SpikeDeep-classifier: a deep-learning based fully automatic offline spike sorting algorithm
Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting electrophysiological recordings, each implanted electrode can record spike activity (SA) of one or more neurons along with background activity (BA). The aim...
Uloženo v:
| Vydáno v: | Journal of neural engineering Ročník 18; číslo 1 |
|---|---|
| Hlavní autoři: | , , , , , , , , , , , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
England
05.02.2021
|
| Témata: | |
| ISSN: | 1741-2552, 1741-2552 |
| On-line přístup: | Zjistit podrobnosti o přístupu |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting electrophysiological recordings, each implanted electrode can record spike activity (SA) of one or more neurons along with background activity (BA). The aim of this study is to isolate SA of each neural source. This process is called spike sorting or spike classification. Advanced spike sorting algorithms are time consuming because of the human intervention at various stages of the pipeline. Current approaches lack generalization because the values of hyperparameters are not fixed, even for multiple recording sessions of the same subject. In this study, a fully automatic spike sorting algorithm called 'SpikeDeep-Classifier' is proposed. The values of hyperparameters remain fixed for all the evaluation data.
The proposed approach is based on our previous study (SpikeDeeptector) and a novel background activity rejector (BAR), which are both supervised learning algorithms and an unsupervised learning algorithm (K-means). SpikeDeeptector and BAR are used to extract meaningful channels and remove BA from the extracted meaningful channels, respectively. The process of clustering becomes straight-forward once the BA is completely removed from the data. Then, K-means with a predefined maximum number of clusters is applied on the remaining data originating from neural sources only. Lastly, a similarity-based criterion and a threshold are used to keep distinct clusters and merge similar looking clusters. The proposed approach is called cluster accept or merge (CAOM) and it has only two hyperparameters (maximum number of clusters and similarity threshold) which are kept fixed for all the evaluation data after tuning.
We compared the results of our algorithm with ground-truth labels. The algorithm is evaluated on data of human patients and publicly available labeled non-human primates (NHPs) datasets. The average accuracy of BAR on datasets of human patients is 92.3% which is further reduced to 88.03% after (K-means + CAOM). In addition, the average accuracy of BAR on a publicly available labeled dataset of NHPs is 95.40% which reduces to 86.95% after (K-mean + CAOM). Lastly, we compared the performance of the SpikeDeep-Classifier with two human experts, where SpikeDeep-Classifier has produced comparable results.
The SpikeDeep-Classifier is evaluated on the datasets of multiple recording sessions of different species, different brain areas and different electrode types without further retraining. The results demonstrate that 'SpikeDeep-Classifier' possesses the ability to generalize well on a versatile dataset and henceforth provides a generalized and fully automated solution to offline spike sorting.
The clinical trial registration number for patients implanted with the Utah array is
For the epilepsy patients, approval from the local ethics committee at the Ruhr-University Bochum, Germany, was obtained prior to implantation. The Clinical trial registration number for the epilepsy patients implanted with microwires is
. |
|---|---|
| AbstractList | Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting electrophysiological recordings, each implanted electrode can record spike activity (SA) of one or more neurons along with background activity (BA). The aim of this study is to isolate SA of each neural source. This process is called spike sorting or spike classification. Advanced spike sorting algorithms are time consuming because of the human intervention at various stages of the pipeline. Current approaches lack generalization because the values of hyperparameters are not fixed, even for multiple recording sessions of the same subject. In this study, a fully automatic spike sorting algorithm called 'SpikeDeep-Classifier' is proposed. The values of hyperparameters remain fixed for all the evaluation data.
The proposed approach is based on our previous study (SpikeDeeptector) and a novel background activity rejector (BAR), which are both supervised learning algorithms and an unsupervised learning algorithm (K-means). SpikeDeeptector and BAR are used to extract meaningful channels and remove BA from the extracted meaningful channels, respectively. The process of clustering becomes straight-forward once the BA is completely removed from the data. Then, K-means with a predefined maximum number of clusters is applied on the remaining data originating from neural sources only. Lastly, a similarity-based criterion and a threshold are used to keep distinct clusters and merge similar looking clusters. The proposed approach is called cluster accept or merge (CAOM) and it has only two hyperparameters (maximum number of clusters and similarity threshold) which are kept fixed for all the evaluation data after tuning.
We compared the results of our algorithm with ground-truth labels. The algorithm is evaluated on data of human patients and publicly available labeled non-human primates (NHPs) datasets. The average accuracy of BAR on datasets of human patients is 92.3% which is further reduced to 88.03% after (K-means + CAOM). In addition, the average accuracy of BAR on a publicly available labeled dataset of NHPs is 95.40% which reduces to 86.95% after (K-mean + CAOM). Lastly, we compared the performance of the SpikeDeep-Classifier with two human experts, where SpikeDeep-Classifier has produced comparable results.
The SpikeDeep-Classifier is evaluated on the datasets of multiple recording sessions of different species, different brain areas and different electrode types without further retraining. The results demonstrate that 'SpikeDeep-Classifier' possesses the ability to generalize well on a versatile dataset and henceforth provides a generalized and fully automated solution to offline spike sorting.
The clinical trial registration number for patients implanted with the Utah array is
For the epilepsy patients, approval from the local ethics committee at the Ruhr-University Bochum, Germany, was obtained prior to implantation. The Clinical trial registration number for the epilepsy patients implanted with microwires is
. Objective.Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting electrophysiological recordings, each implanted electrode can record spike activity (SA) of one or more neurons along with background activity (BA). The aim of this study is to isolate SA of each neural source. This process is called spike sorting or spike classification. Advanced spike sorting algorithms are time consuming because of the human intervention at various stages of the pipeline. Current approaches lack generalization because the values of hyperparameters are not fixed, even for multiple recording sessions of the same subject. In this study, a fully automatic spike sorting algorithm called 'SpikeDeep-Classifier' is proposed. The values of hyperparameters remain fixed for all the evaluation data.Approach.The proposed approach is based on our previous study (SpikeDeeptector) and a novel background activity rejector (BAR), which are both supervised learning algorithms and an unsupervised learning algorithm (K-means). SpikeDeeptector and BAR are used to extract meaningful channels and remove BA from the extracted meaningful channels, respectively. The process of clustering becomes straight-forward once the BA is completely removed from the data. Then, K-means with a predefined maximum number of clusters is applied on the remaining data originating from neural sources only. Lastly, a similarity-based criterion and a threshold are used to keep distinct clusters and merge similar looking clusters. The proposed approach is called cluster accept or merge (CAOM) and it has only two hyperparameters (maximum number of clusters and similarity threshold) which are kept fixed for all the evaluation data after tuning.Main results.We compared the results of our algorithm with ground-truth labels. The algorithm is evaluated on data of human patients and publicly available labeled non-human primates (NHPs) datasets. The average accuracy of BAR on datasets of human patients is 92.3% which is further reduced to 88.03% after (K-means + CAOM). In addition, the average accuracy of BAR on a publicly available labeled dataset of NHPs is 95.40% which reduces to 86.95% after (K-mean + CAOM). Lastly, we compared the performance of the SpikeDeep-Classifier with two human experts, where SpikeDeep-Classifier has produced comparable results.Significance.The SpikeDeep-Classifier is evaluated on the datasets of multiple recording sessions of different species, different brain areas and different electrode types without further retraining. The results demonstrate that 'SpikeDeep-Classifier' possesses the ability to generalize well on a versatile dataset and henceforth provides a generalized and fully automated solution to offline spike sorting.Clinical trial registration numberThe clinical trial registration number for patients implanted with the Utah array isNCT 01849822.For the epilepsy patients, approval from the local ethics committee at the Ruhr-University Bochum, Germany, was obtained prior to implantation. The Clinical trial registration number for the epilepsy patients implanted with microwires is16-5670.Objective.Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting electrophysiological recordings, each implanted electrode can record spike activity (SA) of one or more neurons along with background activity (BA). The aim of this study is to isolate SA of each neural source. This process is called spike sorting or spike classification. Advanced spike sorting algorithms are time consuming because of the human intervention at various stages of the pipeline. Current approaches lack generalization because the values of hyperparameters are not fixed, even for multiple recording sessions of the same subject. In this study, a fully automatic spike sorting algorithm called 'SpikeDeep-Classifier' is proposed. The values of hyperparameters remain fixed for all the evaluation data.Approach.The proposed approach is based on our previous study (SpikeDeeptector) and a novel background activity rejector (BAR), which are both supervised learning algorithms and an unsupervised learning algorithm (K-means). SpikeDeeptector and BAR are used to extract meaningful channels and remove BA from the extracted meaningful channels, respectively. The process of clustering becomes straight-forward once the BA is completely removed from the data. Then, K-means with a predefined maximum number of clusters is applied on the remaining data originating from neural sources only. Lastly, a similarity-based criterion and a threshold are used to keep distinct clusters and merge similar looking clusters. The proposed approach is called cluster accept or merge (CAOM) and it has only two hyperparameters (maximum number of clusters and similarity threshold) which are kept fixed for all the evaluation data after tuning.Main results.We compared the results of our algorithm with ground-truth labels. The algorithm is evaluated on data of human patients and publicly available labeled non-human primates (NHPs) datasets. The average accuracy of BAR on datasets of human patients is 92.3% which is further reduced to 88.03% after (K-means + CAOM). In addition, the average accuracy of BAR on a publicly available labeled dataset of NHPs is 95.40% which reduces to 86.95% after (K-mean + CAOM). Lastly, we compared the performance of the SpikeDeep-Classifier with two human experts, where SpikeDeep-Classifier has produced comparable results.Significance.The SpikeDeep-Classifier is evaluated on the datasets of multiple recording sessions of different species, different brain areas and different electrode types without further retraining. The results demonstrate that 'SpikeDeep-Classifier' possesses the ability to generalize well on a versatile dataset and henceforth provides a generalized and fully automated solution to offline spike sorting.Clinical trial registration numberThe clinical trial registration number for patients implanted with the Utah array isNCT 01849822.For the epilepsy patients, approval from the local ethics committee at the Ruhr-University Bochum, Germany, was obtained prior to implantation. The Clinical trial registration number for the epilepsy patients implanted with microwires is16-5670. |
| Author | Lienkämper, Robin Ali, Omair Klaes, Christian Metzler, Marita Parpaley, Yaroslav Andersen, Richard Kellis, Spencer Lee, Brian Liu, Charles Wellmer, Jörg Iossifidis, Ioannis Dyck, Susanne Glasmachers, Tobias Saif-Ur-Rehman, Muhammad |
| Author_xml | – sequence: 1 givenname: Muhammad orcidid: 0000-0003-1774-7330 surname: Saif-Ur-Rehman fullname: Saif-Ur-Rehman, Muhammad organization: Institute of Informatics, University of Applied Sciences, Bottrop, Germany – sequence: 2 givenname: Omair surname: Ali fullname: Ali, Omair organization: Department of Neurosurgery, University Hospital, Knapschaftskrankenhaus Bochum GmbH, Ruhr-University Bochum, Bochum, Germany – sequence: 3 givenname: Susanne surname: Dyck fullname: Dyck, Susanne organization: Department of Neurosurgery, University Hospital, Knapschaftskrankenhaus Bochum GmbH, Ruhr-University Bochum, Bochum, Germany – sequence: 4 givenname: Robin surname: Lienkämper fullname: Lienkämper, Robin organization: Department of Neurosurgery, University Hospital, Knapschaftskrankenhaus Bochum GmbH, Ruhr-University Bochum, Bochum, Germany – sequence: 5 givenname: Marita surname: Metzler fullname: Metzler, Marita organization: Department of Neurosurgery, University Hospital, Knapschaftskrankenhaus Bochum GmbH, Ruhr-University Bochum, Bochum, Germany – sequence: 6 givenname: Yaroslav surname: Parpaley fullname: Parpaley, Yaroslav organization: Department of Neurosurgery, University Hospital, Knapschaftskrankenhaus Bochum GmbH, Ruhr-University Bochum, Bochum, Germany – sequence: 7 givenname: Jörg surname: Wellmer fullname: Wellmer, Jörg organization: Department of Neurosurgery, University Hospital, Knapschaftskrankenhaus Bochum GmbH, Ruhr-University Bochum, Bochum, Germany – sequence: 8 givenname: Charles surname: Liu fullname: Liu, Charles organization: Neurorestoration Center and Department of Neurosurgery and Neurology, University of Southern California, Los Angeles, United States of America – sequence: 9 givenname: Brian surname: Lee fullname: Lee, Brian organization: Neurorestoration Center and Department of Neurosurgery and Neurology, University of Southern California, Los Angeles, United States of America – sequence: 10 givenname: Spencer orcidid: 0000-0002-5158-1058 surname: Kellis fullname: Kellis, Spencer organization: Division of Biology and Biomedical Engineering, CALTECH, Pasadena, United States of America – sequence: 11 givenname: Richard surname: Andersen fullname: Andersen, Richard organization: Division of Biology and Biomedical Engineering, CALTECH, Pasadena, United States of America – sequence: 12 givenname: Ioannis surname: Iossifidis fullname: Iossifidis, Ioannis organization: Institute of Informatics, University of Applied Sciences, Bottrop, Germany – sequence: 13 givenname: Tobias surname: Glasmachers fullname: Glasmachers, Tobias organization: Institute of Neuroinformatic, Ruhr-University Bochum, Bochum, Germany – sequence: 14 givenname: Christian surname: Klaes fullname: Klaes, Christian organization: Department of Neurosurgery, University Hospital, Knapschaftskrankenhaus Bochum GmbH, Ruhr-University Bochum, Bochum, Germany |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/33166944$$D View this record in MEDLINE/PubMed |
| BookMark | eNpNkLlPwzAYxS1URA_YmZBHllBfiWM2xC1VYqAbQ_TF_lIMzkGcDP3vaUVBTO_QT294czJp2gYJOefsirM8X3KteCLSVCyhtLlTR2T2V03--SmZx_jBmOTasBMylZJnmVFqRt5eO_-Jd4hdYgPE6CuP_TUF6vZVQOgb32xoCREdrcYQthTGoa1h8Ja2VRV8gzTuN2hs-2HPQti0vR_e61NyXEGIeHbQBVk_3K9vn5LVy-Pz7c0qsVLxIbGcWQ6ZQQOlZCkDk3GnnQMts0o4iTo3WjqdpUrkape4rIwQIrdYphLFglz-zHZ9-zViHIraR4shQIPtGAuhUiNTobTcoRcHdCxrdEXX-xr6bfH7h_gGn1Zkvw |
| CitedBy_id | crossref_primary_10_1088_1741_2552_ad8031 crossref_primary_10_1016_j_neucom_2025_131370 crossref_primary_10_1088_2516_1091_ac6b96 crossref_primary_10_3390_brainsci10110835 crossref_primary_10_1088_1741_2552_ad8b6c crossref_primary_10_1088_1741_2552_acfb3a crossref_primary_10_1088_1741_2552_ac0f4a crossref_primary_10_3389_fninf_2022_851024 crossref_primary_10_1109_ACCESS_2023_3242643 crossref_primary_10_1088_1741_2552_adc5a4 crossref_primary_10_1038_s41598_022_07992_w crossref_primary_10_1016_j_compbiomed_2023_107649 |
| ContentType | Journal Article |
| Copyright | Creative Commons Attribution license. |
| Copyright_xml | – notice: Creative Commons Attribution license. |
| DBID | CGR CUY CVF ECM EIF NPM 7X8 |
| DOI | 10.1088/1741-2552/abc8d4 |
| DatabaseName | Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic |
| DatabaseTitle | MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic |
| DatabaseTitleList | MEDLINE MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | no_fulltext_linktorsrc |
| Discipline | Anatomy & Physiology |
| EISSN | 1741-2552 |
| ExternalDocumentID | 33166944 |
| Genre | Research Support, Non-U.S. Gov't Journal Article |
| GroupedDBID | --- 1JI 4.4 53G 5B3 5GY 5VS 5ZH 7.M 7.Q AAGCD AAJIO AAJKP AATNI ABHWH ABJNI ABQJV ABVAM ACAFW ACGFS ACHIP ADEQX AEFHF AENEX AFYNE AKPSB ALMA_UNASSIGNED_HOLDINGS AOAED ASPBG ATQHT AVWKF AZFZN CEBXE CGR CJUJL CRLBU CS3 CUY CVF DU5 EBS ECM EDWGO EIF EMSAF EPQRW EQZZN F5P IHE IJHAN IOP IZVLO KOT LAP M45 N5L N9A NPM P2P PJBAE RIN RO9 ROL RPA SY9 W28 XPP 7X8 AEINN |
| ID | FETCH-LOGICAL-c341t-c10c1a69e9ab3050a961d7dda736f2d3e78973d7654284e7813f92228ceb53e2 |
| IEDL.DBID | 7X8 |
| ISICitedReferencesCount | 25 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000620974700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1741-2552 |
| IngestDate | Fri Sep 05 13:32:13 EDT 2025 Mon Jul 21 06:05:44 EDT 2025 |
| IsDoiOpenAccess | false |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 1 |
| Keywords | deep-learning tunable hyperparameters automatic spike sorting supervised learning unsupervised learning |
| Language | English |
| License | Creative Commons Attribution license. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c341t-c10c1a69e9ab3050a961d7dda736f2d3e78973d7654284e7813f92228ceb53e2 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ORCID | 0000-0003-1774-7330 0000-0002-5158-1058 |
| OpenAccessLink | https://iopscience.iop.org/article/10.1088/1741-2552/abc8d4/pdf |
| PMID | 33166944 |
| PQID | 2459352473 |
| PQPubID | 23479 |
| ParticipantIDs | proquest_miscellaneous_2459352473 pubmed_primary_33166944 |
| PublicationCentury | 2000 |
| PublicationDate | 2021-02-05 |
| PublicationDateYYYYMMDD | 2021-02-05 |
| PublicationDate_xml | – month: 02 year: 2021 text: 2021-02-05 day: 05 |
| PublicationDecade | 2020 |
| PublicationPlace | England |
| PublicationPlace_xml | – name: England |
| PublicationTitle | Journal of neural engineering |
| PublicationTitleAlternate | J Neural Eng |
| PublicationYear | 2021 |
| SSID | ssj0031790 |
| Score | 2.4509459 |
| Snippet | Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting... Objective.Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting... |
| SourceID | proquest pubmed |
| SourceType | Aggregation Database Index Database |
| SubjectTerms | Action Potentials - physiology Algorithms Animals Deep Learning Electrodes, Implanted Humans Neurons - physiology Signal Processing, Computer-Assisted |
| Title | SpikeDeep-classifier: a deep-learning based fully automatic offline spike sorting algorithm |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/33166944 https://www.proquest.com/docview/2459352473 |
| Volume | 18 |
| WOSCitedRecordID | wos000620974700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LSwMxEA5qPXjxVR_1RQTxtrTZZDcbL1LU4kFLwSIFDyWbZGvR7la3Ffrvndlu60kQvAQS2CEk8_gyMztDyIUNAqtFHAPzRvBAkQ3hxc4Jj-uAaeMizYps9-cH2W5HvZ7qlA63vEyrXOjEQlHbzKCPvO6LQAFYEJJfjz887BqF0dWyhcYqqXCAMiiYsreMInCsPjX_IZJ5AJ39MkwJglVfrtV1bCIrfgeYhaFpbf13i9tks4SYtDnniR2y4tJdUm2m8LwezeglLZI-C296lbw8jYdv7ta5sWcQRw8TMJNXVFOLS2VLiQFFW2cpuupnVE-BDtZ5pVmSIEilOdKgeYYFCQZUvw9gU5PX0R7ptu66N_de2W_BM2DLJp5hDcN0qJzSMaiBhlYhs9JaLXmY-JY7GSnJrcQeV5GAGeOJQg-ScXHAnb9P1tIsdYeEYhm4wDmGYEjEiquYcQAWCdAMMLhTI-eLE-wDO2OMQqcum-b9nzOskYP5NfTH87obfc5ZGCohjv7w9THZ8DH7BPOrgxNSSUCY3SlZN1-TYf55VvAJjO3O4zc7z8dF |
| linkProvider | ProQuest |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=SpikeDeep-classifier%3A+a+deep-learning+based+fully+automatic+offline+spike+sorting+algorithm&rft.jtitle=Journal+of+neural+engineering&rft.au=Saif-Ur-Rehman%2C+Muhammad&rft.au=Ali%2C+Omair&rft.au=Dyck%2C+Susanne&rft.au=Lienk%C3%A4mper%2C+Robin&rft.date=2021-02-05&rft.eissn=1741-2552&rft.volume=18&rft.issue=1&rft_id=info:doi/10.1088%2F1741-2552%2Fabc8d4&rft_id=info%3Apmid%2F33166944&rft_id=info%3Apmid%2F33166944&rft.externalDocID=33166944 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1741-2552&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1741-2552&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1741-2552&client=summon |