A systematic review of image segmentation methodology, used in the additive manufacture of patient-specific 3D printed models of the cardiovascular system

Abstract

BackgroundShortcomings in existing methods of image segmentation preclude the widespread adoption of patient-specific 3D printing as a routine decision-making tool in the care of those with congenital heart disease. We sought to determine the range of cardiovascular segmentation methods and how long each of these methods takes.MethodsA systematic review of literature was undertaken. Medical imaging modality, segmentation methods, segmentation time, segmentation descriptive quality (SDQ) and segmentation software were recorded.ResultsTotally 136 studies met the inclusion criteria (1 clinical trial; 80 journal articles; 55 conference, technical and case reports). The most frequently used image segmentation methods were brightness thresholding, region growing and manual editing, as supported by the most popular piece of proprietary software: Mimics (Materialise NV, Leuven, Belgium, 1992–2015). The use of bespoke software developed by individual authors was not uncommon. SDQ indicated that reporting of image segmentation methods was generally poor with only one in three accounts providing sufficient detail for their procedure to be reproduced.Conclusions and implication of key findingsPredominantly anecdotal and case reporting precluded rigorous assessment of risk of bias and strength of evidence. This review finds a reliance on manual and semi-automated segmentation methods which demand a high level of expertise and a significant time commitment on the part of the operator. In light of the findings, we have made recommendations regarding reporting of 3D printing studies. We anticipate that these findings will encourage the development of advanced image segmentation methods.

Cardiac 3D Printing and its Future Directions

“…(22-25) The desired features of the model and the type of clinical imaging data used will generally dictate the choice of segmentation software. The source of the DICOM data (CT, MRI, or echo), desired complexity of the patient-specific model, and extent of operator experience with the software may greatly influence the time required for image segmentation.…”

Structural and congenital heart disease interventions: the role of three-dimensional printing

“…CT and MRI are the most common sources of datasets for 3D printing because of their ability to image the entire heart with detailed intracardiac anatomy. Echocardiography has a superior ability to image fast moving structures such as cardiac valves [7]. Fusion of different modalities (e. g. ventricles from CT, valves from echocardiography) to create a single 3D model has been reported [8].…”

“…Several manual, semi-automatic and automatic image segmentation methods have been used. The most common are: region growing and brightness thresholding followed by manual editing [7]. Region growing examines the relationship of neighbouring pixels to an initial seed point and determines whether the neighbouring pixels should be added as part of that region (Fig.…”

“Black Bone” MRI: a novel imaging technique for 3D printing

“…Success has previously been reported with MRI for some soft tissue structures such Figure 2 The first three-dimensional printed anatomical model ever produced from "Black Bone" MRI of the mandible of an adult volunteer. as the heart and brain; [13][14][15] however, CT remains the imaging modality of choice in the majority of 3D printing examples in the literature. Whilst "Black Bone" MRI goes some way to address the problem by enhancing the soft tissue-bone boundary, at present it is not a perfect solution.…”