Artificial intelligence-assisted wearable porous eutectogel with high-performance NH3 enrichment and visual sensing enables non-invasive monitoring of chronic kidney disease

Chronic kidney disease (CKD) is a major public health disease that threatens human life. To address the urgent need for high-risk populations and patients to monitor CKD in a non-invasive and real-time manner, a novel wearable porous eutectogel with high-performance NH 3 enrichment and sensing is developed for the first time to rapidly and precisely detect the NH 3 concentration in the breath. Herein an eutectogel (EG) containing N-acryloyl glycinamide (NAGA), glycerol (Gly), choline chloride (ChCl) and bromocresol green (BCG) is constructed via digital light processing (DLP) 3D printing. The abundant internal hydrogen bond network in EG/BCG material shows a strong affinity towards NH 3 , leading to remarkable NH 3 capture and enrichment when exposing to the exhaled breath. On the other hand, its porous structure increases significantly the residence time and absorption rate of NH 3 . As a consequence, this new NH 3 sensing material displays fast NH 3 response time (3 s), low detection limit (0.5 ppm), wide NH 3 sensing range (0–10 ppm), high specificity and high temperature/humidity stability, leading to obvious color change within the range of NH 3 concentration in the exhaled breath of healthy individuals and CKD patients. Simply by introducing EG/BCG into a regular medical mask, wearing it for 10 min and monitoring its color change, early screening and continuous monitoring of CKD become feasible by correlating NH 3 concentration with blood urea nitrogen (BUN) levels. With the aid of artificial intelligence (deep learning) algorithms, the detection accuracy of 96.5 % has been achieved in practical clinic test.