Linked to EAHP Statements
Section 1 - Introductory Statements and Governance: Statements 1.3, 1.7
Section 2 - Selection, Procurement and Distribution: Statement 2.2
Section 4 - Clinical Pharmacy Services: Statement 4.8
Section 5 - Patient Safety and Quality Assurance: Statement 5.5
Section 6 - Education and Research: Statement 6.4
ACPE UAN: 0475-0000-21-001-L04-P. A knowledge-based activity.
We live in a quickly evolving world where boundaries are moved continuously toward new horizons. Our knowledge expands and healthcare is capable of more than our ancestors could ever dream of. Technology evolves exponentially, medical devices are smart and robots take their place in healthcare. Meanwhile, the digitisation wave floods the medical record and a lot of structured big data is becoming available. We use databases and build computer algorithms to help us improve quality and patient safety. Most of the computerised physician order entry (CPOE) systems already have clinical decision support systems (CDSS) on board to assist the prescriber. But what if we push the limits and go further?
When Artificial Intelligence (AI) comes into the picture, the future looks amazing and frightening at the same time. These new algorithms can review, interpret and even suggest solutions to complex medical problems. They can help healthcare professionals to augment safety and quality of care. Increase efficiency and free up scarce expert resources by redesigning our approach.
Artificial Intelligence can play a role in drug development, image recognition (e.g. interpreting radiographic images in order to identify patients with chronic pulmonary hypertension), screening patients for eligibility for clinical trials, predict responsiveness to therapy and estimate the likelihood of adverse events (e.g. acute kidney failure, QT-prolongation, heart attack, death). Using machine learning, the outcome and predictive value become more accurate. On top of this, the machine never suffers from fatigue or distractions.
But are algorithms always right? These algorithms are trained by datasets which reflect the variability between health care providers and this approach will undoubtedly influence the potential error rate. Do we not need a real neural network, such as our brain, to identify and treat the exceptional patient?
Will artificial neural networks replace the human healthcare professional or are they just a powerful supporting tool? How can they add to cost-effectiveness and can we trust them while we focus on new tasks? In this keynote, we are introduced to the world of artificial intelligence in relation to healthcare and learn about the endless potentials and possible pitfalls. After all, we cannot stop this evolution and the world will never be the same again.
After the keynote, the participant should be able to:
• understand the definition of artificial intelligence and neural networks;
• discuss the possibilities and limitations of machine learning;
• define the role of artificial intelligence as a tool to help the healthcare professional.
Educational need addressed
As more structured data becomes available algorithms can support the healthcare professional in decision making. At the next level, neural networks can interpret datasets and images and are capable of refining themselves in order to improve the predictive value of their outcomes. Hospital pharmacists must understand the possibilities, role and limitations of artificial intelligence. To be able to use it as an effective tool to provide high-quality and cost-effective care to their patients.
Keywords: Artificial intelligence, machine learning, algorithms, neural network, prediction, decision support, big data, pharmaceutical care, informatics, technology.