A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography
Abstract Machine learning (ML) holds great promise in transforming healthcare. While published studies have shown the utility of ML models in interpreting medical imaging examinations, these are often evaluated under laboratory settings. The importance of real world evaluation is best illustrated by...
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/0a90015727c949899ea11517db07fa81 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:0a90015727c949899ea11517db07fa81 |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:0a90015727c949899ea11517db07fa812021-12-02T15:09:07ZA real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography10.1038/s41598-021-95533-22045-2322https://doaj.org/article/0a90015727c949899ea11517db07fa812021-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-95533-2https://doaj.org/toc/2045-2322Abstract Machine learning (ML) holds great promise in transforming healthcare. While published studies have shown the utility of ML models in interpreting medical imaging examinations, these are often evaluated under laboratory settings. The importance of real world evaluation is best illustrated by case studies that have documented successes and failures in the translation of these models into clinical environments. A key prerequisite for the clinical adoption of these technologies is demonstrating generalizable ML model performance under real world circumstances. The purpose of this study was to demonstrate that ML model generalizability is achievable in medical imaging with the detection of intracranial hemorrhage (ICH) on non-contrast computed tomography (CT) scans serving as the use case. An ML model was trained using 21,784 scans from the RSNA Intracranial Hemorrhage CT dataset while generalizability was evaluated using an external validation dataset obtained from our busy trauma and neurosurgical center. This real world external validation dataset consisted of every unenhanced head CT scan (n = 5965) performed in our emergency department in 2019 without exclusion. The model demonstrated an AUC of 98.4%, sensitivity of 98.8%, and specificity of 98.0%, on the test dataset. On external validation, the model demonstrated an AUC of 95.4%, sensitivity of 91.3%, and specificity of 94.1%. Evaluating the ML model using a real world external validation dataset that is temporally and geographically distinct from the training dataset indicates that ML generalizability is achievable in medical imaging applications.Hojjat SalehinejadJumpei KitamuraNoah DitkofskyAmy LinAditya BharathaSuradech SuthiphosuwanHui-Ming LinJefferson R. WilsonMuhammad MamdaniErrol ColakNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
Medicine R Science Q |
| spellingShingle |
Medicine R Science Q Hojjat Salehinejad Jumpei Kitamura Noah Ditkofsky Amy Lin Aditya Bharatha Suradech Suthiphosuwan Hui-Ming Lin Jefferson R. Wilson Muhammad Mamdani Errol Colak A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography |
| description |
Abstract Machine learning (ML) holds great promise in transforming healthcare. While published studies have shown the utility of ML models in interpreting medical imaging examinations, these are often evaluated under laboratory settings. The importance of real world evaluation is best illustrated by case studies that have documented successes and failures in the translation of these models into clinical environments. A key prerequisite for the clinical adoption of these technologies is demonstrating generalizable ML model performance under real world circumstances. The purpose of this study was to demonstrate that ML model generalizability is achievable in medical imaging with the detection of intracranial hemorrhage (ICH) on non-contrast computed tomography (CT) scans serving as the use case. An ML model was trained using 21,784 scans from the RSNA Intracranial Hemorrhage CT dataset while generalizability was evaluated using an external validation dataset obtained from our busy trauma and neurosurgical center. This real world external validation dataset consisted of every unenhanced head CT scan (n = 5965) performed in our emergency department in 2019 without exclusion. The model demonstrated an AUC of 98.4%, sensitivity of 98.8%, and specificity of 98.0%, on the test dataset. On external validation, the model demonstrated an AUC of 95.4%, sensitivity of 91.3%, and specificity of 94.1%. Evaluating the ML model using a real world external validation dataset that is temporally and geographically distinct from the training dataset indicates that ML generalizability is achievable in medical imaging applications. |
| format |
article |
| author |
Hojjat Salehinejad Jumpei Kitamura Noah Ditkofsky Amy Lin Aditya Bharatha Suradech Suthiphosuwan Hui-Ming Lin Jefferson R. Wilson Muhammad Mamdani Errol Colak |
| author_facet |
Hojjat Salehinejad Jumpei Kitamura Noah Ditkofsky Amy Lin Aditya Bharatha Suradech Suthiphosuwan Hui-Ming Lin Jefferson R. Wilson Muhammad Mamdani Errol Colak |
| author_sort |
Hojjat Salehinejad |
| title |
A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography |
| title_short |
A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography |
| title_full |
A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography |
| title_fullStr |
A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography |
| title_full_unstemmed |
A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography |
| title_sort |
real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/0a90015727c949899ea11517db07fa81 |
| work_keys_str_mv |
AT hojjatsalehinejad arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT jumpeikitamura arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT noahditkofsky arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT amylin arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT adityabharatha arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT suradechsuthiphosuwan arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT huiminglin arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT jeffersonrwilson arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT muhammadmamdani arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT errolcolak arealworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT hojjatsalehinejad realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT jumpeikitamura realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT noahditkofsky realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT amylin realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT adityabharatha realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT suradechsuthiphosuwan realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT huiminglin realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT jeffersonrwilson realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT muhammadmamdani realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography AT errolcolak realworlddemonstrationofmachinelearninggeneralizabilityinthedetectionofintracranialhemorrhageonheadcomputerizedtomography |
| _version_ |
1718387857316380672 |