Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study
In this study. we aimed to detect vestibular schwannomas (VSs) in individual magnetic resonance imaging (MRI) slices by using a 2D-CNN. A pretrained CNN (ResNet-34) was retrained and internally validated using contrast-enhanced T1-weighted (T1c) MRI slices from one institution. In a second step, the...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/14/9/2069 |
| _version_ | 1797505490697060352 |
|---|---|
| author | Carole Koechli Erwin Vu Philipp Sager Lukas Näf Tim Fischer Paul M. Putora Felix Ehret Christoph Fürweger Christina Schröder Robert Förster Daniel R. Zwahlen Alexander Muacevic Paul Windisch |
| author_facet | Carole Koechli Erwin Vu Philipp Sager Lukas Näf Tim Fischer Paul M. Putora Felix Ehret Christoph Fürweger Christina Schröder Robert Förster Daniel R. Zwahlen Alexander Muacevic Paul Windisch |
| author_sort | Carole Koechli |
| collection | DOAJ |
| description | In this study. we aimed to detect vestibular schwannomas (VSs) in individual magnetic resonance imaging (MRI) slices by using a 2D-CNN. A pretrained CNN (ResNet-34) was retrained and internally validated using contrast-enhanced T1-weighted (T1c) MRI slices from one institution. In a second step, the model was externally validated using T1c- and T1-weighted (T1) slices from a different institution. As a substitute, bisected slices were used with and without tumors originating from whole transversal slices that contained part of the unilateral VS. The model predictions were assessed based on the categorical accuracy and confusion matrices. A total of 539, 94, and 74 patients were included for training, internal validation, and external T1c validation, respectively. This resulted in an accuracy of 0.949 (95% CI 0.935–0.963) for the internal validation and 0.912 (95% CI 0.866–0.958) for the external T1c validation. We suggest that 2D-CNNs might be a promising alternative to 2.5-/3D-CNNs for certain tasks thanks to the decreased demand for computational power and the fact that there is no need for segmentations. However, further research is needed on the difference between 2D-CNNs and more complex architectures. |
| first_indexed | 2024-03-10T04:19:20Z |
| format | Article |
| id | doaj.art-cac6677fcd654134aed4adbe5e68dbe6 |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-10T04:19:20Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-cac6677fcd654134aed4adbe5e68dbe62023-11-23T07:54:24ZengMDPI AGCancers2072-66942022-04-01149206910.3390/cancers14092069Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility StudyCarole Koechli0Erwin Vu1Philipp Sager2Lukas Näf3Tim Fischer4Paul M. Putora5Felix Ehret6Christoph Fürweger7Christina Schröder8Robert Förster9Daniel R. Zwahlen10Alexander Muacevic11Paul Windisch12Department of Radiation Oncology, Kantonsspital Winterthur, 8401 Winterthur, SwitzerlandDepartment of Radiation Oncology, Kantonsspital St. Gallen, 9007 St. Gallen, SwitzerlandDepartment of Radiation Oncology, Kantonsspital Winterthur, 8401 Winterthur, SwitzerlandDepartment of Radiology, Kantonsspital St. Gallen, 9007 St. Gallen, SwitzerlandDepartment of Radiology, Kantonsspital St. Gallen, 9007 St. Gallen, SwitzerlandDepartment of Radiation Oncology, Kantonsspital St. Gallen, 9007 St. Gallen, SwitzerlandBerlin Institute of Health at Charité—Universitätsmedizin Berlin, 10117 Berlin, GermanyEuropean Radiosurgery Center, 81377 Munich, GermanyDepartment of Radiation Oncology, Kantonsspital Winterthur, 8401 Winterthur, SwitzerlandDepartment of Radiation Oncology, Kantonsspital Winterthur, 8401 Winterthur, SwitzerlandDepartment of Radiation Oncology, Kantonsspital Winterthur, 8401 Winterthur, SwitzerlandEuropean Radiosurgery Center, 81377 Munich, GermanyDepartment of Radiation Oncology, Kantonsspital Winterthur, 8401 Winterthur, SwitzerlandIn this study. we aimed to detect vestibular schwannomas (VSs) in individual magnetic resonance imaging (MRI) slices by using a 2D-CNN. A pretrained CNN (ResNet-34) was retrained and internally validated using contrast-enhanced T1-weighted (T1c) MRI slices from one institution. In a second step, the model was externally validated using T1c- and T1-weighted (T1) slices from a different institution. As a substitute, bisected slices were used with and without tumors originating from whole transversal slices that contained part of the unilateral VS. The model predictions were assessed based on the categorical accuracy and confusion matrices. A total of 539, 94, and 74 patients were included for training, internal validation, and external T1c validation, respectively. This resulted in an accuracy of 0.949 (95% CI 0.935–0.963) for the internal validation and 0.912 (95% CI 0.866–0.958) for the external T1c validation. We suggest that 2D-CNNs might be a promising alternative to 2.5-/3D-CNNs for certain tasks thanks to the decreased demand for computational power and the fact that there is no need for segmentations. However, further research is needed on the difference between 2D-CNNs and more complex architectures.https://www.mdpi.com/2072-6694/14/9/2069artificial intelligencedeep learningmachine learningvestibularschwannomaneuro-oncology |
| spellingShingle | Carole Koechli Erwin Vu Philipp Sager Lukas Näf Tim Fischer Paul M. Putora Felix Ehret Christoph Fürweger Christina Schröder Robert Förster Daniel R. Zwahlen Alexander Muacevic Paul Windisch Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study Cancers artificial intelligence deep learning machine learning vestibular schwannoma neuro-oncology |
| title | Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study |
| title_full | Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study |
| title_fullStr | Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study |
| title_full_unstemmed | Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study |
| title_short | Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study |
| title_sort | convolutional neural networks to detect vestibular schwannomas on single mri slices a feasibility study |
| topic | artificial intelligence deep learning machine learning vestibular schwannoma neuro-oncology |
| url | https://www.mdpi.com/2072-6694/14/9/2069 |
| work_keys_str_mv | AT carolekoechli convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT erwinvu convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT philippsager convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT lukasnaf convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT timfischer convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT paulmputora convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT felixehret convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT christophfurweger convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT christinaschroder convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT robertforster convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT danielrzwahlen convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT alexandermuacevic convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy AT paulwindisch convolutionalneuralnetworkstodetectvestibularschwannomasonsinglemrislicesafeasibilitystudy |