A Hierarchical Convolutional Neural Network for vesicle fusion event classification

CMOS Perceptron for Vesicle Fusion Classification

Edge computing (processing data close to its source) is one of the fastest developing areas of modern electronics and hardware information technology. This paper presents the implementation process of an analog CMOS preprocessor for use in a distributed environment for processing medical data close to the source. The task of the circuit is to analyze signals of vesicle fusion, which is the basis of life processes in multicellular organisms. The functionality of the preprocessor is based on a classifier of full and partial fusions. The preprocessor is dedicated to operate in amperometric systems, and the analyzed signals are data from carbon nanotube electrodes. The accuracy of the classifier is at the level of 93.67%. The implementation was performed in the 65 nm CMOS technology with a 0.3 V power supply. The circuit operates in the weak-inversion mode and is dedicated to be powered by thermal cells of the human energy harvesting class. The maximum power consumption of the circuit equals 416 nW, which makes it possible to use it as an implantable chip. The results can be used, among others, in the diagnosis of precancerous conditions.

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

This work addresses monitoring vesicle fusions occurring during the exocytosis process, which is the main way of intercellular communication. Certain vesicle behaviors may also indicate certain precancerous conditions in cells. For this purpose we designed a system able to detect two main types of exocytosis: a full fusion and a kiss-and-run fusion, based on data from multiple amperometric sensors at once. It uses many instances of small perceptron neural networks in a massively parallel manner and runs on Jetson TX2 platform, which uses a GPU for parallel processing. Based on performed benchmarking, approximately 140,000 sensors can be processed in real time within the sensor sampling period equal to 10 ms and an accuracy of 99$$\%$$ % . The work includes an analysis of the system performance with varying neural network sizes, input data sizes, and sampling periods of fusion signals.