AI-based atomic force microscopy image analysis allows to predict electrochemical impedance spectra of defects in tethered bilayer membranes

Abstract

Atomic force microscopy (AFM) image analysis of supported bilayers, such as tethered bilayer membranes (tBLMs) can reveal the nature of the membrane damage by pore-forming proteins and predict the electrochemical impedance spectroscopy (EIS) response of such objects. However, automated analysis involving pore detection in such images is often non-trivial and can require AI-based object detection techniques. The specific object-detection algorithm we used to determine the defect coordinates in real AFM images was a convolutional neural network (CNN). Defect coordinates allow to predict the EIS response of tBLMs populated by the pore-forming toxins using finite element analysis (FEA) modeling. We tested if the accuracy of the CNN algorithm affected the EIS spectral features sensitive to defect densities and other physical parameters of tBLMs. We found that the EIS spectra can be predicted sufficiently well, however, systematic errors of characteristic spectral points were observed and need to be taken into account. Importantly, the comparison of predicted EIS curves with experimental ones allowed to estimate important physical parameters of tBLMs such as the specific resistance of submembrane reservoir. This reservoir separates phospholipid bilayer from the solid support. We found that the specific resistance of the reservoir amounts to $$10^{4.25 \pm 0.10}$$ 10 4.25 ± 0.10 $$\Omega \cdot cm$$ Ω · c m which is approximately two orders of a magnitude higher compared to the specific resistance of the buffer bathing tBLMs studied in this work. We hypothesize that such effect may be related in part due to decreased concentration of ionic carriers in the submembrane due to decreased relative dielectric permittivity in this region.

Mesophilic Argonaute-Mediated Polydisperse Droplet Biosensor for Amplification-Free, One-Pot, and Multiplexed Nucleic Acid Detection Using Deep Learning

“…51,52 (3) The well-designed deep learning algorithm, which reduces the number of experimental steps and laborintensive processing, efficiently processes image information to analyze large image samples. 53 The results processed by deep learning demonstrated high accuracy and excellent sensitivity (10 3 to 10 8 CFU/mL for Listeria monocytogenes and 10 2 to 10 8 CFU/mL for Salmonella typhimurium and Staphylococcus aureus) for both single and multiplex detection without signal amplification.…”

“…These open loop and linearized plasmids can activate secondary cleavage activity of CbAgo, , which can multiplex detect target nucleic acid sequences simultaneously. (2) The homogeneous reaction with no complicated separation process results in quick and efficient mass transfer. , (3) The well-designed deep learning algorithm, which reduces the number of experimental steps and labor-intensive processing, efficiently processes image information to analyze large image samples . The results processed by deep learning demonstrated high accuracy and excellent sensitivity (10 3 to 10 8 CFU/mL for Listeria monocytogenes and 10 2 to 10 8 CFU/mL for Salmonella typhimurium and Staphylococcus aureus ) for both single and multiplex detection without signal amplification.…”

On machine learning analysis of atomic force microscopy images for image classification, sample surface recognition

“…For example, regression algorithms can be used to predict complex sample properties using indirect measurements. 18 Examples of specific ML algorithms are listed in Fig. 1.…”

“…13 The AFM technique is fundamentally different from the other imaging methods. [14][15][16][17][18] AFM allows not only imaging a sample surface but also obtaining a large number of physical and chemical parameters of the sample surface. [19][20][21] AFM allows for attaining a very high spatial resolution.…”

Convolutional neural network prediction of the photocurrent–voltage curve directly from scanning electron microscopy images

“…We propose utilizing image data of the photoelectrode materials as an input source, as a previous report showed the successful use of atomic force microscopy image data to predict electrochemical impedance spectra. 25 In this study, scanning electron microscope (SEM) images were used to predict the whole J – V curve using a convolutional neural network (CNN) for one of the typical and well-studied photocatalytic materials, BiVO 4 . 26,27 Although it is challenging that high-dimensional feature space of the CNN model is constructed with only dozens of sample images, as is described in the literature, 8 we could optimize the prediction network function by increasing the number of data by cutting image data and further optimization of the CNN network.…”