TY - CPAPER AB - Neural activity results in chemical changes in theextracellular environment such as variation in pH or potassium/sodium ion concentration. Higher signal to noise ratio makeneurochemical signals an interesting biomarker for closed-loopneuromodulation systems. For such applications, it is importantto reliably classify pH signatures to control stimulationtiming and possibly dosage. For example, the activity of thesubdiaphragmatic vagus nerve (sVN) branch can be monitoredby measuring extracellular neural pH. More importantly, guthormone cholecystokinin (CCK)-specific activity on the sVN canbe used for controllably activating sVN, in order to mimic thegut-brain neural response to food intake. In this paper, we presenta convolutional neural network (CNN) based classification systemto identify CCK-specific neurochemical changes on the sVN,from non-linear background activity. Here we present a novelfeature engineering approach which enables, after training, ahigh accuracy classification of neurochemical signals using CNN. AU - Roever,P AU - Mirza,KB AU - Nikolic,K AU - Toumazou,C DO - 10.1109/ISCAS45731.2020.9180734 EP - 5 PB - IEEE PY - 2020/// SN - 0271-4302 SP - 1 TI - Convolutional neural network for classification of nerve activity based on action potential induced neurochemical signatures UR - http://dx.doi.org/10.1109/ISCAS45731.2020.9180734 UR - http://hdl.handle.net/10044/1/77583 ER -