3D cell nuclei segmentation based on gradient flow tracking
2007

Automated Method for 3D Cell Nuclei Segmentation

publication Evidence: moderate

Author Information

Author(s): Li Gang, Liu Tianming, Tarokh Ashley, Nie Jingxin, Guo Lei, Mara Andrew, Holley Scott, Wong Stephen TC

Primary Institution: Harvard Medical School

Hypothesis

Can a novel automated method accurately segment closely juxtaposed or touching cell nuclei in 3D microscopy images?

Conclusion

The proposed algorithm is able to segment closely juxtaposed or touching cell nuclei obtained from 3D microscopy imaging with reasonable accuracy.

Supporting Evidence

  • The segmentation method achieved over 90% volume overlap compared to expert manual segmentation.
  • The average over-segmentation and under-segmentation rates were around 5% for the proposed method.
  • The method was validated using both synthesized and real biological images.

Takeaway

This study presents a new way to automatically separate touching cell nuclei in 3D images, making it easier for scientists to analyze them.

Methodology

The method involves gradient diffusion, gradient flow tracking, and local adaptive thresholding to segment cell nuclei from 3D images.

Limitations

The method may struggle with images of textured blob objects and requires optimization for larger datasets.

Digital Object Identifier (DOI)

10.1186/1471-2121-8-40

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication