0.3 V 2.5 nW per Channel Current-Mode CMOS Perceptron for Biomedical Signal Processing in Amperometry

CMOS Perceptron for Vesicle Fusion Classification

“…where I 0 is a process dependent constant, κ is the gate coupling coefficient and U T is the thermal voltage. For most CMOS technologies, these parameters take the following values: κ = 0.7, U T = 26 mV and I 0 has a value below 1 pA for NMOS transistor and below 10 fA for PMOS transistor [42]. Contrary to the strong inversion mode in the weak inversion mode, the drain current I D of the MOS transistor does not depend on the gate voltage in relation to the source potential (V GS ) and the drain voltage in relation to the source potential (V DS ), but directly on the potentials in the nodes: source (V S ), drain (V D ), gate (V G ) and in the case of of the PMOS transistor on the well potential (V W ).…”

“…To implement a neural network as a VLSI circuit, the following three circuits were used: a programmable multiplier described in paper [43] which implements the neuron weight, shown in Figure 1, a circuit described in paper [42] which implements a nonlinear neuron activation function, shown in Figure 2, and a circuit for removing the concurrent component (CMRR) described in paper [43] and shown in Figure 3. The multiplier shown in Figure 1 consists of two series-connected six-output current mirrors controlled by twelve keys.…”

“…While this implementation is generally satisfactory with respect to the fidelity of reconstructing the hyperbolic activation function, it appears to be relatively expensive. This implementation uses the circuit shown in Figure 2 designed and described by authors in paper [42]. The advantage of the circuit is its simple architecture based on only six transistors.…”

“…Based on [42], all of the neurons have a uniform activation function described by the Equation ( 6) and Figure 6.…”

Exocytotic vesicle fusion classification for early disease diagnosis using a mobile GPU microsystem

“…Furthermore, typical time waveforms of full and kiss-and-run fusions last 170-180 ms. For this reason, we assumed the duration of a single pattern equal to 180 ms. We checked that for the above-described patterns of both fusion types, the number of points obtained from the patterns making it possible to classify them using the assumed baseline accuracy equals 18. The relatively small size of the input data vector makes it possible to use a perceptron network with an architecture dedicated to the classification of time waveforms [24]. This also determines the size of the input layer equal to 18 neurons and the sampling frequency of the current signals equal to 100 Hz.…”