DiaverumBuilds AI Model to Offer Step-Change in Vascular Access...

DiaverumBuilds AI Model to Offer Step-Change in Vascular Access Thrombosis Prevention

Healthcare Tech Outlook | Wednesday, November 24, 2021

Summary:The vascular access model developed by Diaverum adheres to the concepts of human-centric, explainable, and responsible artificial intelligence.


Vascular Access (VA) thrombosis is a leading cause of pain and mortality in chronic hemodialysis patients, with an annual frequency of 0.11 to 0.5 incidents per patient. It contributes to the unsustainable trajectory of rising expenses and disease burden affecting national healthcare systems worldwide. Diaverum, a prominent worldwide renal care services provider, has created an artificial intelligence (AI) model to help its physicians and nurses anticipate episodes of vascular access thrombosis in hemodialyzed patients with a high degree of accuracy. Based on Diaverum’s historical patient data, simulations predicted that best-in-class clinical assessments and monitoring systems would miss 75 percent of actual cases. The VA model developed by Diaverum adheres to the concepts of human-centric, explainable, and responsible AI.

Dr. Fernando Macário, Diaverum’s Chief Medical Officer, comments, “Diaverum’s renal care model is built around safety-oriented policies and procedures, standardized holistic interventions, strong clinical governance and digitalized workflows to deliver medical outcomes that matter for our patients. With AI, physicians will have better predictive tools, nurses will come closer to the patients, processes will be more effective and patients will be better treated. Data scientists will be a new subset of healthcare professionals, members of a multidisciplinary team, that will also need to interface with patients and other healthcare professionals, to ensure our focus is always on delivering a personalized, life-enhancing renal care.”

It predicts if the patient will have a thrombotic event one week before the episode occurs using individualized input variables such as dialysis treatment data, lab test results, and demographics. The nephrologists are given the AI prognosis together with a set of explainable insights, allowing them to provide preventive individualized care to keep the patient’s vascular access alive.

The innovative VA AI model is integrated into the company’s digitalized clinical processes and integrates with the company’s digital infrastructure, which includes its proprietary renal information platform, d.CARE, and Treatment Guidance System (TGS). The company’s goal is to roll out the approach to its complete network of 452 clinics in 24 countries over the next 18 months, starting with its operations in Saudi Arabia, Portugal, and Spain, which care for around 12,000 renal patients.

According to ZoltanSzepesi, Diaverum’s Chief Transformation Officer, “Today’s announcement is a milestone in a three-year journey to convert all our know-how into a proprietary digital platform so that we can provide the most efficient, standardized, high-quality and highly scalable dialysis service in the industry. Looking ahead, Diaverum is building its own AI development factory to train, validate, deploy and monitor a series of AI models that will address important unmet clinical needs. Our ambition is to have an AI-empowered clinical workforce in all our clinics, improving outcomes that really matter to our patients, like the survival of their vascular access.”

Weekly Brief