Pre-operative planning of minimally invasive mitral valve surgery using deep-learning

Abstract

Aim: We propose a novel pre-operative planning approach based on Deep-Learning techniques in the context of minimally invasive mitral valve surgery (MIMVS), for the identification of the mitral valve and optimal thoracic working port positioning in a patient-specific fashion.

Methods: We used supervised Deep-Learning for the processing of contrast-enhanced computed tomography (CT) scans. Our algorithm consisted of four steps: segmentation on CT scans, localization of the mitral valve, creation of maps under three criteria (distance/angle between the mitral valve plane and candidate working port spots, absence/presence of ribs), and selection of optimal working port.

Results: We compared the performance of the Deep-Learning-based approach vs. the previously described semiautomatic method and conventional user’s planning (Dice mean value: 93.59). The Deep-Learning method outperformed the semiautomatic method [intraclass correlation coefficient (ICC) = 0.206]. We defined two interfaces to navigate among candidate working ports.

Conclusion: We suggest that the Deep-Learning-based approach may help the surgeon in identifying the most appropriate working port (thoracic access) for MIMVS, and to comparatively predict the features of different candidate thoracic accesses in individual patients, and help address individual anatomic issues. It may also help standardize the pre-operative planning and obtain a faster learning curve for trainee surgeons.

Keywords

Machine-learning / deep-learning / minimally invasive cardiac surgery / mitral valve

References

[1] Ito T.Minimally invasive mitral valve surgery through right mini-thoracotomy: recommendations for good exposure, stable cardiopulmonary bypass, and secure myocardial protection.Gen Thorac Cardiovasc Surg.2015;63:371-8

[2] Jung Y,Gietema H,Sardari Nia P.Prediction of poor exposure in endoscopic mitral valve surgery using computed tomography.Eur J Cardiothorac Surg.2024;65:ezae070 PMCID:PMC10927309

[3] Litjens G,Wolterink JM.State-of-the-art deep learning in cardiovascular image analysis.JACC Cardiovasc Imaging.2019;12:1549-65

[4] Wasserthal J,Meyer MT.Totalsegmentator: robust segmentation of 104 anatomic structures in CT images.Radiol Artif Intell.2023;5:e230024 PMCID:PMC10546353

[5] Isensee F,Kohl SAA,Maier-Hein KH.nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation.Nat Methods.2021;18:203-11

[6] Fedorov A,Kalpathy-Cramer J.3D Slicer as an image computing platform for the quantitative imaging network.Magn Reson Imaging.2012;30:1323-41 PMCID:PMC3466397

[7] Finnegan RN,Chlap P.Open-source, fully-automated hybrid cardiac substructure segmentation: development and optimisation.Phys Eng Sci Med.2023;46:377-93 PMCID:PMC10030448

[8] Oktay O, Schlemper J, Folgoc LL, et al. Attention u-net: learning where to look for the pancreas. J Digit Imaging. 2018. Available from: https://www.semanticscholar.org/paper/Attention-U-Net%3A-Learning-Where-to-Look-for-the-Oktay-Schlemper/ae1c89817a3a239e5344293138bdd80293983460. [Last accessed on 23 Mar 2025].

[9] Tao Q,Wang Y.Deep learning-based method for fully automatic quantification of left ventricle function from cine MR images: a multivendor, multicenter study.Radiology.2019;290:81-8

[10] Heuts S,Sardari Nia P.Preoperative planning of left-sided valve surgery with 3D computed tomography reconstruction models: sternotomy or a minimally invasive approach?.Interact Cardiovasc Thorac Surg2016;22:587-93 PMCID:6716458

[11] Miles TJ.Machine learning in cardiac surgery: a narrative review.J Thorac Dis.2024;16:2644-53 PMCID:PMC11087616

[12] Di Perna D, Castro M, Gasc Y, Haigron P, Verhoye JP, Anselmi A. Patient-specific access planning in minimally invasive mitral valve surgery.Med Hypotheses.2020;136:109475

[13] Castro M,Rinnert Y,Verhoye JP.Robot-assisted mitral valve surgery may be supported by patient-specific planning aid in port placement.Med Hypotheses.2024;192:111464