matlab 2021a programming  (MathWorks Inc)

Journal: Cell Reports Methods

Article Title: CartoCell, a high-content pipeline for 3D image analysis, unveils cell morphology patterns in epithelia

doi: 10.1016/j.crmeth.2023.100597

Figure Lengend Snippet: CartoCell pipeline for high-content epithelial cysts segmentation Phase 1: the “high-resolution raw images” consist of confocal z-stack images, where the cell membrane is stained. These images are segmented and proofread using LimeSeg and a custom MATLAB code for curation to obtain the “high-resolution label images” . Together, the raw and the label images encompass the “training high-resolution dataset.” Number of samples = 21. Phase 2: the “training high-resolution dataset” is down-sampled to obtain the “training down-sampled dataset,” which is the training set for the “model M1.” Phase 3: low-resolution images are obtained from confocal z-stack images, stained in a similar way to phase 1. Number of samples = 293. Scale bar, 100 μm. Next, the “work-flow M1” is applied: inference using “model M1” and subsequent post-processing to obtain individual cell instance predictions and cell masks, followed by the 3D Voronoi algorithm to guarantee that predicted cells remain in close contact. As a result, the “low-resolution label images” are generated. Phase 4: training of the “model M2” on the large “training low-resolution dataset.” Number of samples = 314. Phase 5: high-content segmentation of new low-resolution images (unseen by the pipeline) using the “work-flow M2,” which is equivalent to the “work-flow M1” but using the “model M2.” See also Figures S1–S3 ; Tables S2 , , and .

Article Snippet: The output of LimeSeg was processed using an in-house MATLAB program (2021a MathWorks) to detect and curate imperfections during cysts segmentation (see section).

Techniques: Membrane, Staining, Generated

Journal: Cell Reports Methods

Article Title: CartoCell, a high-content pipeline for 3D image analysis, unveils cell morphology patterns in epithelia

doi: 10.1016/j.crmeth.2023.100597

Figure Lengend Snippet:

Article Snippet: The output of LimeSeg was processed using an in-house MATLAB program (2021a MathWorks) to detect and curate imperfections during cysts segmentation (see section).

Techniques: Recombinant, Software