Breath ammonia analysis for the purpose of dialysis efficiency evaluation

It is known that breath ammonia tightly associates with renal failure. One standard treatment for this condition is dialysis to help patient remove more urea and creatinine from the blood. It has been reported that there is a reduction in the ammonia concentration in expired breath of patients as dialysis proceeds. Therefore, this paper attempts to exploit the possibility of evaluation the efficiency of dialysis via measuring the concentration of breath ammonia. In the experiment, we traced the condition of a group of patients with renal failure by collecting their breath samples before and after dialysis, respectively. The breath signals were obtained by a self-designed breath analysis system. To obtain the most accurate classification result, the geometry features, which can represent the most relevant pathological features, were extracted and support vector regression (SVR) method was employed to calculate the classification result. The experimental result showed that there was a quite high accuracy to classify a unknown sample into either pre-dialyzed group or dialyzed group. This approach can provide a fast, painless, and cost-effective in situ monitor the progress of dialysis in real time.