Iterative cell extraction and registration for analysis of time-lapse neural calcium imaging datasets
Abstract/Contents
- Abstract
- Neural calcium imaging provides experimental recordings of the activity patterns of large ensembles of individual neurons in awake, behaving animals. These recordings often span multiple sessions over extended time periods of days to weeks, with the imaging field-of-view kept approximately fixed. In such longitudinal studies, to address biological questions based on the calcium imaging data, one commonly needs to 1) computationally extract individual neurons from the datasets, and 2) track the same individual neurons across multiple recording sessions. Although the extraction of neural activity has been extensively studied for individual imaging sessions, the task of matching cells across multiple sessions has not been well investigated beyond post hoc cell alignment methods based on simple heuristics. To track cells' identities over time, we propose a fast and reliable cell-matching algorithm based on the Wasserstein distance between candidate cell pairs, and we demonstrate the superiority of this metric over other similarity metrics commonly used for cell tracking. We also introduce a new framework, an iterative approach to cell extraction and cell registration, which uses the full set of cells detected across all imaging sessions to improve both the set of the extracted cells and the accuracy with which these cells can be tracked over multiple sessions. Unlike standard analysis pipelines that perform cell matching after cell extraction, our iterative approach uses the set of all detected cells to re-initialize the next iterations of cell extraction. Using both simulated and real datasets, we show that our approach can enable substantial improvements in both the accuracy and numbers of extracted and registered cells. We also show that these improvements in cell quantity and quality increase the accuracy of biological findings in subsequent analyses
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2020; ©2020 |
| Publication date | 2020; 2020 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Tasci, Tugce |
|---|---|
| Degree supervisor | Schnitzer, Mark Jacob, 1970- |
| Thesis advisor | Schnitzer, Mark Jacob, 1970- |
| Thesis advisor | El Gamal, Abbas A |
| Thesis advisor | Özgür, Ayfer |
| Degree committee member | El Gamal, Abbas A |
| Degree committee member | Özgür, Ayfer |
| Associated with | Stanford University, Department of Electrical Engineering. |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Tugce Tasci |
|---|---|
| Note | Submitted to the Department of Electrical Engineering |
| Thesis | Thesis Ph.D. Stanford University 2020 |
| Location | electronic resource |
Access conditions
- Copyright
- © 2020 by Tugce Tasci
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...