Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study

Sato, Junya; Sugimoto, Kento; Suzuki, Yuki; Wataya, Tomohiro; Kita, Kosuke; Nishigaki, Daiki; Tomiyama, Miyuki; Hiraoka, Yu; Hori, Masatoshi; Takeda, Toshihiro; Kido, Shoji; Tomiyama, Noriyuki

doi:https://doi.org/10.1016/j.ebiom.2024.105463

Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study

Sato, Junya; Sugimoto, Kento; Suzuki, Yuki et al.

eBioMedicine, 2024, 110, 105463

Number of Access:1,696（2025-07-23 20:30 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/100628

Link to primary information

File	Format	Terms of use	Size	Views	Date.Available	Description	information
eBioMedicine_110_105463	pdf	None	15.1 MB	18	2025.03.25

Creative Commons : Attribution

Item Information

Output File Export EndNote Basic Export Mendeley

Title	Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study
Creator	CreatorName Sato, Junya
	NRID 1000050984655 1000050984655 ORCID 0000-0002-6874-2399 0000-0002-6874-2399 researchmap permalink kento1109 kento1109 CreatorName Sugimoto, Kento
	NRID 1000020845599 1000020845599 researchmap permalink ykszk ykszk Scopus Author ID 57204202740 57204202740 CreatorName Suzuki, Yuki
	NRID 1000090983023 1000090983023 ORCID 0000-0002-1224-6601 0000-0002-1224-6601 researchmap permalink tomohiro-wataya tomohiro-wataya CreatorName Wataya, Tomohiro
	ORCID 0000-0002-5825-3348 0000-0002-5825-3348 researchmap permalink kosuke-kita kosuke-kita CreatorName Kita, Kosuke
	ORCID 0000-0002-7006-3408 0000-0002-7006-3408 CreatorName Nishigaki, Daiki
	CreatorName Tomiyama, Miyuki
	CreatorName Hiraoka, Yu
	NRID 1000000346206 1000000346206 researchmap permalink read0092260 read0092260 CreatorName Hori, Masatoshi
	NRID 1000070506493 1000070506493 researchmap permalink ttakeda_research ttakeda_research Scopus Author ID 7403572788 7403572788 CreatorName Takeda, Toshihiro
	NRID 1000090314814 1000090314814 ORCID 0000-0001-5293-4110 0000-0001-5293-4110 researchmap permalink read0196068 read0196068 Scopus Author ID 7004389358 7004389358 CreatorName Kido, Shoji
	NRID 1000050294070 1000050294070 researchmap permalink 3430 3430 Scopus Author ID 7006024942 7006024942 CreatorName Tomiyama, Noriyuki
Subject	Anomaly detection
	Artificial intelligence
	Computed tomography
	Named-entity recognition
	Radiology report
Description	Sato J., Sugimoto K., Suzuki Y., et al. Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study. eBioMedicine 110, 105463 (2024); https://doi.org/10.1016/j.ebiom.2024.105463.
Abstract	Background: Artificial intelligence (AI) systems designed to detect abnormalities in abdominal computed tomography (CT) could reduce radiologists’ workload and improve diagnostic processes. However, development of such models has been hampered by the shortage of large expert-annotated datasets. Here, we used information from free-text radiology reports, rather than manual annotations, to develop a deep-learning-based pipeline for comprehensive detection of abdominal CT abnormalities. Methods: In this multicentre retrospective study, we developed a deep-learning-based pipeline to detect abnormalities in the liver, gallbladder, pancreas, spleen, and kidneys. Abdominal CT exams and related free-text reports obtained during routine clinical practice collected from three institutions were used for training and internal testing, while data collected from six institutions were used for external testing. A multi-organ segmentation model and an information extraction schema were used to extract specific organ images and disease information, CT images and radiology reports, respectively, which were used to train a multiple-instance learning model for anomaly detection. Its performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and F1 score against radiologists’ ground-truth labels. Findings: We trained the model for each organ on images selected from 66,684 exams (39,255 patients) and tested it on 300 (295 patients) and 600 (596 patients) exams for internal and external validation, respectively. In the external test cohort, the overall AUC for detecting organ abnormalities was 0.886. Whereas models trained on human-annotated labels performed better with the same number of exams, those trained on larger datasets with labels auto-extracted via the information extraction schema significantly outperformed human-annotated label-derived models. Interpretation: Using disease information from routine clinical free-text radiology reports allows development of accurate anomaly detection models without requiring manual annotations. This approach is applicable to various anatomical sites and could streamline diagnostic processes. Funding: Japan Science and Technology Agency.
Publisher	Elsevier B.V.
Source Title	eBioMedicine
Volume (Issue)	110
Page	105463
Date of Issued	2024-12-01
Language	English
Handle URL	https://hdl.handle.net/11094/100628
eISSN	23523964
Relation.isIdenticalTo	DOI (Publisher Version) https://doi.org/10.1016/j.ebiom.2024.105463
Access Rights	open access
CopyRight	This article is licensed under a Creative Commons Attribution 4.0 International License.
oaire:version	Version of Record
Category	学術雑誌論文 Journal Article

Resource Type	journal article
Local Type	Journal Article
DCTERMS.bibliographicCitation	eBioMedicine. 2024, 110, p. 105463
DC.title	Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study
DC.creator	Sato, Junya
	Sugimoto, Kento
	Suzuki, Yuki
	Wataya, Tomohiro
	Kita, Kosuke
	Nishigaki, Daiki
	Tomiyama, Miyuki
	Hiraoka, Yu
	Hori, Masatoshi
	Takeda, Toshihiro
	Kido, Shoji
	Tomiyama, Noriyuki
DC.publisher	Elsevier B.V.
DC.language' scheme='DCTERMS.RFC1766	English
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2024-12-01
DC.identifier	https://doi.org/10.1016/j.ebiom.2024.105463
DC.subject	Anomaly detection
	Artificial intelligence
	Computed tomography
	Named-entity recognition
	Radiology report
DC.description	Sato J., Sugimoto K., Suzuki Y., et al. Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study. eBioMedicine 110, 105463 (2024); https://doi.org/10.1016/j.ebiom.2024.105463.
DCTERMS.abstract	Background: Artificial intelligence (AI) systems designed to detect abnormalities in abdominal computed tomography (CT) could reduce radiologists’ workload and improve diagnostic processes. However, development of such models has been hampered by the shortage of large expert-annotated datasets. Here, we used information from free-text radiology reports, rather than manual annotations, to develop a deep-learning-based pipeline for comprehensive detection of abdominal CT abnormalities. Methods: In this multicentre retrospective study, we developed a deep-learning-based pipeline to detect abnormalities in the liver, gallbladder, pancreas, spleen, and kidneys. Abdominal CT exams and related free-text reports obtained during routine clinical practice collected from three institutions were used for training and internal testing, while data collected from six institutions were used for external testing. A multi-organ segmentation model and an information extraction schema were used to extract specific organ images and disease information, CT images and radiology reports, respectively, which were used to train a multiple-instance learning model for anomaly detection. Its performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and F1 score against radiologists’ ground-truth labels. Findings: We trained the model for each organ on images selected from 66,684 exams (39,255 patients) and tested it on 300 (295 patients) and 600 (596 patients) exams for internal and external validation, respectively. In the external test cohort, the overall AUC for detecting organ abnormalities was 0.886. Whereas models trained on human-annotated labels performed better with the same number of exams, those trained on larger datasets with labels auto-extracted via the information extraction schema significantly outperformed human-annotated label-derived models. Interpretation: Using disease information from routine clinical free-text radiology reports allows development of accurate anomaly detection models without requiring manual annotations. This approach is applicable to various anatomical sites and could streamline diagnostic processes. Funding: Japan Science and Technology Agency.
DC.rights	This article is licensed under a Creative Commons Attribution 4.0 International License.
DC.format	100628
citation_title	Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study
citation_author	Sato, Junya
	Sugimoto, Kento
	Suzuki, Yuki
	Wataya, Tomohiro
	Kita, Kosuke
	Nishigaki, Daiki
	Tomiyama, Miyuki
	Hiraoka, Yu
	Hori, Masatoshi
	Takeda, Toshihiro
	Kido, Shoji
	Tomiyama, Noriyuki
citation_publisher	Elsevier B.V.
citation_language	English
citation_date	2024-12-01
citation_journal_title	eBioMedicine
citation_volume	110
citation_firstpage	105463
citation_doi	https://doi.org/10.1016/j.ebiom.2024.105463
citation_public_url	https://hdl.handle.net/11094/100628

Annotation-free multi-organ anomaly detection in abdominal CT using free-text radiology reports: a multi-centre retrospective study

Link to primary information

Item Information

Show details