iTREND: Intelligent Technologies for Renal Dialysis and Diagnostics

l-r iTREND Principal Investigators: Maarten Taal, Paul Stewart, Nick Selby

 The iTrend project, funded by the MStart Trust is a collaborative project between the University of Derby, Royal Derby Hospital and the University of Nottingham, and seeks to improve patient outcomes by delivering real-time control via novel sensing technologies. The project officially started on April 4th 2016, runs for 3 years, and is led by PIs Prof Paul Stewart (Endowed Research Chair in Intelligent Systems, University of Derby) and Prof Maarten Taal and Dr Nick Selby (Department of Renal Medicine, Royal Derby Hospital; Faculty of Medicine & Health Sciences, University of Nottingham).

In people with acute or chronic kidney failure, waste products and excess water from metabolism must be removed by dialysis to compensate for loss of kidney function and sustain life. The kidneys maintain an equilibrium level of water and solutes inside the body, function as part of the endocrine system, and excrete acidic metabolic end-products. Dialysis performs these functions through diffusion and ultrafiltration but cannot correct the compromised endocrine functions. Positive long-term outcomes for patients are significantly compromised by blood pressure instability during dialysis treatments, which can impair blood supply (hypoperfusion) of vulnerable organs such as the heart and the brain, which over time has a degrading effect on long term organ function. This may impact not just patient outcomes but also quality of life and symptom burden. The prime objective of this project is to develop computational and technology hardware combined with best-practice solutions applied to the dialysis procedure to make significant improvements to patient outcomes and quality of life. The project integrates new data collection methods, sensors and intelligent analysis and control to deliver a hardware/software expert system to sit alongside existing dialysis equipment to perform closed loop control to achieve positive outcomes through improved stability during dialysis.

A number of technology interventions exist to improve blood pressure stability in patients by modifying dialysis delivery, including;

  • HemoControl – blood volume and water volume control
  • DiaControl – Solute and water removal
  • Thermal Balance – Cooling dialysate temperature

Such approaches which have been shown to reduce but not prevent blood pressure instability and end-organ hypoperfusion during dialysis. However significant challenges exist to make the paradigm shift necessary to achieve a generalized step-change in outcomes, namely:

  • Cardiovascular monitoring (blood pressure and heart rate) during dialysis is currently ad-hoc and infrequent, resulting in delayed detection of instability and failure to intervene early to prevent instability
  • Due to the ad-hoc nature of data collection and archiving, closed-loop control of stability has not been developed to achieve its true potential.
  • Sensor integration and validation is not well understood or developed
  • There is a huge gap in systems and control understanding and implementation when comparing healthcare with engineering (e.g. aerospace) particularly with respect to intelligent systems, advanced control, advanced sensors, remote prognostics and diagnostics, multisensory data fusion etc.
  • There is a distinct technology gap in terms of a sensor and data analysis unit which can provide decision support to practitioners in the support of dialysis patients both on an individual patient and dialysis population basis.

The proposed project is based on an unprecedented collaboration in this field between healthcare practitioners and engineering. By adopting a multidisciplinary approach to the challenge of dialysis outcomes, we believe we can achieve the necessary paradigm shift that will deliver tangible positive results. We have identified monitoring equipment (Finometer) that will provide beat to beat analysis of heart rate, blood pressure and cardiac output via a simple finger probe. Data from monitoring of a large number of dialysis treatment sessions will be analysed and integrated with other clinical data to develop a computer algorithm to predict and detect early evidence of instability.

The predictive algorithm will then be used to initiate and coordinate responses from the different dialysis machine-based interventions currently available to prevent instability. Different interventions will be tested and the most effective combination determined. The final system will tested in patients under different conditions to ensure that it is robust and able to improve stability for a wide range of patients.