Possibilities in the application of machine learning on bioimpedance time-series

Abstract

The relation between a biological process and the changes in passive electrical properties of the tissue is often non-linear, in which developing prediction models based on bioimpedance spectra is not trivial. Relevant information on tissue status may also lie in characteristic developments in the bioimpedance spectra over time, often neglected by conventional methods. The aim of this study was to explore possibilities in machine learning methods for time series of bioimpedance spectra, where we used organ ischemia as an example. Based on published data on the development of the bioimpedance spectrum during liver ischemia, a simulation model was made and used to generate sets of synthetic data with different levels of organ-to-organ variation, measurement noise and drift. Three types of artificial neural networks were employed in learning to predict the ischemic duration, based on the simulated datasets. The simulated prediction performance was very dependent on the amount of training examples, the organ-to-organ variation and the selection of input variables from the bioimpedance spectrum. The performance was also affected by noise and drift in the measurement, but a recurrent neural network with long short-term memory units could obtain good predictions even on noisy and drifting measurements. This approach may be relevant for further exploration on several applications of bioimpedance having the purpose of predicting a biological state based on spectra measured over time.

Stochastic Bioimpedance-Based Channel Model of the Human Body for Galvanic Coupling

“…For example, glucose concentration affects body analyte composition [ 27 ]). [ 28 ] showed bioimpedance is related to biological state of liver ischemia. [ 29 , 30 ] showed bioimpdeance body parameters such as relative permittivity differ in the various body parts and subsections such as the dermis and epidermis skin layers.…”

Supervised binary classification methods for strawberry ripeness discrimination from bioimpedance data

“…An alternative solution to the above-mentioned issues, employed mostly in a biomedical context, is represented by the use of machine learning based classification methods. Within this framework, the most widely used and effective discrimination techniques are represented by LR 19 , artificial neural networks (ANN) 20 , DT 21 , SVM 22 , NBC 23 and KNN 24 .…”

Machine learning for stem cell differentiation and proliferation classification on electrical impedance spectroscopy

“…While this is an intelligent way of both dealing with electrode contributions and also reducing the size of the samples, we argue that machine learning methods offer a viable mechanism to automate some of the post processing methods since it can be used to perform automatic feature extraction and enabling new insights over the processes subjacent to cell differentiation by allowing increased granularity through post processing automation. Also, as CRM becomes an established field, clinical monitoring will take precedence over scientific interprettation [16], making the prediction performance more important than drawing inference from the data. The overall aim can typically be to assess cell state or levels of a) b) c) pathological changes related to situations where there is a need for improvement of the cultured cells.…”

“…This can be seen as a highly non-linear system characterized by conflicting events that simultaneously compete such as proliferation (impedance increase) or the increase in the total area of cell membrane (increasing electrical capacitance) created by the dendritic geometry competing with decreased cell density as neurons are sparser when compared to the precursor stem cells (less cells and therefore less total membrane and also big morphological changes). As with tissue [ 16 ], this leads to the possibility that the value of electrical parameters changes at different rates and overlap at different time durations of the underlying biological processes, making the ability to discern the state of the cell culture dependent on memory and the historical development of the parameters as not every cell will not undergo the same process at the same time. This represents a big challenge even for the application of machine learning as the differentiation phenomenon is non-linear both in time and also regarding the subjacent processes that compose it.…”

“…Recent studies from our group demonstrated promising results using a recurrent neural network (RNN) with long short-term memory (LSTM). Through hours long measurements, they obtained high classification accuracies using repeated measurements in tissue related studies suggesting that this type of machine learning approach may be useful in a wider sense for impedance time series problems [ 16 , 17 ].…”