11/7/2023 0 Comments Fracture photosMoreover, pelvic fracture is a key cause of mortality in the elderly 1, 2, 3, 4. Pelvic fracture can be considered as a significant health concern, representing one of the most common causes of hospitalization and mobility loss 1. The proposed method successfully detected no bone fractures in all except 4 non-fracture subjects (96.4%). This method was then applied to 112 subjects without bone fractures to evaluate over-detection. The AUC with a single orientation was 0.652. Area under the curve (AUC) was 0.824, with 0.805 recall and 0.907 precision. The proposed method was validated in 93 subjects with bone fractures. The 3D fracture region is then obtained by integrating the 2D fracture candidates. Multiple YOLOv3 models in parallel detect 2D fracture candidates in different orientations simultaneously. We assume that an appropriate reconstruction orientation would exist to optimally characterize image features of bone fractures on 3D-CT. We propose a method that evaluates multiple, 2D, real-time object detection systems (YOLOv3 models) in parallel, in which each YOLOv3 model is trained using differently orientated 2D slab images reconstructed from 3D-CT. However, training a DCNN directly using 3D images is complicated, computationally costly, and requires large amounts of training data. ![]() Deep convolutional neural networks (DCNNs) have been used for lesion detection on 2D and 3D medical images. This study proposes an automated method to detect pelvic fractures on 3-dimensional computed tomography (3D-CT). Beyond the regional implications, this study emphasizes the importance of uncertainty reduction in the early stages of fracture analysis, and its effects on establishing a reliable fracture history.Pelvic fracture is one of the leading causes of death in the elderly, carrying a high risk of death within 1 year of fracture. The fifth fracture set is believed to be fault-related and was only observed in one of the wells. In the Geneva Basin, this event is represented by a pair of NE-SW striking conjugates, which forms the majority of observed fractures. The latter is considered the main deformation event of the region: the Late Miocene fold-and-thrust tectonics. This led to the identification of four Mesozoic to Cenozoic stress regimes: 1) a normal or strike-slip regime, 2) a NE-SW reverse regime, 3) a normal regime, and 4) a NW-SE reverse regime. In the two wells, a total of eight fracture sets were defined based on movement and relative chronology, and used to reconstruct the fracture history of the basin. Research has shown that these rocks form a tight reservoir, and fluid flow will rely on the presence of fracture networks. The created workflow was then applied to two wells in the Geneva Basin, Switzerland associated with a geothermal project aiming to produce heat and electricity from the Upper Cretaceous limestone rocks. Based on the results from the analysis, a workflow was created to reduce uncertainties in fracture interpretations from image logs. ![]() This study aims to address this gap, by performing a statistical analysis on five interpretations of the same well, highlighting their differences and similarities. It has been widely recognized that interpretations come with uncertainties nevertheless, there is a current lack of both quantitative and qualitative understanding regarding differences among interpretations. Therefore, fracture analysis has become a fundamental aspect of geosciences, often starting with fracture interpretation from borehole image logs. ![]() Natural fractures play a crucial role in many sustainable subsurface applications, particularly in reservoirs with low primary porosity and permeability. Uncertainty reduction in image log fracture interpretation, and its implications to the geological history of the Geneva Basin, SwitzerlandÄoesburg, Myrthe (TU Delft Civil Engineering & Geosciences)Äaniilidis, Alexandros (graduation committee)
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |