Deep Learning Application in Spinal Implant Identification
Retrospective observational study. To demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning. Deep learning has recently been actively applied to medical images. However, despite many attempts to apply deep learning to medica...
Saved in:
| Published in: | Spine (Philadelphia, Pa. 1976) Vol. 46; no. 5; p. E318 |
|---|---|
| Main Authors: | , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
United States
01.03.2021
|
| Subjects: | |
| ISSN: | 1528-1159, 1528-1159 |
| Online Access: | Get more information |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Retrospective observational study.
To demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.
Deep learning has recently been actively applied to medical images. However, despite many attempts to apply deep learning to medical images, the application has rarely been successful. We aimed to demonstrate the effectiveness and usefulness of deep learning in the medical field. The goal of this study was to demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.
For deep learning algorithm development, radiographs were retrospectively obtained from clinical cases in which the patients had lumbar spine one-segment instrument surgery. A total of 2894 lumbar spine anteroposterior (AP: 1446 cases) and lateral (1448 cases) radiographs were collected. Labeling work was conducted for five different implants. We conducted experiments using three deep learning algorithms. The traditional deep neural network model built by coding the transfer learning algorithm, Google AutoML, and Apple Create ML. Recall (sensitivity) and precision (specificity) were measured after training.
Overall, each model performed well in identifying each pedicle screw implant. In conventional transfer learning, AP radiography showed 97.0% precision and 96.7% recall. Lateral radiography showed 98.7% precision and 98.2% recall. In Google AutoML, AP radiography showed 91.4% precision and 87.4% recall; lateral radiography showed 97.9% precision and 98.4% recall. In Apple Create ML, AP radiography showed 76.0% precision and 73.0% recall; lateral radiography showed 89.0% precision and 87.0% recall. In all deep learning algorithms, precision and recall were higher in lateral than in AP radiography.
The deep learning application is effective for spinal implant identification. This demonstrates that clinicians can use ML-based deep learning applications to improve clinical practice and patient care.Level of Evidence: 3. |
|---|---|
| AbstractList | Retrospective observational study.
To demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.
Deep learning has recently been actively applied to medical images. However, despite many attempts to apply deep learning to medical images, the application has rarely been successful. We aimed to demonstrate the effectiveness and usefulness of deep learning in the medical field. The goal of this study was to demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.
For deep learning algorithm development, radiographs were retrospectively obtained from clinical cases in which the patients had lumbar spine one-segment instrument surgery. A total of 2894 lumbar spine anteroposterior (AP: 1446 cases) and lateral (1448 cases) radiographs were collected. Labeling work was conducted for five different implants. We conducted experiments using three deep learning algorithms. The traditional deep neural network model built by coding the transfer learning algorithm, Google AutoML, and Apple Create ML. Recall (sensitivity) and precision (specificity) were measured after training.
Overall, each model performed well in identifying each pedicle screw implant. In conventional transfer learning, AP radiography showed 97.0% precision and 96.7% recall. Lateral radiography showed 98.7% precision and 98.2% recall. In Google AutoML, AP radiography showed 91.4% precision and 87.4% recall; lateral radiography showed 97.9% precision and 98.4% recall. In Apple Create ML, AP radiography showed 76.0% precision and 73.0% recall; lateral radiography showed 89.0% precision and 87.0% recall. In all deep learning algorithms, precision and recall were higher in lateral than in AP radiography.
The deep learning application is effective for spinal implant identification. This demonstrates that clinicians can use ML-based deep learning applications to improve clinical practice and patient care.Level of Evidence: 3. Retrospective observational study.STUDY DESIGNRetrospective observational study.To demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.OBJECTIVETo demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.Deep learning has recently been actively applied to medical images. However, despite many attempts to apply deep learning to medical images, the application has rarely been successful. We aimed to demonstrate the effectiveness and usefulness of deep learning in the medical field. The goal of this study was to demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.SUMMARY OF BACKGROUND DATADeep learning has recently been actively applied to medical images. However, despite many attempts to apply deep learning to medical images, the application has rarely been successful. We aimed to demonstrate the effectiveness and usefulness of deep learning in the medical field. The goal of this study was to demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning.For deep learning algorithm development, radiographs were retrospectively obtained from clinical cases in which the patients had lumbar spine one-segment instrument surgery. A total of 2894 lumbar spine anteroposterior (AP: 1446 cases) and lateral (1448 cases) radiographs were collected. Labeling work was conducted for five different implants. We conducted experiments using three deep learning algorithms. The traditional deep neural network model built by coding the transfer learning algorithm, Google AutoML, and Apple Create ML. Recall (sensitivity) and precision (specificity) were measured after training.METHODSFor deep learning algorithm development, radiographs were retrospectively obtained from clinical cases in which the patients had lumbar spine one-segment instrument surgery. A total of 2894 lumbar spine anteroposterior (AP: 1446 cases) and lateral (1448 cases) radiographs were collected. Labeling work was conducted for five different implants. We conducted experiments using three deep learning algorithms. The traditional deep neural network model built by coding the transfer learning algorithm, Google AutoML, and Apple Create ML. Recall (sensitivity) and precision (specificity) were measured after training.Overall, each model performed well in identifying each pedicle screw implant. In conventional transfer learning, AP radiography showed 97.0% precision and 96.7% recall. Lateral radiography showed 98.7% precision and 98.2% recall. In Google AutoML, AP radiography showed 91.4% precision and 87.4% recall; lateral radiography showed 97.9% precision and 98.4% recall. In Apple Create ML, AP radiography showed 76.0% precision and 73.0% recall; lateral radiography showed 89.0% precision and 87.0% recall. In all deep learning algorithms, precision and recall were higher in lateral than in AP radiography.RESULTSOverall, each model performed well in identifying each pedicle screw implant. In conventional transfer learning, AP radiography showed 97.0% precision and 96.7% recall. Lateral radiography showed 98.7% precision and 98.2% recall. In Google AutoML, AP radiography showed 91.4% precision and 87.4% recall; lateral radiography showed 97.9% precision and 98.4% recall. In Apple Create ML, AP radiography showed 76.0% precision and 73.0% recall; lateral radiography showed 89.0% precision and 87.0% recall. In all deep learning algorithms, precision and recall were higher in lateral than in AP radiography.The deep learning application is effective for spinal implant identification. This demonstrates that clinicians can use ML-based deep learning applications to improve clinical practice and patient care.Level of Evidence: 3.CONCLUSIONThe deep learning application is effective for spinal implant identification. This demonstrates that clinicians can use ML-based deep learning applications to improve clinical practice and patient care.Level of Evidence: 3. |
| Author | Yang, Hee-Seok Park, Jeong-Yoon Kim, Sungjun Kim, Kwang-Ryeol |
| Author_xml | – sequence: 1 givenname: Hee-Seok surname: Yang fullname: Yang, Hee-Seok organization: Department of Neurosurgery, Seoul Barunsesang Hospital, Seoul, South Korea – sequence: 2 givenname: Kwang-Ryeol surname: Kim fullname: Kim, Kwang-Ryeol organization: Department of Neurosurgery, International St. Mary's Hospital, Catholic Kwandong University, Incheon, South Korea – sequence: 3 givenname: Sungjun surname: Kim fullname: Kim, Sungjun organization: Department of Radiology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea – sequence: 4 givenname: Jeong-Yoon surname: Park fullname: Park, Jeong-Yoon organization: Department of Neurosurgery, Spine and Spinal Cord Institute, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/33534442$$D View this record in MEDLINE/PubMed |
| BookMark | eNpNj01LxDAQhoOsuB_6D0R69NI1k6Qx9bauX4WC4Oq5pOlUIm0am_bgv7dghZ3DO-_hYZhnTRauc0jIJdAt0PT25v7tsKVHw5UQJ2QFCVMxQJIujvqSrEP4miDJIT0jS84TLoRgK3L3gOijHHXvrPuMdt431ujBdi6yLjp463QTZa1vtBuirEI32HoGzslprZuAF_PekI-nx_f9S5y_Pmf7XR4bnio2JUgoU0UZ1BJrlkhRUoqirDiCFLwqMWEUTIoSjUExeWhleG1YXRomkW3I9d9d33ffI4ahaG0w2EwvYTeGggklIRGg-IRezehYtlgVvret7n-Kf1_2C8kpWRA |
| CitedBy_id | crossref_primary_10_1007_s10278_023_00932_8 crossref_primary_10_1007_s00256_021_03879_5 crossref_primary_10_3389_fpsyg_2021_751234 crossref_primary_10_1002_jsp2_70001 crossref_primary_10_3389_fonc_2022_814667 crossref_primary_10_1007_s00256_021_03884_8 crossref_primary_10_3390_diagnostics12040836 crossref_primary_10_3171_2022_1_FOCUS21652 crossref_primary_10_1186_s12893_024_02646_2 crossref_primary_10_1007_s00586_025_09167_3 crossref_primary_10_1016_j_jocn_2024_110853 |
| ContentType | Journal Article |
| Copyright | Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved. |
| Copyright_xml | – notice: Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved. |
| DBID | CGR CUY CVF ECM EIF NPM 7X8 |
| DOI | 10.1097/BRS.0000000000003844 |
| 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 | Medicine |
| EISSN | 1528-1159 |
| ExternalDocumentID | 33534442 |
| Genre | Multicenter Study Journal Article Observational Study |
| GroupedDBID | --- .-D .XZ .Z2 01R 0R~ 123 1J1 354 40H 4Q1 4Q2 4Q3 53G 5RE 5VS 6PF 71W 77Y 7O~ A9M AAAAV AAAXR AAGIX AAHPQ AAIQE AAJCS AAMOA AAMTA AAQKA AAQQT AARTV AASCR AASOK AAUEB AAWTL AAXQO ABASU ABBUW ABDIG ABJNI ABOCM ABVCZ ABXVJ ABZAD ACDDN ACEWG ACGFO ACGFS ACIJW ACILI ACLDA ACWDW ACWRI ACXJB ACXNZ ADGGA ADHPY AE3 AE6 AEBDS AENEX AFDTB AFEXH AFSOK AFUWQ AGINI AHOMT AHQNM AHVBC AIJEX AINUH AJCLO AJIOK AJNWD AJZMW AKCTQ AKULP ALKUP ALMA_UNASSIGNED_HOLDINGS ALMTX AMJPA AMKUR AMNEI AOHHW AWKKM BOYCO BQLVK BYPQX C45 CGR CS3 CUY CVF DIWNM DU5 E.X EBS ECM EEVPB EIF ERAAH EX3 F2K F2L F2M F2N F5P FCALG FL- GNXGY GQDEL H0~ HLJTE HZ~ IKREB IKYAY IN~ IPNFZ JF9 JG8 JK3 JK8 K8S KD2 KMI L-C L7B N9A NPM N~7 N~B O9- OAG OAH ODA ODMTH OHH OHYEH OJAPA OL1 OLG OLH OLU OLV OLW OLY OLZ OPUJH OVD OVDNE OVIDH OVLEI OVOZU OWBYB OWU OWV OWW OWX OWY OWZ OXXIT P2P R2J RIG RLZ S4R S4S SJN TEORI TSPGW V2I VVN W3M WH7 WOQ WOW X3V X3W XXN XYM YCJ YFH YOC ZB8 ZFV ZY1 7X8 ABPXF ABXYN ABZZY ACDOF ACZKN ADKSD ADSXY AFBFQ AFMBP AFNMH AHQVU AOQMC |
| ID | FETCH-LOGICAL-c3982-c3161b98021f6ef2564b00e4bd3e1643dbe5201c9e6ecce4384a8c3fc2fbc26e2 |
| IEDL.DBID | 7X8 |
| ISICitedReferencesCount | 18 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=00007632-202103010-00012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1528-1159 |
| IngestDate | Sun Sep 28 09:54:46 EDT 2025 Wed Feb 19 02:29:06 EST 2025 |
| IsDoiOpenAccess | false |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 5 |
| Language | English |
| License | Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c3982-c3161b98021f6ef2564b00e4bd3e1643dbe5201c9e6ecce4384a8c3fc2fbc26e2 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 ObjectType-Undefined-3 |
| OpenAccessLink | https://journals.lww.com/spinejournal/Fulltext/2021/03010/Deep_Learning_Application_in_Spinal_Implant.12.aspx |
| PMID | 33534442 |
| PQID | 2486154183 |
| PQPubID | 23479 |
| ParticipantIDs | proquest_miscellaneous_2486154183 pubmed_primary_33534442 |
| PublicationCentury | 2000 |
| PublicationDate | 2021-Mar-01 20210301 |
| PublicationDateYYYYMMDD | 2021-03-01 |
| PublicationDate_xml | – month: 03 year: 2021 text: 2021-Mar-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States |
| PublicationTitle | Spine (Philadelphia, Pa. 1976) |
| PublicationTitleAlternate | Spine (Phila Pa 1976) |
| PublicationYear | 2021 |
| SSID | ssj0006319 |
| Score | 2.4608388 |
| Snippet | Retrospective observational study.
To demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep... Retrospective observational study.STUDY DESIGNRetrospective observational study.To demonstrate the clinical usefulness of deep learning by identifying previous... |
| SourceID | proquest pubmed |
| SourceType | Aggregation Database Index Database |
| StartPage | E318 |
| SubjectTerms | Adult Algorithms Deep Learning - trends Female Humans Internal Fixators - trends Lumbar Vertebrae - diagnostic imaging Lumbar Vertebrae - surgery Male Middle Aged Neural Networks, Computer Radiography - trends Retrospective Studies |
| Title | Deep Learning Application in Spinal Implant Identification |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/33534442 https://www.proquest.com/docview/2486154183 |
| Volume | 46 |
| WOSCitedRecordID | wos00007632-202103010-00012&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/eLvHCXMwpV1LT8MwDLaAIcSF92O8FCSu0bYm7VIuaDwmDmyaGEi7VUnqoF26woDfj9Nm2gkJiR5yaqrKduIvsf0Z4CpWRrV1GnNEm3Op0w43MjZcoyb_Z3zf0Yoy_6k7HKrJJB2FC7d5SKtc7InVRp3PrL8jb0VSkfOVZIE35Tv3XaN8dDW00FiFhiAo41O6upMlW3giqsYe5KLopER-e1E6l3Zbt8_jmrowPEJJ-TvIrJxNf_u_v7kDWwFmsl5tF7uwgsUebAxCIH0fru8RSxbIVd9YbxnGZtOCjUvfK4t55mASPKuLeV144QBe-w8vd488tFHglkQS0UiozqSKvLlL0BHGkbTWUJpcIB2WRG4wJhhgU0xInyhJFFpZ4WzkjI0SjA5hrZgVeAxMJ1a5hECbbpNGLS3-2HRyX05PH3FaN-FyIZWMzNTHHnSBs695tpRLE45q0WZlzaeRCRELKWV08ofZp7AZ-aySKgvsDBqOFimew7r9_pzOPy4q_dM4HA1-AMR5uLY |
| 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=Deep+Learning+Application+in+Spinal+Implant+Identification&rft.jtitle=Spine+%28Philadelphia%2C+Pa.+1976%29&rft.au=Yang%2C+Hee-Seok&rft.au=Kim%2C+Kwang-Ryeol&rft.au=Kim%2C+Sungjun&rft.au=Park%2C+Jeong-Yoon&rft.date=2021-03-01&rft.issn=1528-1159&rft.eissn=1528-1159&rft.volume=46&rft.issue=5&rft.spage=E318&rft_id=info:doi/10.1097%2FBRS.0000000000003844&rft.externalDBID=NO_FULL_TEXT |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1528-1159&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1528-1159&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1528-1159&client=summon |