When the library is located in prime real estate: a case study on the loss of space from the Duke University Medical Center Library and Archives.

The Duke University Medical Center Library and Archives is located in the heart of the Duke Medicine campus, surrounded by Duke Hospital, ambulatory clinics, and numerous research facilities. Its location is considered prime real estate, given its adjacency to patient care, research, and educational activities. In 2005, the Duke University Library Space Planning Committee had recommended creating a learning center in the library that would support a variety of educational activities. However, the health system needed to convert the library's top floor into office space to make way for expansion of the hospital and cancer center. The library had only five months to plan the storage and consolidation of its journal and book collections, while working with the facilities design office and architect on the replacement of key user spaces on the top floor. Library staff worked together to develop plans for storing, weeding, and consolidating the collections and provided input into renovation plans for users spaces on its mezzanine level. The library lost 15,238 square feet (29%) of its net assignable square footage and a total of 16,897 (30%) gross square feet. This included 50% of the total space allotted to collections and over 15% of user spaces. The top-floor space now houses offices for Duke Medicine oncology faculty and staff. By storing a large portion of its collection off-site, the library was able to remove more stacks on the remaining stack level and convert them to user spaces, a long-term goal for the library. Additional space on the mezzanine level had to be converted to replace lost study and conference room spaces. While this project did not match the recommended space plans for the library, it underscored the need for the library to think creatively about the future of its facility and to work toward a more cohesive master plan.

Criminally Involved Parents Who Misuse Substances and Children’s Odds of Being Arrested as a Young Adult: Do Drug Treatment Courts Mitigate the Risk?

© 2016, Springer Science+Business Media New York.This paper examined (1) the association between parents who are convicted of a substance-related offense and their children’s probability of being arrested as a young adult and (2) whether or not parental participation in an adult drug treatment court program mitigated this risk. The analysis relied on state administrative data from North Carolina courts (2005–2013) and from birth records (1988–2003). The dependent variable was the probability that a child was arrested as a young adult (16–21). Logistic regression was used to compare groups and models accounted for the clustering of multiple children with the same mother. Findings revealed that children whose parents were convicted on either a substance-related charge on a non-substance-related charge had twice the odds of being arrested as young adult, relative to children whose parents had not been observed having a conviction. While a quarter of children whose parents participated in a drug treatment court program were arrested as young adults, parental completion this program did not reduce this risk. In conclusion, children whose parents were convicted had an increased risk of being arrested as young adults, irrespective of whether or not the conviction was on a substance-related charge. However, drug treatment courts did not reduce this risk. Reducing intergenerational links in the probability of arrest remains a societal challenge.

Using Image Processing Methods to Improve the Detection of Buried Explosive Threats in GPR Data

Current state of the art techniques for landmine detection in ground penetrating radar (GPR) utilize statistical methods to identify characteristics of a landmine response. This research makes use of 2-D slices of data in which subsurface landmine responses have hyperbolic shapes. Various methods from the field of visual image processing are adapted to the 2-D GPR data, producing superior landmine detection results. This research goes on to develop a physics-based GPR augmentation method motivated by current advances in visual object detection. This GPR specific augmentation is used to mitigate issues caused by insufficient training sets. This work shows that augmentation improves detection performance under training conditions that are normally very difficult. Finally, this work introduces the use of convolutional neural networks as a method to learn feature extraction parameters. These learned convolutional features outperform hand-designed features in GPR detection tasks. This work presents a number of methods, both borrowed from and motivated by the substantial work in visual image processing. The methods developed and presented in this work show an improvement in overall detection performance and introduce a method to improve the robustness of statistical classification.