{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T21:42:07Z","timestamp":1779918127921,"version":"3.53.1"},"reference-count":67,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2025,10,14]],"date-time":"2025-10-14T00:00:00Z","timestamp":1760400000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/501100002241","name":"Japan Science and Technology Agency","doi-asserted-by":"publisher","award":["JPMJCR20D5"],"award-info":[{"award-number":["JPMJCR20D5"]}],"id":[{"id":"10.13039\/501100002241","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001691","name":"Japan Society for the Promotion of Science","doi-asserted-by":"publisher","award":["25K21247"],"award-info":[{"award-number":["25K21247"]}],"id":[{"id":"10.13039\/501100001691","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Int J Imaging Syst Tech"],"published-print":{"date-parts":[[2025,11]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>Minimally invasive surgery can benefit significantly from automated surgical tool detection, enabling advanced analysis and assistance. However, the limited availability of annotated data in surgical settings poses a challenge for training robust deep learning models. This paper introduces a novel staged adaptive fine\u2010tuning approach consisting of two steps: a linear probing stage to condition additional classification layers on a pre\u2010trained CNN\u2010based architecture and a gradual freezing stage to dynamically reduce the fine\u2010tunable layers, aiming to regulate adaptation to the surgical domain. This strategy reduces network complexity and improves efficiency, requiring only a single training loop and eliminating the need for multiple iterations. We validated our method on the Cholec80 dataset, employing CNN architectures (ResNet\u201050 and DenseNet\u2010121) pre\u2010trained on ImageNet for detecting surgical tools in cholecystectomy endoscopic videos. Our results demonstrate that our method improves detection performance compared to existing approaches and established fine\u2010tuning techniques, achieving a mean average precision (mAP) of 96.4%. To assess its broader applicability, the generalizability of the fine\u2010tuning strategy was further confirmed on the CATARACTS dataset, a distinct domain of minimally invasive ophthalmic surgery. These findings suggest that gradual freezing fine\u2010tuning is a promising technique for improving tool presence detection in diverse surgical procedures and may have broader applications in general image classification tasks.<\/jats:p>","DOI":"10.1002\/ima.70218","type":"journal-article","created":{"date-parts":[[2025,10,14]],"date-time":"2025-10-14T14:06:29Z","timestamp":1760450789000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Adaptive Transfer Learning for Surgical Tool Presence Detection in Laparoscopic Videos Through Gradual Freezing Fine\u2010Tuning"],"prefix":"10.1002","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2076-6842","authenticated-orcid":false,"given":"Ana","family":"Davila","sequence":"first","affiliation":[{"name":"Institutes of Innovation for Future Society Nagoya University  Nagoya Aichi Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8833-2215","authenticated-orcid":false,"given":"Jacinto","family":"Colan","sequence":"additional","affiliation":[{"name":"Department of Micro\u2010Nano Mechanical Science and Engineering Nagoya University  Nagoya Aichi Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9917-098X","authenticated-orcid":false,"given":"Yasuhisa","family":"Hasegawa","sequence":"additional","affiliation":[{"name":"Institutes of Innovation for Future Society Nagoya University  Nagoya Aichi Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,10,14]]},"reference":[{"key":"e_1_2_12_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2593957"},{"key":"e_1_2_12_3_1","doi-asserted-by":"publisher","DOI":"10.1080\/13645706.2019.1584116"},{"key":"e_1_2_12_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11548-024-03101-6"},{"key":"e_1_2_12_5_1","doi-asserted-by":"publisher","DOI":"10.3390\/s23249865"},{"key":"e_1_2_12_6_1","doi-asserted-by":"crossref","unstructured":"Y.Yamada J.Colan A.Davila andY.Hasegawa \u201cTask Segmentation Based on Transition State Clustering for Surgical Robot Assistance \u201din 2023 8th International Conference on Control and Robotics Engineering (ICCRE) (2023) 260\u2013264.","DOI":"10.1109\/ICCRE57112.2023.10155581"},{"key":"e_1_2_12_7_1","doi-asserted-by":"crossref","unstructured":"A.Jin S.Yeung J.Jopling et\u00a0al. \u201cTool Detection and Operative Skill Assessment in Surgical Videos Using Region\u2010Based Convolutional Neural Networks \u201din 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) (2018) 691\u2013699.","DOI":"10.1109\/WACV.2018.00081"},{"key":"e_1_2_12_8_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00464-015-4330-7"},{"key":"e_1_2_12_9_1","doi-asserted-by":"publisher","DOI":"10.3109\/10929081003647239"},{"key":"e_1_2_12_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jss.2013.06.022"},{"key":"e_1_2_12_11_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-024-10979-w"},{"key":"e_1_2_12_12_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12880-022-00793-7"},{"key":"e_1_2_12_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.imavis.2024.105012"},{"key":"e_1_2_12_14_1","first-page":"3324","article-title":"Transfusion: Understanding Transfer Learning for Medical Imaging","volume":"32","author":"Raghu M.","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_12_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bbe.2021.11.004"},{"key":"e_1_2_12_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2020.3004555"},{"key":"e_1_2_12_17_1","doi-asserted-by":"crossref","unstructured":"A.Vardazaryan D.Mutter J.Marescaux andN.Padoy \u201cWeakly\u2010Supervised Learning for Tool Localization in Laparoscopic Videos \u201din Intravascular Imaging and Computer Assisted Stenting and Large\u2010Scale Annotation of Biomedical Data and Expert Label Synthesis (2018) 169\u2013179.","DOI":"10.1007\/978-3-030-01364-6_19"},{"key":"e_1_2_12_18_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11548-019-01958-6"},{"key":"e_1_2_12_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102433"},{"key":"e_1_2_12_20_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41551-023-01010-8"},{"key":"e_1_2_12_21_1","first-page":"1897","article-title":"Catastrophic Forgetting Meets Negative Transfer: Batch Spectral Shrinkage for Safe Transfer Learning","volume":"32","author":"Chen X.","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_12_22_1","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1611835114"},{"key":"e_1_2_12_23_1","unstructured":"H.Wang H.Xu J.Wang J.Zhou andK.Deng \u201cEfficient Surgical Tool Recognition via HMM\u2010Stabilized Deep Learning \u201d(2024) ArXiv:2404.04992."},{"key":"e_1_2_12_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2023.3279838"},{"key":"e_1_2_12_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2017.277"},{"key":"e_1_2_12_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2018.05.001"},{"key":"e_1_2_12_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2021.102801"},{"key":"e_1_2_12_28_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00464-021-08336-x"},{"key":"e_1_2_12_29_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.3046258"},{"key":"e_1_2_12_30_1","doi-asserted-by":"publisher","DOI":"10.1080\/21681163.2020.1835550"},{"key":"e_1_2_12_31_1","doi-asserted-by":"crossref","unstructured":"W.Chen J.Feng J.Lu andJ.Zhou \u201cEndo3d: Online Workflow Analysis for Endoscopic Surgeries Based on 3d cnn and lstm \u201din OR 2.0 Context\u2010Aware Operating Theaters Computer Assisted Robotic Endoscopy Clinical Image\u2010Based Procedures and Skin Image Analysis (2018) 97\u2013107.","DOI":"10.1007\/978-3-030-01201-4_12"},{"key":"e_1_2_12_32_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-20351-1_36"},{"key":"e_1_2_12_33_1","unstructured":"M.Sahu A.Mukhopadhyay A.Szengel andS.Zachow \u201cTool and Phase Recognition Using Contextual CNN Features \u201d(2016) ArXiv:1610.08854."},{"key":"e_1_2_12_34_1","doi-asserted-by":"publisher","DOI":"10.1186\/s40537-021-00509-8"},{"key":"e_1_2_12_35_1","doi-asserted-by":"crossref","unstructured":"A.Kanakatte A.Ramaswamy J.Gubbi A.Ghose andB.Purushothaman \u201cSurgical Tool Segmentation and Localization Using Spatio\u2010Temporal Deep Network \u201din 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) (2020) 1658\u20131661.","DOI":"10.1109\/EMBC44109.2020.9176676"},{"key":"e_1_2_12_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101572"},{"key":"e_1_2_12_37_1","doi-asserted-by":"publisher","DOI":"10.3390\/s23041958"},{"key":"e_1_2_12_38_1","doi-asserted-by":"publisher","DOI":"10.1002\/ima.22791"},{"key":"e_1_2_12_39_1","doi-asserted-by":"publisher","DOI":"10.3390\/app9142865"},{"key":"e_1_2_12_40_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.1127647"},{"key":"e_1_2_12_41_1","first-page":"18583","article-title":"Measuring Robustness to Natural Distribution Shifts in Image Classification","volume":"33","author":"Taori R.","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_12_42_1","doi-asserted-by":"publisher","DOI":"10.1109\/WACV45572.2020.9093635"},{"key":"e_1_2_12_43_1","unstructured":"A.Kumar A.Raghunathan R.Jones T.Ma andP.Liang \u201cFine\u2010Tuning Can Distort Pretrained Features and Underperform out\u2010of\u2010Distribution \u201d(2022) ArXiv:2202.10054."},{"key":"e_1_2_12_44_1","first-page":"3320","volume-title":"Advances in Neural Information Processing Systems","author":"Yosinski J.","year":"2014"},{"key":"e_1_2_12_45_1","unstructured":"S.MukherjeeandA. H.Awadallah \u201cDistilling BERT Into Simple Neural Networks With Unlabeled Transfer Data \u201d(2020) ArXiv:1910.01769."},{"key":"e_1_2_12_46_1","doi-asserted-by":"crossref","unstructured":"J.HowardandS.Ruder \u201cUniversal Language Model Fine\u2010Tuning for Text Classification \u201d(2018).","DOI":"10.18653\/v1\/P18-1031"},{"key":"e_1_2_12_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00494"},{"key":"e_1_2_12_48_1","unstructured":"Y.Lee A. S.Chen F.Tajwar et\u00a0al. \u201cSurgical Fine\u2010Tuning Improves Adaptation to Distribution Shifts \u201d(2023) ArXiv:2210.11466."},{"key":"e_1_2_12_49_1","unstructured":"Y.Liu S.Agarwal andS.Venkataraman \u201cAutoFreeze: Automatically Freezing Model Blocks to Accelerate Fine\u2010Tuning \u201d(2021) ArXiv:2102.01386."},{"key":"e_1_2_12_50_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i11.17155"},{"key":"e_1_2_12_51_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i3.16350"},{"key":"e_1_2_12_52_1","unstructured":"H.Li P.Chaudhari H.Yang et\u00a0al. \u201cRethinking the Hyperparameters for Fine\u2010Tuning \u201d(2020) ArXiv:2002.11770."},{"key":"e_1_2_12_53_1","first-page":"2825","volume-title":"Proceedings of the 35th International Conference on Machine Learning","author":"Li X.","year":"2018"},{"key":"e_1_2_12_54_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.3034343"},{"key":"e_1_2_12_55_1","doi-asserted-by":"publisher","DOI":"10.3390\/s23020570"},{"key":"e_1_2_12_56_1","doi-asserted-by":"crossref","unstructured":"A.Davila J.Colan andY.Hasegawa \u201cGradient\u2010Based Fine\u2010Tuning Strategy for Improved Transfer Learning on Surgical Images \u201din 2023 International Symposium on Micro\u2010NanoMechatronics and Human Science (2023) 1\u20135.","DOI":"10.1109\/MHS59931.2023.10510130"},{"key":"e_1_2_12_57_1","unstructured":"E. J.Hu Y.Shen P.Wallis et\u00a0al. \u201cLoRA: Low\u2010Rank Adaptation of Large Language Models \u201d(2021) ArXiv:2106.09685."},{"key":"e_1_2_12_58_1","unstructured":"J.Qu X.Han T.Xiao et\u00a0al. \u201cAdapting Vision\u2010Language Foundation Model for Next Generation Medical Ultrasound Image Analysis \u201d(2025)."},{"key":"e_1_2_12_59_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-023-09194-5"},{"key":"e_1_2_12_60_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2024.128378"},{"key":"e_1_2_12_61_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2025.3587524"},{"key":"e_1_2_12_62_1","doi-asserted-by":"publisher","DOI":"10.1093\/jcde\/qwae047"},{"key":"e_1_2_12_63_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3418206"},{"key":"e_1_2_12_64_1","unstructured":"H.Alhajj M.Lamard P.Henri Conze B.Cochener andG.Quellec \u201cCataracts \u201dIEEE Dataport (2021) https:\/\/doi.org\/10.21227\/ac97\u20108m18."},{"key":"e_1_2_12_65_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_12_66_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"e_1_2_12_67_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00432"},{"key":"e_1_2_12_68_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00020"}],"container-title":["International Journal of Imaging Systems and Technology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/ima.70218","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,26]],"date-time":"2025-11-26T11:16:44Z","timestamp":1764155804000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/ima.70218"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,14]]},"references-count":67,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,11]]}},"alternative-id":["10.1002\/ima.70218"],"URL":"https:\/\/doi.org\/10.1002\/ima.70218","archive":["Portico"],"relation":{},"ISSN":["0899-9457","1098-1098"],"issn-type":[{"value":"0899-9457","type":"print"},{"value":"1098-1098","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,14]]},"assertion":[{"value":"2024-12-10","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-21","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-10-14","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70218"}}