Probabilistic Method to Fuse Artificial Intelligence-Generated Underground Utility Mapping.

Utility as-built plans, which typically provide information about underground utilities' position and spatial locations, are known to comprise inaccuracies. Over the years, the reliance on utility investigations using an array of sensing equipment has increased in an attempt to resolve utility as-built inaccuracies and mitigate the high rate of accidental underground utility strikes during excavation activities. Adapting data fusion into utility engineering and investigation practices has been shown to be effective in generating information with improved accuracy. However, the complexities in data interpretation and associated prohibitive costs, especially for large-scale projects, are limiting factors. This paper addresses the problem of data interpretation, costs, and large-scale utility mapping with a novel framework that generates probabilistic inferences by fusing data from an automatically generated initial map with as-built data. The probabilistic inferences expose regions of high uncertainty, highlighting them as prime targets for further investigations. The proposed model is a collection of three main processes. First, the automatic initial map creation is a novel contribution supporting rapid utility mapping by subjecting identified utility appurtenances to utility inference rules. The second and third processes encompass the fusion of the created initial utility map with available knowledge from utility as-builts or historical satellite imagery data and then evaluating the uncertainties using confidence value estimators. The proposed framework transcends the point estimation of buried utility locations in previous works by producing a final probabilistic utility map, revealing a confidence level attributed to each segment linking aboveground features. In this approach, the utility infrastructure is rapidly mapped at a low cost, limiting the extent of more detailed utility investigations to low-confidence regions. In resisting obsolescence, another unique advantage of this framework is the dynamic nature of the mapping to automatically update information upon the arrival of new knowledge. This ultimately minimizes the problem of utility as-built accuracies dwindling over time.

Review of Artificial Intelligence Applications for Virtual Sensing of Underground Utilities.

Accurately identifying the location and depth of buried utility assets became a considerable challenge in the construction industry, for which accidental strikes can cause important economic losses and safety concerns. While the collection of as-built utility locations is becoming more accurate, there still exists an important need to be capable of accurately detecting buried utilities in order to eliminate risks associated with digging. Current practices typically involve the use of trained agents to survey and detect underground utilities at locations of interest, which is a costly and time-consuming process. With advances in artificial intelligence (AI), an opportunity arose in conducting virtual sensing of buried utilities by combining robotics (e.g., drones), knowledge, and logic. This paper reviewed methods that are based on AI in mapping underground infrastructure. In particular, the use of AI in aerial and terrestrial mapping of utility assets was reviewed, followed by a summary of AI techniques used in fusing multi-source data in creating underground infrastructure maps. Key observations from the consolidated literature were that (1) when leveraging computer vision methods, automatic mapping techniques vastly focus on manholes localized from aerial imagery; (2) when applied to non-intrusive sensing, AI methods vastly focus on empowering ground-penetrating radar (GPR)-produced data; and (3) data fusion techniques to produce utility maps should be extended to any utility assets/types. Based on these observations, a universal utility mapping model was proposed, one that could enable mapping of underground utilities using limited information available in the form of different sources of data and knowledge.

Biomechanical testing of locking and nonlocking plates in the canine scapula.

Locking plates have been shown to offer improved fixation in fractures involving either osteoporotic bone or bone with lesser screw pullout strength, such as thin and flat bones. Fractures of the scapular body are one type of fracture where the screw pullout strength using conventional plate fixation may not be sufficient to overcome physiologic forces. The purpose of this study was to compare the pullout strengths of locking plates to conventional nonlocking plates in the canine scapula. A 2.7 mm string of pearls plate (SOP) and a 2.7 mm limited contact dynamic compression plate (LC-DCP) were applied with similar divergent screws to the supraspinatus fossa of the scapula. Forces perpendicular to the plates were applied and both the loads at failure and modes of failure were recorded. No differences were noted in loads at failure between the two plating systems. Although the modes of failure were not significantly different, the SOP constructs tended to fail more often by bone slicing and coring, whereas the LC-DCP constructs failed primarily by screw stripping. Neither of the plate systems used in this study demonstrated a distinct mechanical advantage. The application and limitations of locking plate systems in various clinical situations require further study.

Perpendicular pull-out force of locking versus non-locking plates in thin cortical bone using a canine mandibular ramus model.

OBJECTIVE: To compare the holding strength of a conventional plate-screw construct with a locking plate-screw construct in the thin cortical bone of the canine mandibular ramus. STUDY DESIGN: Mechanical study. ANIMALS: Paired cadaveric canine mandibles (n = 10 pairs). METHODS: Perpendicular pull-out testing was performed on 2.0-mm limited-contact dynamic compression plate (LC-DCP)-screw constructs and 2.0-mm string-of-pearls (SOP) plate-screw constructs applied to the ramus of the canine mandible. Pull-out force was applied perpendicular to the long axis of the plates. Construct stiffness and load at failure were determined from load-displacement curves and method of failure was determined from high speed digital video recordings. A paired t-test was used to compare mean construct stiffness and load at failure between groups. RESULTS: SOP plate-screw constructs had a significantly higher mean construct stiffness and load at failure than did LC-DCP-screw constructs. LC-DCP constructs failed by screw pull-out while bone slicing and fracture were prominent mechanisms of failure for SOP constructs. CONCLUSIONS: SOP plate-screw constructs sustained a significantly higher perpendicular load at failure than did LC-DCP-screw constructs.