DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks.

Development of automated analysis tools for "single ion channel" recording is hampered by the lack of available training data. For machine learning based tools, very large training sets are necessary with sample-by-sample point labelled data (e.g., 1 sample point every 100microsecond). In an experimental context, such data are labelled with human supervision, and whilst this is feasible for simple experimental analysis, it is infeasible to generate the enormous datasets that would be necessary for a big data approach using hand crafting. In this work we aimed to develop methods to generate simulated ion channel data that is free from assumptions and prior knowledge of noise and underlying hidden Markov models. We successfully leverage generative adversarial networks (GANs) to build an end-to-end pipeline for generating an unlimited amount of labelled training data from a small, annotated ion channel "seed" record, and this needs no prior knowledge of theoretical dynamical ion channel properties. Our method utilises 2D CNNs to maintain the synchronised temporal relationship between the raw and idealised record. We demonstrate the applicability of the method with 5 different data sources and show authenticity with t-SNE and UMAP projection comparisons between real and synthetic data. The model would be easily extendable to other time series data requiring parallel labelling, such as labelled ECG signals or raw nanopore sequencing data.

Deep-Channel uses deep neural networks to detect single-molecule events from patch-clamp data.

Single-molecule research techniques such as patch-clamp electrophysiology deliver unique biological insight by capturing the movement of individual proteins in real time, unobscured by whole-cell ensemble averaging. The critical first step in analysis is event detection, so called "idealisation", where noisy raw data are turned into discrete records of protein movement. To date there have been practical limitations in patch-clamp data idealisation; high quality idealisation is typically laborious and becomes infeasible and subjective with complex biological data containing many distinct native single-ion channel proteins gating simultaneously. Here, we show a deep learning model based on convolutional neural networks and long short-term memory architecture can automatically idealise complex single molecule activity more accurately and faster than traditional methods. There are no parameters to set; baseline, channel amplitude or numbers of channels for example. We believe this approach could revolutionise the unsupervised automatic detection of single-molecule transition events in the future.

Measurement of Core Body Temperature Using Graphene-Inked Infrared Thermopile Sensor.

Continuous and reliable measurements of core body temperature (CBT) are vital for studies on human thermoregulation. Because tympanic membrane directly reflects the temperature of the carotid artery, it is an accurate and non-invasive method to record CBT. However, commercial tympanic thermometers lack portability and continuous measurements. In this study, graphene inks were utilized to increase the accuracy of the temperature measurements from the ear by coating graphene platelets on the lens of an infrared thermopile sensor. The proposed ear-based device was designed by investigating ear canal geometry and developed with 3D printing technology using the Computer-Aided Design (CAD) Software, SolidWorks 2016. It employs an Arduino Pro Mini and a Bluetooth module. The proposed system runs with a 3.7 V, 850 mAh rechargeable lithium-polymer battery that allows long-term, continuous monitoring. Raw data are continuously and wirelessly plotted on a mobile phone app. The test was performed on 10 subjects under resting and exercising in a total period of 25 min. Achieved results were compared with the commercially available Braun Thermoscan, Original Thermopile, and Cosinuss One ear thermometers. It is also comprehended that such system will be useful in personalized medicine as wearable in-ear device with wireless connectivity.

Graphene-Enabled Electrodes for Electrocardiogram Monitoring.

The unique parameters of graphene (GN)-notably its considerable electron mobility, high surface area, and electrical conductivity-are bringing extensive attention into the wearable technologies. This work presents a novel graphene-based electrode for acquisition of electrocardiogram (ECG). The proposed electrode was fabricated by coating GN on top of a metallic layer of a Ag/AgCl electrode using a chemical vapour deposition (CVD) technique. To investigate the performance of the fabricated GN-based electrode, two types of electrodes were fabricated with different sizes to conduct the signal qualities and the skin-electrode contact impedance measurements. Performances of the GN-enabled electrodes were compared to the conventional Ag/AgCl electrodes in terms of ECG signal quality, skin-electrode contact impedance, signal-to-noise ratio (SNR), and response time. Experimental results showed the proposed GN-based electrodes produced better ECG signals, higher SNR (improved by 8%), and lower contact impedance (improved by 78%) values than conventional ECG electrodes.