Rapid prototyping of medical image data based on distance field concept

Ueng, Shyh‐Kuang; Huang, Hsin‐Cheng; Chou, Chieh-Shih

doi:10.1109/icasi.2018.8394339

Cited by 1 publication

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar procedures for creating implants, bio-models, and related applications can be found in the work of [16]. In the work of [7], Ueng et al proposed a manufacturing method to fabricate scanned objects from magnetic resonance imaging (MRI) and CT-scan data. Their method computes the iso-surface of the object at first.…”

Section: Related Workmentioning

confidence: 97%

“…In the research presented in [6,7,[13][14][15][16][17], the scanned objects are contained in volume data sets. Before the LM processes starts, they have to be converted into polygonal representations to match the input data format of the slicers.…”

Section: Related Workmentioning

confidence: 99%

“…In the paper of [18], Oropallo and Piegl regarded G-code generation for voxel-based models as one of the top challenges in LM. The work of [6,7,[13][14][15][16][17] and the proposition of [18] inspired us to develop a G-code generator, aiming at translating voxel-based models into G-code programs to enhance the usefulness of LM in medical and other industrial applications.…”

Section: Related Workmentioning

confidence: 99%

“…These geometries cannot be converted into G-codes by using conventional slicers. One can transform these voxel-based models into surface representations at first and then processes the intermediate geometries by using conventional G-code generators [6,7]. However, this format transformation process increases computational costs and induces extra geometric errors.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A G-Code Generator for Volumetric Models

Ueng

Huang

2019

Applied Sciences

Self Cite

View full text Add to dashboard Cite

In layered manufacturing (LM), slicers are employed to convert input geometric models into G-codes. Conventional slicers accept only surface models as input data. Thus, volumetric models have to be converted into polygonal representations to fit the data format of the slicers. This results in extra computational costs and geometric errors. In this article, we present an efficient slicer aiming to generate G-codes for volumetric models. At first, our slicer computes the printing direction by exploring the inertia tensor of the input model to enhance the stability of the printed part and to decrease the build time. Then, it detects and classifies overhangs in the input model and generates necessary support structures by using a pattern-based method. Thirdly, the proposed slicer divides the input model into the skin and internal regions and cuts the model into 2D images. Subsequently, these images are transformed into toolpaths by utilizing texture mapping and graph traversal methods. Finally, the resultant toolpaths are smoothed to reduce staircases and encoded into G-codes. Test results verify that the proposed slicer produces decent G-codes for volumetric models. Scanned objects hidden in volume data can be directly manufactured without generating intermediate polygonal representations. LM processes become more efficient.

show abstract

Section: Related Workmentioning

confidence: 97%