ABSTRACT
Background: The BONEBRIDGE® (Med-El GmbH) is a bone-conduction device comprising an external audio processor and an internal Bone Conduction-Floating Mass Transducer (BC-FMT) surgically anchored to the temporal bone. Due to the implant's size, its placement may be challenging in certain anatomies, necessitating thorough surgical planning. Manual planning methods are laborious, time-intensive, and prone to errors. This study aimed to develop and validate an automated algorithm for determining skull thickness, aiding in the surgical planning of the BONEBRIDGE and other devices requiring similar bone thickness estimations. Materials and methods: Twelve cadaveric temporal bones underwent clinical computed tomography (CT). A custom Python algorithm was developed to automatically segment bone from soft tissue, generate 3D models, and perform ray-tracing to estimate bone thickness. Two thickness colormaps were generated for each sample: the cortical thickness to the first air cell and the total thickness down to the dura. The algorithm was validated against expert manual measurements to achieve consensus interpretation. Results: The algorithm estimated bone-to-air thicknesses (mean = 4.7 mm, 95% Confidence Interval [CI] of 4.3-5.0 mm) that closely matched the expert measurements (mean = 4.7 mm, CI of 4.4-5.0 mm), with a mean absolute difference (MAD) of 0.3 mm. Similarly, the algorithm's estimations to the dura (6.0 mm, CI of 5.4-6.5 mm) were comparable to the expert markings (5.9 mm, CI of 5.4-6.5 mm), with a MAD of 0.3 mm. Conclusions: The first automated algorithm to calculate skull thickness to both the air cells and dura in the temporal bone was developed. Colormaps were optimized to aid with the surgical planning of BONEBRIDGE implantation, however the tool can be generalized to aid in the surgical planning of any bone thickness application. The tool was published as a freely available extension to the open-source 3D Slicer software program (www.slicer.org).
ABSTRACT
OBJECTIVE: Safe storage of firearms has been recommended as a means of preventing gun-related pediatric injuries, yet few interventions have led to significant improvements in storage practices. This study examined a multifaceted community education campaign to promote safe handgun storage and the campaign's impact on firearm locking and loading practices in households with children. METHODS: Beginning in 1997, a safe-storage campaign consisting of television and radio announcements, educational materials, billboards, and discount coupons for lock boxes was conducted in King County, Washington. The campaign evaluation used a quasi-experimental design and compared the intervention site with 9 control counties outside Washington State and west of the Mississippi River. Cross-sectional, random-digit-dial telephone surveys of handgun-owning households with children were conducted in all study counties both before the intervention in 1996 (n = 302) and again in 2001 (n = 255). The main analyses assessed whether greater improvements in household firearm-storage practices occurred between 1996 and 2001 in the intervention, compared with the control, counties. Primary outcomes were based on up to 3 handguns per household and included (1) all stored with trigger locks, lock boxes, or gun safes (formal locking devices), (2) all stored in lock boxes or gun safes, (3) any stored loaded, (4) any stored loaded without a formal locking device, and (5) any stored loaded and not in a lock box or gun safe. Data were also collected on up to 1 long gun per household; long-gun outcomes included (1) stored with a trigger lock or gun safe and (2) stored loaded. RESULTS: Overall, handguns and long guns were generally more likely to be stored locked and less likely to be loaded in 2001 compared with 1996, with these trends seeming to be more consistent in the intervention county. Even so, more than one quarter of households with children and handguns in 2001 failed to store all of their handguns with a formal locking device, and up to 8% continued to possess at least 1 loaded handgun that was not stored with a formal device. The majority of households that stored their handguns with formal devices used lock boxes or gun safes. Storage of handguns in lock boxes or gun safes became more common in both the intervention (adjusted odds ratio [aOR]: 1.71; 95% confidence interval [CI]: 1.03-2.84) and control households (aOR: 1.66; 95% CI: 1.01-2.72) between 1996 and 2001. None of the other measured changes reached statistical significance, such as storing any household handgun loaded (aOR: 0.71; 95% CI: 0.35-1.42 [intervention]; aOR: 1.08; 95% CI: 0.58-2.00 [control]) or keeping any household handgun loaded and not stored in a lock box or gun safe (aOR: 0.59; 95% CI: 0.22-1.55 [intervention]; aOR: 0.67; 95% CI: 0.30-1.49 [control]). Moreover, the intervention county did not experience significantly greater overall improvements in household storage practices for handguns or long guns than did control counties. CONCLUSIONS: In both the intervention and control counties, households were more likely to lock all handguns in 2001 compared with 1996. After accounting for temporal trends, this educational campaign, combined with economic incentives to purchase lock boxes, did not seem to significantly change safe storage practices in households with handguns and children. Even if the campaign did result in small improvements in firearm safe storage, simultaneous national and state-specific gun-safety activities or legislative efforts may have drawn increasing attention to gun-related issues in the control counties, thereby making it more difficult to identify effects of our specific handgun storage intervention.
