Academic Journal
Optimizing and Benchmarking Machine Learning and Traditional Synaptic Event Detection Pipelines in Neurophysiology Experiments.
| Τίτλος: | Optimizing and Benchmarking Machine Learning and Traditional Synaptic Event Detection Pipelines in Neurophysiology Experiments. |
|---|---|
| Συγγραφείς: | Sevigny JP; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607.; Graduate Program in Neuroscience, University of Illinois at Chicago, Chicago, Illinois 60607., Schrank S; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607., Donka RM; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607., Aguilar OD; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607., Yunus NI; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607., Valchinova MR; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607., Fyke Z; Graduate Program in Neuroscience, University of Illinois at Chicago, Chicago, Illinois 60607.; Department of Biology, University of Illinois at Chicago, Chicago, Illinois 60607., Zak JD; Graduate Program in Neuroscience, University of Illinois at Chicago, Chicago, Illinois 60607.; Department of Biology, University of Illinois at Chicago, Chicago, Illinois 60607., Roitman JD; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607.; Graduate Program in Neuroscience, University of Illinois at Chicago, Chicago, Illinois 60607., Sparta DR; Department of Psychology, University of Illinois at Chicago, Chicago, Illinois 60607 dsparta@uic.edu.; Graduate Program in Neuroscience, University of Illinois at Chicago, Chicago, Illinois 60607. |
| Πηγή: | ENeuro [eNeuro] 2026 Apr 17; Vol. 13 (5). Date of Electronic Publication: 2026 Apr 17 (Print Publication: 2026). |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Society for Neuroscience Country of Publication: United States NLM ID: 101647362 Publication Model: Electronic-Print Cited Medium: Internet ISSN: 2373-2822 (Electronic) Linking ISSN: 23732822 NLM ISO Abbreviation: eNeuro Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: [Washington, DC] : Society for Neuroscience, [2014]- |
| Ιατρικοί όροι (MeSH): | Excitatory Postsynaptic Potentials*/physiology , Neurophysiology*/methods , Neurophysiology*/standards , Inhibitory Postsynaptic Potentials*/physiology , Synapses*/physiology , Benchmarking* , Machine Learning*, Animals |
| Περίληψη: | Synaptic physiology experiments are fundamental to neuroscience research. Consequently, accurate detection of synaptic currents is crucial for conducting high-quality experiments. Traditionally, detecting inhibitory and excitatory postsynaptic currents (sIPSCs/sEPSCs) relied on hand-counting individual events. Although sEPSCs and sIPSCs are clear to the trained eye, hand analysis is time and labor intensive. Recent advances in applied machine learning promise faster, superior event detectors that may improve data quality and reduce or even completely negate the need for hand curation. While many strategies for sIPSC and sEPSC detection exist, rarely have they been quantitatively compared for accuracy within an experiment. Our study aims to establish practical ground-truth event detection in a large experimental dataset through meticulous hand counting and to assess variance in detection results across different laboratories, analysis techniques, and cell types. Using thoroughly hand-counted data as our ground-truth comparison, we benchmark current popular detection methods, including a modern supervised deep learning approach. Our results suggest that current analysis strategies vary widely in their results and that a supervised machine learning approach rivals manual event counting performed by expert electrophysiologists better than other automated approaches. (Copyright © 2026 Sevigny et al.) |
| Competing Interests: | The authors declare no competing financial interests. |
| References: | J Physiol. 1952 Apr;116(4):449-72. (PMID: 14946713) Nat Rev Neurosci. 2015 Jan;16(1):5-16. (PMID: 25524119) Front Synaptic Neurosci. 2019 Jul 24;11:20. (PMID: 31396073) PLoS One. 2010 Nov 24;5(11):e15517. (PMID: 21124805) J Neurosci. 1993 Jun;13(6):2323-31. (PMID: 8099123) Biophys J. 1997 Jul;73(1):220-9. (PMID: 9199786) J Neurosci Methods. 2021 Jun 1;357:109125. (PMID: 33711356) Nat Commun. 2024 Jun 4;15(1):4745. (PMID: 38834553) J Physiol. 1952 Aug;117(4):500-44. (PMID: 12991237) Alcohol Clin Exp Res. 2012 Jul;36(7):1180-7. (PMID: 22309159) Biosens Bioelectron. 2002 Sep;17(9):773-82. (PMID: 12191925) Nature. 1976 Apr 29;260(5554):799-802. (PMID: 1083489) Nat Neurosci. 1999 Jan;2(1):44-9. (PMID: 10195179) Gigascience. 2021 Jan 22;10(1):. (PMID: 33481004) Neuron. 2008 Nov 6;60(3):469-76. (PMID: 18995821) Biophys J. 2012 Oct 3;103(7):1429-39. (PMID: 23062335) Neuropsychopharmacology. 2023 Jan;48(1):113-120. (PMID: 35810199) Proc Natl Acad Sci U S A. 2024 Feb 6;121(6):e2315804121. (PMID: 38294937) Elife. 2025 Mar 05;13:. (PMID: 40042890) Mol Brain. 2021 Sep 8;14(1):138. (PMID: 34496933) Curr Protoc. 2025 Jul;5(7):e70161. (PMID: 40607617) J Neurosci Methods. 1994 Apr;52(1):87-100. (PMID: 8090022) |
| Grant Information: | R00 DC017754 United States DC NIDCD NIH HHS; R01 AA027516 United States AA NIAAA NIH HHS |
| Contributed Indexing: | Keywords: electrophysiology; machine learning; synaptic physiology |
| Entry Date(s): | Date Created: 20260417 Date Completed: 20260715 Latest Revision: 20260813 |
| Update Code: | 20260813 |
| PubMed Central ID: | PMC13233908 |
| DOI: | 10.1523/ENEURO.0410-25.2026 |
| PMID: | 41997847 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 2373-2822 |
|---|---|
| DOI: | 10.1523/ENEURO.0410-25.2026 |