Exploratory machine learning analysis to characterize angioscopic features associated with atherosclerosis-related aortic dissection: an exploratory single-center angioscopic study - Report - MDSpire

Exploratory machine learning analysis to characterize angioscopic features associated with atherosclerosis-related aortic dissection: an exploratory single-center angioscopic study

  • By

  • Satoru Takahashi

  • Sei Komatsu

  • Chikao Yutani

  • Hiroyuki Nishi

  • Yoshiharu Higuchi

  • Nobuzo Iwa

  • Tomoki Ohara

  • Mitsuhiko Takewa

  • Kazuhisa Kodama

  • May 7, 2026

  • 0 min

Share

Clinical Report: Investigation of Machine Learning Techniques to Identify Angioscopic Characteristics Linked to Atherosclerosis-Induced Aortic Dissection

Overview

Revise to specify the machine learning techniques used and their specific contributions.

Background

Aortic dissection (AD) is a critical cardiovascular emergency that poses significant risks to life, necessitating early detection and prompt intervention. Traditional imaging modalities often fall short in identifying early signs of AD, highlighting the need for advanced techniques. Non-obstructive general angioscopy (NOGA) offers a promising approach to visualize atherosclerotic changes that may contribute to AD, potentially improving diagnostic accuracy.

Data Highlights

FeatureImportance
Intramural Blood (IB)Most consistent SRAPI associated with AD
Puff Sign (P)Strong importance
Salmon-Pink Appearance (SP)Strong importance
Fissure Bleeding (FB)Highly frequent but variable importance

Key Findings

  • Intramural blood (IB) was identified as the most consistent SRAPI linked to AD.
  • Puff sign (P) and salmon-pink appearance (SP) also demonstrated significant relevance.
  • Fissure bleeding (FB) was common but showed variable importance across different analytical methods.
  • Machine learning techniques effectively managed imbalanced data and identified key SRAPI features.
  • Network analysis revealed distinct patterns between AD patients and controls.

Clinical Implications

The findings suggest that specific angioscopic characteristics, particularly IB and SP, may serve as important indicators of aortic injury in AD. Clinicians should consider the potential of NOGA as a diagnostic tool to enhance early detection and management of aortic dissection.

Conclusion

Highlight specific future research directions needed to validate the study's findings.

Related Resources & Content

  1. A non-invasive approach utilizing machine learning for the identification of atherosclerotic coronary artery aneurysms, Springer, 2022 -- https://link.springer.com/article/10.1007/s11548-022-02725-w
  2. Developing a carotid ultrasound radiomics-semantic fusion model to identify aortic dissection: a two-center retrospective study, Frontiers in Medicine, 2026 -- https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2026.1813428/full
  3. Link Between Aortic Valve Sclerosis and Subsequent Myocardial Infarction Following Initial Event: Insights from an Observational Study Utilizing Topological Data Analysis, European Journal of Preventive Cardiology, 2024 -- https://academic.oup.com/eurjpc/article/32/17/1789/7619300
  4. Fibro-calcific remodeling of non-stenotic aortic valves as an indicator of recurrent myocardial infarction risk, European Journal of Preventive Cardiology, 2024 -- https://academic.oup.com/eurjpc/article/32/17/1779/7607904
  5. 2024 ESC Guidelines for PAD and Aortic Diseases: Key Points - American College of Cardiology -- https://www.acc.org/Latest-in-Cardiology/ten-points-to-remember/2024/09/03/18/59/2024-esc-guidelines-for-pad-esc-2024
  6. 2024 ESC Guidelines for PAD and Aortic Diseases: Key Points - American College of Cardiology
  7. https://www.jstage.jst.go.jp/article/jcad/30/4/30_30.24-00016/_pdf
  8. Segmentation-model-based framework to detect aortic dissection on non-contrast CT images: a retrospective study | Insights into Imaging | Springer Nature Link

Original Source(s)

Exploratory machine learning analysis to characterize angioscopic features associated with atherosclerosis-related aortic dissection: an exploratory single-center angioscopic study

  • by Satoru Takahashi, Sei Komatsu, Chikao Yutani, Hiroyuki Nishi, Yoshiharu Higuchi, Nobuzo Iwa, Tomoki Ohara, Mitsuhiko Takewa, Kazuhisa Kodama

  • May 7, 2026

Related Content

Tenecteplase Fails to Boost EVT Outcomes

Phase 3 stroke trial leaves open the possibility of clinically meaningful benefit or harm in selected late-window patients.

Alignment Between Cardiologists and AI-Driven Diagnostic Systems: Mixed Methods Study

Who Decides When a Doctor Steps Back?

Cognitive decline, credentialing, and why “I’ll know” often isn’t enough.