Unleashing the Power of Artificial Intelligence in Pharmacovigilance
Welcome to the Pharmacovigilance-AI – cutting-edge insights into the intersection of Artificial Intelligence (AI) and Pharmacovigilance. In this rapidly evolving era of healthcare, the integration of AI technologies has emerged as a game-changer, that will revolutionize the field of pharmacovigilance. At Ottosen Pharmacovigilance, we aim to provide you access to Pharmacovigilance AI capabilities and a broader understanding of how AI is reshaping the landscape of drug safety monitoring and optimizing patient care.
The Pharmacovigilance AI has all the features of the most advanced artificial intelligence models available and we are constantly training the Pharmacovigilance-ai with specific information related to global drug safety, pharmacovigilance, regulatory and clinical development.
Below you can see examples of practical use of the Pharmacovigilance AI
Write risk section of the Informed Consent Form (ICF) in clinical trials. (see more in the ICF blog post HERE)
The description of risks in a clinical trial must be written in lay language and at an 8th grade level in the ICF. This is an area where a well trained AI can really help.
A real life example is a study where we know that the trial drug can cause “changes in calcium levels” and we have to explain in the patient information (ICF) what that means in lay language. Asking the right question to a capable AI can provide an excellent text for the information sheet.
I asked the AI this specific question: “In layman’s terms what is changes in calcium levels. Short response please”
The Pharmacovigilance AI responded: “Changes in calcium levels meaning the amount of calcium in your body is either too high or too low. Calcium is important for strong bones, muscle function, and nerve signaling. If calcium levels are too high, it can cause symptoms like fatigue, nausea, and kidney stones. If calcium levels are too low, it can lead to muscle cramps, weak bones, and mood changes.”
Specification of normal lab ranges
Often when you set up a clinical trial you will need the normal rages for a large number of lab parameters and they are not always easy to get. This is another area where the AI provide excellent help.
Question to the AI: What is the normal range for potassium?
Pharmacovigilance AI response: The normal range for potassium levels in the blood is typically between 3.5 and 5.0 milliequivalents per liter (mEq/L). Potassium is an essential electrolyte that plays a crucial role in various bodily functions, including nerve and muscle cell functioning. Maintaining the appropriate balance of potassium is vital for proper heart function, nerve transmission, and muscle contraction. However, it is important to note that the reference range for potassium levels may vary slightly depending on the laboratory, so it’s always best to consult with a healthcare professional for an accurate interpretation of your specific potassium levels.
This is probably too much information for your purpose so you can either ask the AI – shorter – or ask the question differently
Question to the AI: What is the normal range for potassium – short response please
The Pharmacovigilance AI responds: The normal range for potassium levels in the blood is typically between 3.5 and 5.0 mEq/L
This on the other hand may be too brief as there are varying ranges of serum potassium depending on age group so you may want to add another modifying question where you simply add this in the entry box
Modifying question to the AI: By agegroup
New response from the AI: The normal range for potassium levels in the blood can vary slightly by age group. Here are the general ranges:
Infants (0-6 months): 3.9-5.5 mEq/L
Children (7 months – 14 years): 3.4-4.7 mEq/L
Adults: 3.5-5.0 mEq/L
There are many other examples but it is probably best you try it out yourself by entering your question below where you have full access to the Pharmacovigilance AI
Keep in mind that this is still early days for artificial intelligence and pharmacovigilance so answers may not always be correct. Always double check any response with other sources.