EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Development and evaluation of a deep learning-assisted diagnostic support system for radiographer preliminary clinical evaluation of intracranial hemorrhage

Kazuma Tsukamoto, Mizuho Nishio, Ryo Kurosaki, Yasuyuki Kojita, Hidetoshi Matsuo iD, Munenobu Nogami iD, Kazuki Ishikawa, Ryosuke Hatano, Yuka Shimo, Yuki Komon, Yuya Ueki, Izumi Imaoka und 7 weitere

PeerJ · 17.06.2026 · 14 · e21414

DOI: 10.7717/peerj.21414 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ, online erschienen am 17.06.2026. Verfasst von Kazuma Tsukamoto und 18 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medical Device Engineering, Kobe University Graduate School of Medicine, Kusunoki-cho, Chuo-ku, Kobe, Japan. Hauptthemen laut MEDLINE: Intracranial Hemorrhages, Deep Learning, Tomography, X-Ray Computed. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: JSPS KAKENHI, Cross-ministerial Strategic Innovation Promotion Program and Construction of an Integrated Health Care System, AMED.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Intracranial hemorrhage is life-threatening and requires prompt and accurate diagnosis. Non-contrast head computed tomography is the standard first-line examination, but detecting small hemorrhages and classifying multiple subtypes require substantial expertise. Workforce shortages and increasing diagnostic workloads, especially in emergency settings, further challenge timely decision-making. Artificial intelligence (AI)-assisted interpretation has shown promise for improving accuracy and efficiency. This retrospective study evaluated the effect of AI assistance on the diagnostic performance of radiologic technologists (RTs).

Methods

We analyzed the data for 100 non-contrast head computed tomography examinations (50 positive and 50 negative for hemorrhage) obtained from the Japan Medical Image Database. The interpretations of the five RTs (5–12 years of experience) with and without AI assistance were compared with those of two radiologists. The detection targets were intraparenchymal, intraventricular, subarachnoid, subdural, epidural, and any hemorrhages. We calculated the Area Under the Receiver Operating Characteristic Curve (AUC), accuracy, sensitivity, and specificity. The differences in the AUC for the AI-assisted and unassisted readings were tested using the DeLong method with Bonferroni correction.

Results

Significant AUC improvements were observed for five of the 30 reader–task comparisons (17%) after Bonferroni correction. These improvements were all related to intraventricular ( p = 0.0001 to 0.0071) and subdural ( p = 0.0022 to 0.0071) hemorrhages.

Conclusion

AI assistance significantly improved RT detection of challenging subtypes such as intraventricular and subdural hemorrhages. However, it did not improve the diagnostic accuracy for detecting any hemorrhage overall ( p = 0.0689 to 0.9669). AI can strengthen the role of RTs within task-sharing models and help stabilize preliminary assessments, especially in emergency care and resource-constrained environments.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Intracranial HemorrhagesDeep LearningTomography, X-Ray ComputedHumansRetrospective StudiesSensitivity and SpecificityIntelligent SystemsJapanROC Curve

Stichwörter der Autor:innen

Artificial intelligenceComputed tomographyDeep learningIntracranial hemorrhagePerformance evaluation

Autor:innen und Institutionen

  • Department of Medical Device Engineering, Kobe University Graduate School of Medicine, Kusunoki-cho, Chuo-ku, Kobe, Japan.
  • Center for Radiology and Radiation Oncology, Kobe University Hospital, Kusunoki-cho, Chuo-ku, Kobe, Japan.
  • Department of Radiology, Kobe University Graduate School of Medicine, Chuo-ku, Kobe, Japan.
  • Division of Medical Imaging, Biomedical Imaging Research Center, University of Fukui, Yoshida, Fukui, Japan.
  • GE HealthCare (Japan) Research Division, Hino-shi, Tokyo, Japan.
  • Department of Radiology, Juntendo University Graduate School of Medicine, Bunkyo-ku, Tokyo, Japan.

ORCID-Profile: Hidetoshi Matsuo, Munenobu Nogami, Toshiaki Akashi, Takamichi Murakami

Förderung

  • JSPS KAKENHI
  • Cross-ministerial Strategic Innovation Promotion Program and Construction of an Integrated Health Care System
  • AMED
  • GE Healthcare (Japan), Hino, Japan

Zitieren

APA

Tsukamoto, K., Nishio, M., Kurosaki, R., Kojita, Y., Matsuo, H., Nogami, M., Ishikawa, K., Hatano, R., Shimo, Y., Komon, Y., Ueki, Y., Imaoka, I., Ohmura, K., Matsuo, K., Akashi, T., Aoki, S., Kusaka, A., Murakami, T., & Muragaki, Y. (2026). Development and evaluation of a deep learning-assisted diagnostic support system for radiographer preliminary clinical evaluation of intracranial hemorrhage. PeerJ, 14, e21414. https://doi.org/10.7717/peerj.21414

Vancouver

Tsukamoto K, Nishio M, Kurosaki R, Kojita Y, Matsuo H, Nogami M, et al. Development and evaluation of a deep learning-assisted diagnostic support system for radiographer preliminary clinical evaluation of intracranial hemorrhage. PeerJ. 2026;14:e21414. doi:10.7717/peerj.21414

BibTeX
@article{Tsukamoto2026,
  author = {Tsukamoto, Kazuma and Nishio, Mizuho and Kurosaki, Ryo and Kojita, Yasuyuki and Matsuo, Hidetoshi and Nogami, Munenobu and Ishikawa, Kazuki and Hatano, Ryosuke and Shimo, Yuka and Komon, Yuki and Ueki, Yuya and Imaoka, Izumi and Ohmura, Kazuyuki and Matsuo, Kazuya and Akashi, Toshiaki and Aoki, Shigeki and Kusaka, Akiko and Murakami, Takamichi and Muragaki, Yoshihiro},
  title = {Development and evaluation of a deep learning-assisted diagnostic support system for radiographer preliminary clinical evaluation of intracranial hemorrhage},
  journal = {PeerJ},
  year = {2026},
  volume = {14},
  pages = {e21414},
  doi = {10.7717/peerj.21414}
}
Kostenlos starten →