Fine-tuning TrailMap: The utility of transfer learning to improve the performance of deep learning in axon segmentation of light-sheet microscopy images.
| العنوان: | Fine-tuning TrailMap: The utility of transfer learning to improve the performance of deep learning in axon segmentation of light-sheet microscopy images. |
|---|---|
| المؤلفون: | Oostrom M; AI & Data Analytics Division, Pacific Northwest National Laboratory, Richland, WA USA., Muniak MA; Vollum Institute, Oregon Health & Science University, Portland, OR USA., Eichler West RM; AI & Data Analytics Division, Pacific Northwest National Laboratory, Richland, WA USA., Akers S; AI & Data Analytics Division, Pacific Northwest National Laboratory, Richland, WA USA., Pande P; Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA USA., Obiri M; AI & Data Analytics Division, Pacific Northwest National Laboratory, Richland, WA USA., Wang W; Appel Alzheimer's Disease Research Institute, Feil Family Brain and Mind Research Institute, Weill Cornell Medicine, New York, NY USA., Bowyer K; Appel Alzheimer's Disease Research Institute, Feil Family Brain and Mind Research Institute, Weill Cornell Medicine, New York, NY USA., Wu Z; Appel Alzheimer's Disease Research Institute, Feil Family Brain and Mind Research Institute, Weill Cornell Medicine, New York, NY USA., Bramer LM; Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA USA., Mao T; Vollum Institute, Oregon Health & Science University, Portland, OR USA., Webb-Robertson BJ; Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA USA. |
| المصدر: | BioRxiv : the preprint server for biology [bioRxiv] 2023 Oct 23. Date of Electronic Publication: 2023 Oct 23. |
| نوع المنشور: | Preprint |
| اللغة: | English |
| بيانات الدورية: | Country of Publication: United States NLM ID: 101680187 Publication Model: Electronic Cited Medium: Internet NLM ISO Abbreviation: bioRxiv Subsets: PubMed not MEDLINE |
| مستخلص: | Light-sheet microscopy has made possible the 3D imaging of both fixed and live biological tissue, with samples as large as the entire mouse brain. However, segmentation and quantification of that data remains a time-consuming manual undertaking. Machine learning methods promise the possibility of automating this process. This study seeks to advance the performance of prior models through optimizing transfer learning. We fine-tuned the existing TrailMap model using expert-labeled data from noradrenergic axonal structures in the mouse brain. By fine-tuning the final two layers of the neural network at a lower learning rate of the TrailMap model, we demonstrate an improved recall and an occasionally improved adjusted F1-score within our test dataset over using the originally trained TrailMap model. |
| التعليقات: | Update in: PLoS One. 2024 Mar 29;19(3):e0293856. (PMID: 38551935) |
| References: | Prog Biophys Mol Biol. 2022 Jan;168:81-93. (PMID: 34216639) Nat Methods. 2023 Oct;20(10):1593-1604. (PMID: 37770711) Cell Metab. 2018 Jan 9;27(1):226-236.e3. (PMID: 29320703) Cell. 2015 Jul 30;162(3):648-61. (PMID: 26232230) Nat Methods. 2013 Jun;10(6):483-90. (PMID: 23866325) J Neurosci Methods. 2018 Aug 1;306:45-56. (PMID: 29689283) Nat Methods. 2012 Jun 28;9(7):676-82. (PMID: 22743772) Nat Methods. 2021 Feb;18(2):203-211. (PMID: 33288961) Cell. 2014 Feb 27;156(5):1096-111. (PMID: 24581503) Neuroinformatics. 2023 Apr;21(2):269-286. (PMID: 36809643) BMC Bioinformatics. 2012 Nov 27;13:316. (PMID: 23181553) Nat Neurosci. 2014 Sep;17(9):1276-85. (PMID: 25086607) Bioinformatics. 2011 Jul 1;27(13):i239-47. (PMID: 21685076) Nature. 2021 Oct;598(7879):174-181. (PMID: 34616072) PLoS One. 2012;7(6):e38011. (PMID: 22723842) J Histochem Cytochem. 2011 Feb;59(2):129-38. (PMID: 21339178) Brain Struct Funct. 2020 Mar;225(2):785-803. (PMID: 32065256) Proc Natl Acad Sci U S A. 2020 May 19;117(20):11068-11075. (PMID: 32358193) Ann N Y Acad Sci. 2011 Apr;1224:109-125. (PMID: 21251014) J Vis Exp. 2018 Jul 28;(137):. (PMID: 30102289) Neuron. 2017 Jun 21;94(6):1085-1100.e6. (PMID: 28641108) |
| معلومات مُعتمدة: | R01 NS081071 United States NS NINDS NIH HHS; R01 NS104944 United States NS NINDS NIH HHS; RF1 MH120119 United States MH NIMH NIH HHS; RF1 MH128969 United States MH NIMH NIH HHS |
| تواريخ الأحداث: | Date Created: 20231114 Latest Revision: 20240408 |
| رمز التحديث: | 20240408 |
| مُعرف محوري في PubMed: | PMC10634742 |
| DOI: | 10.1101/2023.10.23.563546 |
| PMID: | 37961439 |
| قاعدة البيانات: | MEDLINE |
| DOI: | 10.1101/2023.10.23.563546 |
|---|