AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT
Abstract The aim of the study was to evaluate the impact of the newly developed Similar patient search (SPS) Web Service, which supports reading complex lung diseases in computed tomography (CT), on the diagnostic accuracy of residents. SPS is an image-based search engine for pre-diagnosed cases alo...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-03-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-29949-3 |
| _version_ | 1797864954866434048 |
|---|---|
| author | Johannes Haubold Ke Zeng Sepehr Farhand Sarah Stalke Hannah Steinberg Denise Bos Mathias Meetschen Anisa Kureishi Sebastian Zensen Tim Goeser Sandra Maier Michael Forsting Felix Nensa |
| author_facet | Johannes Haubold Ke Zeng Sepehr Farhand Sarah Stalke Hannah Steinberg Denise Bos Mathias Meetschen Anisa Kureishi Sebastian Zensen Tim Goeser Sandra Maier Michael Forsting Felix Nensa |
| author_sort | Johannes Haubold |
| collection | DOAJ |
| description | Abstract The aim of the study was to evaluate the impact of the newly developed Similar patient search (SPS) Web Service, which supports reading complex lung diseases in computed tomography (CT), on the diagnostic accuracy of residents. SPS is an image-based search engine for pre-diagnosed cases along with related clinical reference content ( https://eref.thieme.de ). The reference database was constructed using 13,658 annotated regions of interest (ROIs) from 621 patients, comprising 69 lung diseases. For validation, 50 CT scans were evaluated by five radiology residents without SPS, and three months later with SPS. The residents could give a maximum of three diagnoses per case. A maximum of 3 points was achieved if the correct diagnosis without any additional diagnoses was provided. The residents achieved an average score of 17.6 ± 5.0 points without SPS. By using SPS, the residents increased their score by 81.8% to 32.0 ± 9.5 points. The improvement of the score per case was highly significant (p = 0.0001). The residents required an average of 205.9 ± 350.6 s per case (21.9% increase) when SPS was used. However, in the second half of the cases, after the residents became more familiar with SPS, this increase dropped to 7%. Residents’ average score in reading complex chest CT scans improved by 81.8% when the AI-driven SPS with integrated clinical reference content was used. The increase in time per case due to the use of the SPS was minimal. |
| first_indexed | 2024-04-09T23:01:01Z |
| format | Article |
| id | doaj.art-ab15be3d5a064ff5a8fb94d62d9d9b1e |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-09T23:01:01Z |
| publishDate | 2023-03-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-ab15be3d5a064ff5a8fb94d62d9d9b1e2023-03-22T10:59:57ZengNature PortfolioScientific Reports2045-23222023-03-0113111210.1038/s41598-023-29949-3AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CTJohannes Haubold0Ke Zeng1Sepehr Farhand2Sarah Stalke3Hannah Steinberg4Denise Bos5Mathias Meetschen6Anisa Kureishi7Sebastian Zensen8Tim Goeser9Sandra Maier10Michael Forsting11Felix Nensa12Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenSiemens Medical Solutions Inc.Siemens Medical Solutions Inc.Georg Thieme Verlag KGDepartment of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenDepartment of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenDepartment of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenInstitute of Artificial Intelligence in Medicine, University Hospital EssenDepartment of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenDepartment of Radiology and Neuroradiology, Kliniken Maria HilfDepartment of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenDepartment of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenDepartment of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital EssenAbstract The aim of the study was to evaluate the impact of the newly developed Similar patient search (SPS) Web Service, which supports reading complex lung diseases in computed tomography (CT), on the diagnostic accuracy of residents. SPS is an image-based search engine for pre-diagnosed cases along with related clinical reference content ( https://eref.thieme.de ). The reference database was constructed using 13,658 annotated regions of interest (ROIs) from 621 patients, comprising 69 lung diseases. For validation, 50 CT scans were evaluated by five radiology residents without SPS, and three months later with SPS. The residents could give a maximum of three diagnoses per case. A maximum of 3 points was achieved if the correct diagnosis without any additional diagnoses was provided. The residents achieved an average score of 17.6 ± 5.0 points without SPS. By using SPS, the residents increased their score by 81.8% to 32.0 ± 9.5 points. The improvement of the score per case was highly significant (p = 0.0001). The residents required an average of 205.9 ± 350.6 s per case (21.9% increase) when SPS was used. However, in the second half of the cases, after the residents became more familiar with SPS, this increase dropped to 7%. Residents’ average score in reading complex chest CT scans improved by 81.8% when the AI-driven SPS with integrated clinical reference content was used. The increase in time per case due to the use of the SPS was minimal.https://doi.org/10.1038/s41598-023-29949-3 |
| spellingShingle | Johannes Haubold Ke Zeng Sepehr Farhand Sarah Stalke Hannah Steinberg Denise Bos Mathias Meetschen Anisa Kureishi Sebastian Zensen Tim Goeser Sandra Maier Michael Forsting Felix Nensa AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT Scientific Reports |
| title | AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT |
| title_full | AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT |
| title_fullStr | AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT |
| title_full_unstemmed | AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT |
| title_short | AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT |
| title_sort | ai co pilot content based image retrieval for the reading of rare diseases in chest ct |
| url | https://doi.org/10.1038/s41598-023-29949-3 |
| work_keys_str_mv | AT johanneshaubold aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT kezeng aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT sepehrfarhand aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT sarahstalke aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT hannahsteinberg aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT denisebos aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT mathiasmeetschen aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT anisakureishi aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT sebastianzensen aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT timgoeser aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT sandramaier aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT michaelforsting aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct AT felixnensa aicopilotcontentbasedimageretrievalforthereadingofrarediseasesinchestct |