Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence

Background and objectivesTraumatic optic neuropathy (TON) caused by optic canal fractures (OCF) can result in severe visual impairment, even blindness. Timely and accurate diagnosis and treatment are crucial for preserving visual function. However, diagnosing OCF can be challenging for inexperienced...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu-Lin Li, Yu-Hao Li, Mu-Yang Wei, Guang-Yu Li
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-05-01
Series:Frontiers in Cell and Developmental Biology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fcell.2025.1609028/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850122431265505280
author Yu-Lin Li
Yu-Hao Li
Mu-Yang Wei
Guang-Yu Li
author_facet Yu-Lin Li
Yu-Hao Li
Mu-Yang Wei
Guang-Yu Li
author_sort Yu-Lin Li
collection DOAJ
description Background and objectivesTraumatic optic neuropathy (TON) caused by optic canal fractures (OCF) can result in severe visual impairment, even blindness. Timely and accurate diagnosis and treatment are crucial for preserving visual function. However, diagnosing OCF can be challenging for inexperienced clinicians due to atypical OCF changes in imaging studies and variability in optic canal anatomy. This study aimed to develop an artificial intelligence (AI) image recognition system for OCF to assist in diagnosing OCF and segmenting important anatomical structures in the orbital apex.MethodsUsing the YOLOv7 neural network, we implemented OCF localization and assessment in CT images. To achieve more accurate segmentation of key anatomical structures, such as the internal carotid artery, cavernous sinus, and optic canal, we introduced Selective Kernel Convolution and Transformer encoder modules into the original UNet structure.ResultsThe YOLOv7 model achieved an overall precision of 79.5%, recall of 74.3%, F1 score of 76.8%, and mAP@0.5 of 80.2% in OCF detection. For segmentation tasks, the improved UNet model achieved a mean Intersection over Union (mIoU) of 92.76% and a mean Dice coefficient (mDice) of 90.19%, significantly outperforming the original UNet. Assisted by AI, ophthalmology residents improved their diagnostic AUC-ROC from 0.576 to 0.795 and significantly reduced diagnostic time.ConclusionThis study developed an AI-based system for the diagnosis and treatment of optic canal fractures. The system not only enhanced diagnostic accuracy and reduced surgical collateral damage but also laid a solid foundation for the continuous development of future intelligent surgical robots and advanced smart healthcare systems.
format Article
id doaj-art-eb6e158f4f704e6a808362cf189b2aa5
institution OA Journals
issn 2296-634X
language English
publishDate 2025-05-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Cell and Developmental Biology
spelling doaj-art-eb6e158f4f704e6a808362cf189b2aa52025-08-20T02:34:50ZengFrontiers Media S.A.Frontiers in Cell and Developmental Biology2296-634X2025-05-011310.3389/fcell.2025.16090281609028Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligenceYu-Lin Li0Yu-Hao Li1Mu-Yang Wei2Guang-Yu Li3Department of Ophthalmology, The Second Norman Bethune Hospital of Jilin University, Changchun, ChinaInternational School, Beijing University of Posts and Telecommunications, Bei Jing, ChinaDepartment of Ophthalmology, The Second Norman Bethune Hospital of Jilin University, Changchun, ChinaDepartment of Ophthalmology, The Second Norman Bethune Hospital of Jilin University, Changchun, ChinaBackground and objectivesTraumatic optic neuropathy (TON) caused by optic canal fractures (OCF) can result in severe visual impairment, even blindness. Timely and accurate diagnosis and treatment are crucial for preserving visual function. However, diagnosing OCF can be challenging for inexperienced clinicians due to atypical OCF changes in imaging studies and variability in optic canal anatomy. This study aimed to develop an artificial intelligence (AI) image recognition system for OCF to assist in diagnosing OCF and segmenting important anatomical structures in the orbital apex.MethodsUsing the YOLOv7 neural network, we implemented OCF localization and assessment in CT images. To achieve more accurate segmentation of key anatomical structures, such as the internal carotid artery, cavernous sinus, and optic canal, we introduced Selective Kernel Convolution and Transformer encoder modules into the original UNet structure.ResultsThe YOLOv7 model achieved an overall precision of 79.5%, recall of 74.3%, F1 score of 76.8%, and mAP@0.5 of 80.2% in OCF detection. For segmentation tasks, the improved UNet model achieved a mean Intersection over Union (mIoU) of 92.76% and a mean Dice coefficient (mDice) of 90.19%, significantly outperforming the original UNet. Assisted by AI, ophthalmology residents improved their diagnostic AUC-ROC from 0.576 to 0.795 and significantly reduced diagnostic time.ConclusionThis study developed an AI-based system for the diagnosis and treatment of optic canal fractures. The system not only enhanced diagnostic accuracy and reduced surgical collateral damage but also laid a solid foundation for the continuous development of future intelligent surgical robots and advanced smart healthcare systems.https://www.frontiersin.org/articles/10.3389/fcell.2025.1609028/fulloptic canal fracturesCTenhanced orbital CTdeep learningAIYOLOv7
spellingShingle Yu-Lin Li
Yu-Hao Li
Mu-Yang Wei
Guang-Yu Li
Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence
Frontiers in Cell and Developmental Biology
optic canal fractures
CT
enhanced orbital CT
deep learning
AI
YOLOv7
title Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence
title_full Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence
title_fullStr Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence
title_full_unstemmed Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence
title_short Automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence
title_sort automatic detection of optic canal fractures and recognition and segmentation of anatomical structures in the orbital apex based on artificial intelligence
topic optic canal fractures
CT
enhanced orbital CT
deep learning
AI
YOLOv7
url https://www.frontiersin.org/articles/10.3389/fcell.2025.1609028/full
work_keys_str_mv AT yulinli automaticdetectionofopticcanalfracturesandrecognitionandsegmentationofanatomicalstructuresintheorbitalapexbasedonartificialintelligence
AT yuhaoli automaticdetectionofopticcanalfracturesandrecognitionandsegmentationofanatomicalstructuresintheorbitalapexbasedonartificialintelligence
AT muyangwei automaticdetectionofopticcanalfracturesandrecognitionandsegmentationofanatomicalstructuresintheorbitalapexbasedonartificialintelligence
AT guangyuli automaticdetectionofopticcanalfracturesandrecognitionandsegmentationofanatomicalstructuresintheorbitalapexbasedonartificialintelligence