Respiration based prediction of intradialytic hypotension
| Reference number | |
| Coordinator | Lunds universitet - Institutionen för Biomedicinsk Teknik |
| Funding from Vinnova | SEK 180 000 |
| Project duration | January 2016 - July 2017 |
| Status | Completed |
| Venture | Personal mobility between societal sectors |
Purpose and goal
Intradialytic hypotension is the most common complication during hemodialysis, and occur in about 25% of all treatments. To reduce patient suffering it is desirable to predict and prevent it before the blood pressure drops. The aim of the project was to investigate whether respiratory information extracted from pressure sensors the dialysis machine can be used to predict hypotension.
Expected results and effects
In this project, a robust method for extracting respiratory information and estimation of respiratory rate from pressure sensor signals have been developed and tested. Publication of the method in scientific journal is planned. The possibilities of developing methods for predicting blood pressure drops based on extracted respiratory information have unfortunately been limited by data access.
Planned approach and implementation
During the grant period, PI spent time equivalent to 3 months full-time work at Gambro-Lundia, and thus gained valuable industry experience. In agreement with the project plan, an existing database of over 600 dialysis treatments has been analyzed. Unfortunately, only a few of these include completely annotated episodes of acute symptomatic blood pressure drops, so the possibilities for developing methods for predicting blood pressure drops have been very limited. New ideas for joint projects have come up during the work.