Automated Classification of Airborne Pollen using Neural Networks.

Pollen allergies are considered as a global epidemic nowadays, as they influence more than a quarter of the worldwide population, with this percentage expected to rapidly increase because of ongoing climate change. To date, alerts on high-risk allergenic pollen exposure have been provided only via forecasting models and conventional monitoring methods that are laborious. The aim of this study is to develop and evaluate our own pollen classification model based on deep neural networks. Airborne allergenic pollen have been monitored in Augsburg, Bavaria, Germany, since 2015, using a novel automatic Bio-Aerosol Analyzer (BAA 500, Hund GmbH). The automatic classification system is compared and evaluated against our own, newly developed algorithm. Our model achieves an unweighted average precision of 83.0 % and an unweighted average recall of 77.1 % across 15 classes of pollen taxa. Automatic, real-time information on concentrations of airborne allergenic pollen will significantly contribute to the implementation of timely, personalized management of allergies in the future. It is already clear that new methods and sophisticated models have to be developed so as to successfully switch to novel operational pollen monitoring techniques serving the above need.

Improving Intensive Care Unit and Ward Utilization by Adapting Master Surgery Schedules.

With increasing organizational and financial pressure on hospitals, each individual surgical treatment has to be reviewed and planned thoroughly. Apart from the expensive operating room facilities, proper staffing and planning of downstream units, like the wards or the intensive care units (ICUs), should be considered as well. In this article, we outline the relationship between a master surgery schedule (MSS), i.e., the assignment of surgical blocks to medical specialties, and the bed demand in the downstream units using an analytical model. By using historical data retrieved from the clinical information system and a patient flow model, we applied a recently developed algorithm for predicting bed demand based on the MSSs for patients of 3 surgical subspecialties of a hospital. Simulations with 3 different MSSs were performed. The impact on the required amount of beds in the downstream units was analyzed. We show the potential improvements of the current MSS considering 2 main goals: leveling workload among days and reduction of weekend utilization. We discuss 2 different MSSs, one decreasing the weekend ICU utilization by 20% and the other one reducing maximum ward bed demand by 7%. A test with 12 months of real-life data validates the results. The application of the algorithm provides detailed insights for the hospital into the impact of MSS designs on the bed demand in downstream units. It allowed creating MSSs that avoid peaks in bed demand and high weekend occupancy levels in the ICU and the ward.