Deep learning automated segmentation of middle skull‐base structures for enhanced navigation

Neves, Caio Athayde; Tran, Emma; Blevins, Nikolas H.; Hwang, Peter H.

doi:10.1002/alr.22856

Cited by 12 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Likewise, graphical information undergoes heavy preprocessing before AI submission. For example, only three reviewed articles used three-dimensional native volume information to allow segmentation of sinonasal structures [ 25 , 27 , 33 ], eight studies used native bidimensional images, and all others used some form of data manipulation.…”

Section: Discussionmentioning

confidence: 99%

Artificial intelligence, machine learning, and deep learning in rhinology: a systematic review

Bulfamante

Ferella

Miller

et al. 2022

Eur Arch Otorhinolaryngol

View full text Add to dashboard Cite

Purpose This PRISMA-compliant systematic review aims to analyze the existing applications of artificial intelligence (AI), machine learning, and deep learning for rhinological purposes and compare works in terms of data pool size, AI systems, input and outputs, and model reliability. Methods MEDLINE, Embase, Web of Science, Cochrane Library, and ClinicalTrials.gov databases. Search criteria were designed to include all studies published until December 2021 presenting or employing AI for rhinological applications. We selected all original studies specifying AI models reliability. After duplicate removal, abstract and full-text selection, and quality assessment, we reviewed eligible articles for data pool size, AI tools used, input and outputs, and model reliability. Results Among 1378 unique citations, 39 studies were deemed eligible. Most studies (n = 29) were technical papers. Input included compiled data, verbal data, and 2D images, while outputs were in most cases dichotomous or selected among nominal classes. The most frequently employed AI tools were support vector machine for compiled data and convolutional neural network for 2D images. Model reliability was variable, but in most cases was reported to be between 80% and 100%. Conclusions AI has vast potential in rhinology, but an inherent lack of accessible code sources does not allow for sharing results and advancing research without reconstructing models from scratch. While data pools do not necessarily represent a problem for model construction, presently available tools appear limited in allowing employment of raw clinical data, thus demanding immense interpretive work prior to the analytic process.

show abstract

Section: Discussionmentioning

confidence: 99%

Artificial intelligence, machine learning, and deep learning in rhinology: a systematic review

Bulfamante

Ferella

Miller

et al. 2022

Eur Arch Otorhinolaryngol

View full text Add to dashboard Cite

show abstract

“…For example, manual segmentation of the input data (i.e., temporal bone imaging) is often required to identify critical anatomical structures to highlight during surgery. To improve the scalability of AR in the operating room, novel methods for fully automated segmentation of temporal bone CT scans to identify key anatomical structures have recently been described 34–36 . Similar advances in these technical components may decrease the overall technical and logistical barriers to large‐scale adoption of AR in otology and neurotology.…”

Section: Discussionmentioning

confidence: 99%

“…To improve the scalability of AR in the operating room, novel methods for fully automated segmentation of temporal bone CT scans to identify key anatomical structures have recently been described. [34][35][36] Similar advances in these technical components may decrease the overall technical and logistical barriers to large-scale adoption of AR in otology and neurotology.…”

Section: Barriers To Adoption In the Field Of Otology/ Neurotologymentioning

confidence: 99%

Augmented Reality in Otology/Neurotology: A Scoping Review with Implications for Practice and Education

Chen

Ding

et al. 2022

The Laryngoscope

View full text Add to dashboard Cite

Objective To determine how augmented reality (AR) has been applied to the field of otology/neurotology, examine trends and gaps in research, and provide an assessment of the future potential of this technology within surgical practice and education. Data Sources PubMed, EMBASE, and Cochrane Library were assessed from their inceptions through October 2022. A manual bibliography search was also conducted. Review Methods A scoping review was conducted and reported according to PRISMA‐ScR guidelines. Data from studies describing the application of AR to the field of otology/neurotology were evaluated, according to a priori inclusion/exclusion criteria. Exclusion criteria included non‐English language articles, abstracts, letters/commentaries, conference papers, and review articles. Results Eighteen articles covering a diverse range of AR platforms were included. Publication dates spanned from 2007 to 2022 and the rate of publication increased over this time. Six of 18 studies were case series in human patients although the remaining were proof of concepts in cadaveric/artificial/animal models. The most common application of AR was for surgical navigation (14 of 18 studies). Computed tomography was the most common source of input data. Few studies noted potential applications to surgical training. Conclusion Interest in the application of AR to otology/neurotology is growing based on the number of recent publications that use a broad range of hardware, software, and AR platforms. Large gaps in research such as the need for submillimeter registration error must be addressed prior to adoption in the operating room and for educational purposes. Level of Evidence N/A Laryngoscope, 133:1786–1795, 2023

show abstract

“…For both VS and IE models, preprocessing techniques that emphasized structure enhancement compared with their surroundings, as well as data augmentation techniques tested in previous studies (13)(14)(15), were used to improve model training.…”

Section: Model Trainingmentioning

confidence: 99%

Automated Radiomic Analysis of Vestibular Schwannomas and Inner Ears Using Contrast-Enhanced T1-Weighted and T2-Weighted Magnetic Resonance Imaging Sequences and Artificial Intelligence

et al. 2023

Self Cite

View full text Add to dashboard Cite

Objective To objectively evaluate vestibular schwannomas (VSs) and their spatial relationships with the ipsilateral inner ear (IE) in magnetic resonance imaging (MRI) using deep learning. Study Design Cross-sectional study. Patients A total of 490 adults with VS, high-resolution MRI scans, and no previous neurotologic surgery. Interventions MRI studies of VS patients were split into training (390 patients) and test (100 patients) sets. A three-dimensional convolutional neural network model was trained to segment VS and IE structures using contrast-enhanced T1-weighted and T2-weighted sequences, respectively. Manual segmentations were used as ground truths. Model performance was evaluated on the test set and on an external set of 100 VS patients from a public data set (Vestibular-Schwannoma-SEG). Main Outcome Measure(s) Dice score, relative volume error, average symmetric surface distance, 95th-percentile Hausdorff distance, and centroid locations. Results Dice scores for VS and IE volume segmentations were 0.91 and 0.90, respectively. On the public data set, the model segmented VS tumors with a Dice score of 0.89 ± 0.06 (mean ± standard deviation), relative volume error of 9.8 ± 9.6%, average symmetric surface distance of 0.31 ± 0.22 mm, and 95th-percentile Hausdorff distance of 1.26 ± 0.76 mm. Predicted VS segmentations overlapped with ground truth segmentations in all test subjects. Mean errors of predicted VS volume, VS centroid location, and IE centroid location were 0.05 cm3, 0.52 mm, and 0.85 mm, respectively. Conclusions A deep learning system can segment VS and IE structures in high-resolution MRI scans with excellent accuracy. This technology offers promise to improve the clinical workflow for assessing VS radiomics and enhance the management of VS patients.

show abstract

Deep learning automated segmentation of middle skull‐base structures for enhanced navigation

Abstract: Deep learning automated segmentation of middle skull-base structures for enhanced navigation.

Cited by 12 publications

References 10 publications

Artificial intelligence, machine learning, and deep learning in rhinology: a systematic review

Artificial intelligence, machine learning, and deep learning in rhinology: a systematic review

Augmented Reality in Otology/Neurotology: A Scoping Review with Implications for Practice and Education

Automated Radiomic Analysis of Vestibular Schwannomas and Inner Ears Using Contrast-Enhanced T1-Weighted and T2-Weighted Magnetic Resonance Imaging Sequences and Artificial Intelligence

Contact Info

Product

Resources

About