Academic Journal of Computing & Information Science, 2023, 6(12); doi: 10.25236/AJCIS.2023.061205.
Qinfang Zhang
School of Information Science and Engineering, Chongqing Jiaotong University, Chongqing, China
The effectiveness of deep learning-based high-density localization techniques in expediting applications of single-molecule localization microscopy is well-acknowledged. However, existing methods within this domain grapple with speed limitations, particularly when confronted with extensive raw data sets. To address this challenge, this study introduces a high-density localization approach founded on TensorRT, incorporating an enhanced U-shaped network for the swift reconstruction of raw images. Experimental verification conducted on a dataset featuring microtubule proteins illustrates that the proposed algorithm significantly amplifies the overall pace of reconstruction without compromising the precision of image reconstruction. This not only showcases a noteworthy advancement in speed but also establishes a practical technical remedy for the swift and high-fidelity reconstruction of fluorescence single-molecule microscopy images.
microscopy; fluorescence microscopy; super-resolution; image reconstruction; TensorRT
Qinfang Zhang. Accelerated Single-Molecule Microscopy Image Reconstruction Algorithm Based on TensorRT. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 12: 45-49. https://doi.org/10.25236/AJCIS.2023.061205.
[1] Deschout H, Zanacchi F C, Mlodzianoski M, et al. Precisely and accurately localizing single emitters in fluorescence microscopy[J]. Nature methods, 2014, 11(3): 253-266.
[2] W. Ouyang, A. Aristov, M. Lelek, X. Hao, and C. Zimmer, “Deep learning massively accelerates super-resolution localization microscopy,” Nat. Biotechnol. 36(5), 460–468 (2018).
[3] S. A. Jones, S.-H. Shim, J. He, and X. Zhuang, “Fast, three-dimensional super-resolution imaging of live cells,” Nat. Methods 8(6), 499–505 (2011).
[4] Nehme E, Weiss L E, Michaeli T, et al. Deep-STORM: super-resolution single-molecule microscopy by deep learning [J]. Optica, 2018, 5(4): 458-464.
[5] Zhang Q, Lou L, Li Q, et al. Super-resolution single-molecule microscopic reconstruction based on improved deep-STORM[C]//5th International Conference on Computer Information Science and Application Technology (CISAT 2022). SPIE, 2022, 12451: 458-463.
[6] Liu Y, Wang Y, Li N, et al. An attention-based approach for single image super resolution[C]//2018 24 Th international conference on pattern recognition (ICPR). IEEE, 2018: 2777-2784.
[7] Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]//International conference on machine learning. pmlr, 2015: 448-456.
[8] Sage D, Kirshner H, Pengo T, et al. Quantitative evaluation of software packages for single-molecule localization microscopy[J]. Nature methods, 2015, 12(8): 717-724.
[9] Kingma D, Ba J .Adam: A Method for Stochastic Optimization[J].Computer Science, 2014. DOI:10. 48550/arXiv.1412.6980.
[10] Abadi M, Agarwal A, Barham P, et al. TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems[J]. 2016.DOI:10.48550/arXiv.1603.04467.
[11] Gazagnes S, Soubies E, Blanc-Féraud L. High density molecule localization for super-resolution microscopy using CEL0 based sparse approximation[C]//2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017). IEEE, 2017: 28-31.
[12] Min J, Vonesch C, Kirshner H, et al. FALCON: fast and unbiased reconstruction of high-density super-resolution microscopy data[J]. Scientific reports, 2014, 4(1): 1-9.