or consultants to solve key problems
Become the public face of the new technology, both internally and externally
Understanding your needs and identifying opportunity to exploit
Evaluate your data
Knowledge Gaps | Costs v Benefits | Your Competitors
Latest Industry Tech and how we can enhance it just for you
Our key recommendations and suggested next steps