A novel automatic segmentation and tracking method to measure cellular dielectrophoretic mobility from individual cell trajectories for high throughput assay

Highlights

• Automatic segmentation and tracking method to measure DEP-induced trajectories of numerous cells at the single cell level.

• Evaluation of the method with cell count capability, cell trapping efficiency, and cell trapping Occupancy.

• Measurement of dielectric mobility of each cell using the automatic method in a robust, efficient, and accurate manner.

• Verification in comparison of the measured cellular dielectric mobility with previous results and statistical consistency.

Abstract

Background and Objective

The dielectrophoresis (DEP) technique is increasingly being recognised as a potentially valuable tool for non-contact manipulation of numerous cells as well as for biological single cell analysis with non-invasive characterisation of a cell's electrical properties. Several studies have attempted to track multiple cells to characterise their cellular DEP mobility. However, they encountered difficulties in simultaneously tracking the movement of a large number of individual cells in a bright-field image sequence because of interference from the background electrode pattern. Consequently, this present study aims to develop an automatic system for imaging-based characterisation of cellular DEP mobility, which enables the simultaneous tracking of several hundred of cells inside a microfluidic device.

Methods

: The proposed method for segmentation and tracking of cells consists of two main stages: pre-processing and particle centre localisation. In the pre-processing stage, background subtraction and contrast enhancement were performed to distinguish the cell region from the background image. In the particle centre localisation stage, the unmarked cell was automatically detected via graph-cut algorithm-based K-means clustering.

Results

: Our algorithm enabled segmentation and tracking of numerous Michigan Cancer Foundation-7 (MCF-7) cell trajectories while the DEP force was oscillated between positive and negative. The cell tracking accuracy and cell count capability was at least 90% of the total number of cells with the newly developed algorithm. In addition, the cross-over frequency was measured by analysing the segmented and tracked trajectory data of the cellular movements caused by the positive and negative DEP force. The measured cross-over frequency was compared with previous results. The multi-cellular movements investigation based on the measured cross-over frequency was repeated until the viability of cells was unchanged in the same environment as in a microfluidic device. The results were statistically consistent, indicating that the developed algorithm was reliable for the investigation of DEP cellular mobility.

Conclusion

: This study developed a powerful platform to simultaneously measure the DEP-induced trajectories of numerous cells, and to investigate in a robust, efficient, and accurate manner the DEP properties at both the single cell and cell ensemble level.

Introduction

Since Pohl's first report in the mid-20th century of cellular movement induced by the interaction between a non-uniform electric field and dipoles inside a cell, the behaviour, called dielectrophoresis (DEP), has been widely employed for many diverse practical applications such as manipulation of cells, characterisation of cells, detection and identification of cells, and drug and phenotype screening inside a microfluidic device [1], [2]. For example, the cellular movement caused by the DEP force has been used for flexible trapping and manipulation of single cells, high-throughput separation and trapping of cells, and precise control of a single cell [1], [2], [3], [4], [5], [6]. DEP has also been used to characterise cellular properties such as polarity, capacitance, conductance, and electrochemical activity [[7], [8]–9]. Moreover, the detection and identification of cancer cells, mycobacterium semegmatis single cells, and human embryonic stem cells [[10], [11]–12], and the screening of anticancer and antibiotic drugs and phenotypes [[13], [14]–15] have also been accomplished by DEP movement techniques. Success in those applications requires accurate movement of cells with a flexible direction inside a microfluidic device and precise tracking of the trajectories of cells when recording cell movement image sequences. To achieve the accurate movement of cells, many researchers have modified the fabricated structure of electrodes, which generate a non-uniform field that leads to the DEP force, in order to accurately and flexibly control the non-uniform electric field inside their microfluidic devices. Coplanar electrodes [2,7,9,16], interdigitated electrode arrays [[10], [11]–12,17], bipolar electrodes [4], a quadrupole electrode [5], and even an optically induced electrode [8] have been developed. These structure modifications in microfluidic devices resulted in significant progress and many achievements in these applications.

However, the critical issue of tracking the trajectories of cells has not been seriously considered even though that issue plays a pivotal role in the successful achievement of a high-throughput DEP experiments. For example, while the cells are moving under a DEP force, only a single cell is tracked and recorded in each time-lapse image because of the constrained field of view, where the area captured is around just one cell of interest. In this approach, cellular trajectories can be easily examined and tracked from the image stack because there is only one cell in each image. Nevertheless, a tremendous number of images are needed to obtain statistically reliable trajectory data, which are used for the characterisation of cellular properties. Moreover, to generalize a certain cell type property, at least more than 100 cells should be observed. That is, the tracking and recording method of a single cell in each time-lapse image leads to an extremely large cost [2]. To avoid this cost, one approach is to record multiple cell trajectories in time-lapse images that are segmented into individual cells from each time-lapse image, and examine single cell trajectories on each segmented image using image processing software such as ImageJ and Cell Profiler and simple, customised segmentation algorithms [3–5,8,11,12,17,18]. The cost can be dramatically reduced by examining single cell trajectories from the images with only one cell. However, the following issues still remain for the examination of cellular trajectories: the bright-field microscope images used for tracking cellular trajectories always include noise, which originates from the partially low and blurred contrast boundary between the transparent and non-spherical features of most cells, and the intensity variation caused by overlapping cells and electrode patterns on a substrate, which result in distortion of cell segmentation and poor tracking accuracy [19]. Moreover, while numerous cells are moving in a solution under a DEP force, collision and detachment events between adjacent cells frequently occur. These congested events should be considered in the continuous analysis of consecutive images in order to accurately track cellular trajectories [20]. These drawbacks cause only a few cells to be tracked in each captured image and become a barrier to simultaneously tracking a large number of parallel single cell trajectories in an automatic, robust, efficient, and accurate manner, resulting in a lack of accurate segmentation and tracking algorithms of a multitude of cells moving under a DEP force in a microfluidic device [4].

In this paper, we report a novel automatic segmentation and tracking method for multiple single cell trajectories induced by a DEP force acting on the cells inside a microfluidic device. For the successful demonstration of the novel method, we developed a cell segmentation and tracking algorithm that allows for cell identification and segmentation, linking of the segmented cells between consecutive frames, and the elimination of a periodic electrode patterns. Using this algorithm, several hundred MCF-7 cellular trajectories on an interdigitated electrode array were successfully segmented and tracked from the images in which the movements of cells induced by varying between a positive DEP (pDEP) force and a negative DEP (nDEP) force in the same environment were recorded. The segmentation and tracking algorithm was also verified by a comparison between the cross-over frequency (f_co) of MCF-7 cells using the data from cellular trajectories with results reported in the previous work [21]. In addition, using the newly developed algorithm we tracked numerous cellular movements between traps, caused by varying between a pDEP and a nDEP force, until the viability of cells remained constant, resulting in a good statistical agreement between the repeated results. These results indicate that the developed algorithm is reliable for studying cellular DEP mobility using cell trajectories measured with the algorithm.