To what extent can 3D model replicate dimensions of individual mitral valve prolapse?

Takashi Shirakawa, Masao Yoshitatsu, Yasushi Koyama, Akira Kurata, Toru Miyoshi, Hiroki Mizoguchi, Takafumi Masai, Koichi Toda, Yoshiki Sawa

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


Determining the complex geometry of mitral valve prolapse is often difficult. We constructed 3D models of six prolapsed mitral valves for surgical assessment, and evaluated how accurately the models could replicate individual valve dimensions. 3D polygon data were constructed based on an original segmentation method for computed tomography images. The model’s replication performance was confirmed via dimensional comparison between the actual hearts during surgery and those models. The results revealed that the prolapsed segments matched in all cases; however, torn chordae were replicated in four cases. The mean height differences were 0.0 mm (SD 1.6, range − 2 to + 2 mm) for the anterolateral side, 0.0 mm (SD 1.7, range − 2 to + 2 mm) for the prolapsed leaflet center, and − 1.5 mm (SD 0.6, range − 1 to − 2 mm) for the posteromedial side. Regression analysis showed a strong and positive correlation, and Bland–Altman plots indicated quantitative similarity of the models to the actual hearts. We concluded that our 3D valve models could replicate the actual mitral valve prolapses within acceptable dimensional differences. Our concepts are useful for better 3D valve creation and better surgical planning with reliable 3D valve models.

Original languageEnglish
Pages (from-to)348-355
Number of pages8
JournalJournal of Artificial Organs
Issue number3
Publication statusPublished - Sep 1 2018


  • 3D model
  • Mitral valve
  • Prolapse
  • Prototyping
  • Surgical simulation

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Biomaterials
  • Biomedical Engineering
  • Cardiology and Cardiovascular Medicine


Dive into the research topics of 'To what extent can 3D model replicate dimensions of individual mitral valve prolapse?'. Together they form a unique fingerprint.

Cite this