Healthcare systems face huge problems, with growing numbers of patients, too few healthcare professionals to treat them and increasing cost of care. The Healthcare in one sector where productivity has generally not improved in the past with technical innovation. Rather quite the opposite: the use of newer, more effective, but also more expensive technologies to treat the same disease has reduced productivity in health care. Artificial intelligence brings the promise to invert the trend, offering advances ranging from more efficient diagnoses to safer treatments at relatively cheaper prices.
Digital transformation has affected all areas of society. In healthcare, computer systems are not only designed to support documentation and administrative tasks but expected to efficiently assist health professionals in complex clinical situations.
Advances in artificial intelligence (AI) present an opportunity to optimize pathways of diagnosis and prognosis, and to develop personalized strategies for treatment. For instance, analyses which capture potential risk-factors – from underlying genetics to specific environments – could aid in the development of prophylactic strategies, and more accurate diagnosis. In addition, both structural and functional imaging techniques can provide patient specific insight into current health and inform treatment.
Despite the recent surge in scientific production around AI in medicine, very few applications made their way to regulatory certification and even fewer are used consistently in clinical practice.
The field of AI in Nephrology faces several major challenges. In order to protect public safety and trust, the most important values of medicine, the healthcare system is highly regulated, a circumstance that certainly affect the uptake and costs of innovation; healthcare professionals should be knowledgeable around the efficacy and safety of AI applications, a new competence that should be nurtured through medical education and the production of high-quality evidence; ethical concerns and equity should be addressed by manufacturers, scientists and users with an authentic strive to listen and understand minorities and underprivileged groups; and finally, a deep transformation of healthcare system toward value-based healthcare would allow valuing the benefits of AI in terms of improved outcomes and efficiency gains.
At the Renal Research Institute we apply a rigorous development and clinical integration process to solve orgnizational and clinical challenge in clinical nephrology with AI based and computational matematical models. In this presentation we will conclude by providing few examples of succesful AI clinical integration.
Bullet points
• Advances in artificial intelligence (AI) present an opportunity to optimise pathways of diagnosis and prognosis, and to develop personalised strategies for treatment
• Despite the recent surge in scientific production around AI in medicine, very few applications made their way to regulatory certification and even fewer are consistently used in clinical practice.
• The field of AI in Medicine faces several major challenges: complex regulatory pathways; public concerns around ethics, accountability, efficacy and safety; lack of clinical integration of AI solutions as a main topic in medical education; misalignment of incentive structure governing healthcare
• Examples of AI integration in clinical practice.
