RRI, in collaboration with Fresenius Medical Care GRD, has developed advanced techniques to create realistic simulations of large clinical trials—providing valuable safety information, revealing possible treatment protocol limitations, and predicting potential patient benefits. At their core, virtual trials are comprehensive mathematical models that simulate actual patient physiology. RRI applies advanced mathematical and computational techniques to adapt essential characteristics of individual patients using real-life dialysis clinical data, producing a model referred to as an avatar. Thousands of avatars are generated and comprise the population used in virtual trial.

Recording and acting upon instantaneous bio-signals is an important way to create smoother dialysis treatments to minimize patient physiological stress. RRI researchers were the first to analyze high-frequency recordings of arterial oxygen saturation to detect sleep-associated hypoxemic episodes during dialysis. Using physiological models, advanced analytics, and bio-signals—such as intradialytic central venous oxygen saturation and hematocrit in combination with other patient-specific data—researchers were also able to follow the maturation of newly created arteriovenous fistulas and patient hemodynamics to predict intradialytic hypotensive episodes.

RRI created a mobile application that uses vascular access images to identify and classify aneurysms that pose a high risk to patients. This app is an example of how artificial intelligence can help recognize patterns in data, automatically flag potential problems, and improve personalized patient care. By quickly analyzing smartphone images taken by a caregiver, the app can classify the severity of the patient’s vascular access problem and then immediately send a message to the clinical team. This innovative tool is currently being piloted in 20 RRI clinics.

Peritonitis, a serious complication for peritoneal dialysis (PD) patients, can cause morbidity and mortality. Swift diagnosis and reduced time-to-treatment are essential for therapy. Cloudiness, resulting from increased white blood cell levels in spent PD dialysate, is a significant indicator of potential peritonitis. Often early detection of cloudy dialysate relies on patient judgment, which may be compromised due to poor visual acuity and/or cognitive challenges. RRI scientists developed a smartphone- based tool to objectively measure turbidity in spent PD fluid and automatically report the results to the care provider. This system is economical because it utilizes built-in light sensors and is possibly more sensitive than the turbidity recognized and reported by the patient.

Intradialytic hypotension (IDH), a common complication of hemodialysis, is associated with high morbidity and mortality. Despite significant effort to develop strategies to prevent IDH, there is limited data describing dynamic changes in blood pressure during hemodialysis. A model developed at RRI and based on machine learning methods can predict IDH, defined as a systolic blood pressure < 90 mm Hg during hemodialysis, with clinically meaningful accuracy. The model used a host of data, including pre-treatment data (e.g., demographics, routine dialysis-specific measurements, lab values, and comorbidities), data captured during dialysis in the Chairside system (e.g., blood pressure, heart rate), and Crit-Line® data (relative blood volume, oxygen saturation).

Arteriovenous fistula (AVF) and arteriovenous graph (AVG) complications are frequent in chronic hemodialysis patients, resulting in compromised dialysis efficiency, increased hospitalization, mortality, and higher costs. Clinical diagnosis of AVF patency is difficult, and it was hypothesized that patient clinical and laboratory parameters might be indicative of impending AVF failure. RRI researchers developed a predictive model using the temporal dynamics of those parameters, identifying patients at risk of AVF or AVG failure and allowing for timely interventions to reduce conversions from AVF or AVG to central venous catheters.

RRI is exploring smartphone-based technologies to address issues related to AVFs. There is a pressing need to develop point-of-care devices that can be used by patients at home to monitor their AVF daily without assistance from clinical staff. RRI’s previous clinical study indicated that signals derived from post-processed smartphone video recordings of AVFs correlate with their flow rates. With this technology, RRI may be able to develop a smartphone-based application utilizing built-in camera functionalities to detect early AVF failure.

RRI's novel concept for “displacer-enhanced dialysis” was named a winner of the Kidney Innovation Accelerator (KidneyX) Redesign Dialysis competition, a public-private partnership between the U.S. Department of Health and Human Services and the American Society of Nephrology. RRI’s proposal aims to develop a new displacer substance that can rid the blood of protein-bound uremic toxins that can negatively affect patient health and are notoriously difficult to remove by hemodialysis. There has been little progress in the past decade in medicine’s ability to remove these toxins during hemodialysis, and this award recognizes RRI’s research toward finding a practical solution that would help improve the quality of life for patients requiring life-sustaining hemodialysis treatment. The KidneyX competition focuses on accelerating innovation in the prevention, diagnosis, and treatment of kidney diseases. RRI was one of 15 winners for this first phase of the Redesign Dialysis competition.