Needle detection and localisation for robot‐assisted subretinal injection using deep learning
Abstract Subretinal injection is a complicated task for retinal surgeons to operate manually. In this paper we demonstrate a robust framework for needle detection and localisation in robot‐assisted subretinal injection using microscope‐integrated Optical Coherence Tomography with deep learning. Five...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-06-01
|
| Series: | CAAI Transactions on Intelligence Technology |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/cit2.12242 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850121630164975616 |
|---|---|
| author | Mingchuan Zhou Xiangyu Guo Matthias Grimm Elias Lochner Zhongliang Jiang Abouzar Eslami Juan Ye Nassir Navab Alois Knoll Mohammad Ali Nasseri |
| author_facet | Mingchuan Zhou Xiangyu Guo Matthias Grimm Elias Lochner Zhongliang Jiang Abouzar Eslami Juan Ye Nassir Navab Alois Knoll Mohammad Ali Nasseri |
| author_sort | Mingchuan Zhou |
| collection | DOAJ |
| description | Abstract Subretinal injection is a complicated task for retinal surgeons to operate manually. In this paper we demonstrate a robust framework for needle detection and localisation in robot‐assisted subretinal injection using microscope‐integrated Optical Coherence Tomography with deep learning. Five convolutional neural networks with different architectures were evaluated. The main differences between the architectures are the amount of information they receive at the input layer. When evaluated on ex‐vivo pig eyes, the top performing network successfully detected all needles in the dataset and localised them with an Intersection over Union value of 0.55. The algorithm was evaluated by comparing the depth of the top and bottom edge of the predicted bounding box to the ground truth. This analysis showed that the top edge can be used to predict the depth of the needle with a maximum error of 8.5 μm. |
| format | Article |
| id | doaj-art-37b43dbad41a4d24976d3f623e76dfd6 |
| institution | OA Journals |
| issn | 2468-2322 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Wiley |
| record_format | Article |
| series | CAAI Transactions on Intelligence Technology |
| spelling | doaj-art-37b43dbad41a4d24976d3f623e76dfd62025-08-20T02:35:01ZengWileyCAAI Transactions on Intelligence Technology2468-23222025-06-0110370371510.1049/cit2.12242Needle detection and localisation for robot‐assisted subretinal injection using deep learningMingchuan Zhou0Xiangyu Guo1Matthias Grimm2Elias Lochner3Zhongliang Jiang4Abouzar Eslami5Juan Ye6Nassir Navab7Alois Knoll8Mohammad Ali Nasseri9Robotic Micro‐nano Manipulation Lab College of Biosystems Engineering and Food Science Zhejiang University Hangzhou ChinaRobotic Micro‐nano Manipulation Lab College of Biosystems Engineering and Food Science Zhejiang University Hangzhou ChinaSchool of Computation, Information and Technology Technische Universität München München GermanySchool of Computation, Information and Technology Technische Universität München München GermanySchool of Computation, Information and Technology Technische Universität München München GermanyCarl Zeiss Meditec AG München GermanyDepartment of Ophthalmology Second Affiliated Hospital of Zhejiang University College of Medicine Hangzhou ChinaSchool of Computation, Information and Technology Technische Universität München München GermanySchool of Computation, Information and Technology Technische Universität München München GermanyAugenklinik und Poliklinik Klinikum rechts der Isar der Technische Universität München München GermanyAbstract Subretinal injection is a complicated task for retinal surgeons to operate manually. In this paper we demonstrate a robust framework for needle detection and localisation in robot‐assisted subretinal injection using microscope‐integrated Optical Coherence Tomography with deep learning. Five convolutional neural networks with different architectures were evaluated. The main differences between the architectures are the amount of information they receive at the input layer. When evaluated on ex‐vivo pig eyes, the top performing network successfully detected all needles in the dataset and localised them with an Intersection over Union value of 0.55. The algorithm was evaluated by comparing the depth of the top and bottom edge of the predicted bounding box to the ground truth. This analysis showed that the top edge can be used to predict the depth of the needle with a maximum error of 8.5 μm.https://doi.org/10.1049/cit2.12242deep learningoptical coherence tomographyrobot‐assisted surgerysubretinal injection |
| spellingShingle | Mingchuan Zhou Xiangyu Guo Matthias Grimm Elias Lochner Zhongliang Jiang Abouzar Eslami Juan Ye Nassir Navab Alois Knoll Mohammad Ali Nasseri Needle detection and localisation for robot‐assisted subretinal injection using deep learning CAAI Transactions on Intelligence Technology deep learning optical coherence tomography robot‐assisted surgery subretinal injection |
| title | Needle detection and localisation for robot‐assisted subretinal injection using deep learning |
| title_full | Needle detection and localisation for robot‐assisted subretinal injection using deep learning |
| title_fullStr | Needle detection and localisation for robot‐assisted subretinal injection using deep learning |
| title_full_unstemmed | Needle detection and localisation for robot‐assisted subretinal injection using deep learning |
| title_short | Needle detection and localisation for robot‐assisted subretinal injection using deep learning |
| title_sort | needle detection and localisation for robot assisted subretinal injection using deep learning |
| topic | deep learning optical coherence tomography robot‐assisted surgery subretinal injection |
| url | https://doi.org/10.1049/cit2.12242 |
| work_keys_str_mv | AT mingchuanzhou needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT xiangyuguo needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT matthiasgrimm needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT eliaslochner needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT zhongliangjiang needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT abouzareslami needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT juanye needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT nassirnavab needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT aloisknoll needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning AT mohammadalinasseri needledetectionandlocalisationforrobotassistedsubretinalinjectionusingdeeplearning |