Deep learning-based real-time detection of neurons in brain slices for in vitro physiology

Abstract

A common electrophysiology technique used in neuroscience is patch clamp: a method in which a glass pipette electrode facilitates single cell electrical recordings from neurons. Typically, patch clamp is done manually in which an electrophysiologist views a brain slice under a microscope, visually selects a neuron to patch, and moves the pipette into close proximity to the cell to break through and seal its membrane. While recent advances in the field of patch clamping have enabled partial automation, the task of detecting a healthy neuronal soma in acute brain tissue slices is still a critical step that is commonly done manually, often presenting challenges for novices in electrophysiology. To overcome this obstacle and progress towards full automation of patch clamp, we combined the differential interference microscopy optical technique with an object detection-based convolutional neural network (CNN) to detect healthy neurons in acute slice. Utilizing the YOLOv3 convolutional neural network architecture, we achieved a 98% reduction in training times to 18 min, compared to previously published attempts. We also compared networks trained on unaltered and enhanced images, achieving up to 77% and 72% mean average precision, respectively. This novel, deep learning-based method accomplishes automated neuronal detection in brain slice at 18 frames per second with a small data set of 1138 annotated neurons, rapid training time, and high precision. Lastly, we verified the health of the identified neurons with a patch clamp experiment where the average access resistance was 29.25 M$$\Omega$$ Ω (n = 9). The addition of this technology during live-cell imaging for patch clamp experiments can not only improve manual patch clamping by reducing the neuroscience expertise required to select healthy cells, but also help achieve full automation of patch clamping by nominating cells without human assistance.

Patch-walking: Coordinated multi-pipette patch clamp for efficiently finding synaptic connections

“…A third alternative could be that one pipette patches a deep cell and stays patched onto it while secondary pipettes continue to automatically patch other cells and search for connections. Experiments could be done to probe specific layer to layer connections, and machine learning algorithms to detect specific neurons could be introduced as well [39]. From these potential uses of patch-walking, future work of automated multi-patching could greatly enhance the different methods used to study functional synaptic connectivity and local neuronal networks.…”

A deep learning framework for automated and generalized synaptic event analysis

“…Convolutional neural networks (CNNs) are especially effective for image classification, but can also be used for one-dimensional data (Fawaz et al, 2019;Wang et al, 2016). CNNs have been successfully applied in neuroscience to segment brain regions (Iqbal et al, 2019), detect synaptic vesicles in electron microscopy images (Imbrosci et al, 2022), identify spikes in Ca 2+ imaging data (Rupprecht, Carta, et al, 2021), and localize neurons in brain slices (Yip et al, 2021), or neurons with fluorescence signals in time-series data (Denis et al, 2020;Sit á et al, 2022).…”

“…Convolutional neural networks (CNNs) are particularly effective for image classification and can also be used for one-dimensional data 22,23 . CNNs have been successfully applied in neuroscience to segment brain regions 24 , detect synaptic vesicles in electron microscopy images 25 , identify spikes in Ca 2+ imaging data 26 , and localize neurons in brain slices 27 , or neurons with fluorescence signals in time-series data 28,29 . Machine learning, especially deep learning, can thus significantly improve biological data analysis and contribute to a better understanding of neural function.…”