A novel framework for predicting patients at risk of readmission

Uncertainty in decision-making for patients’ risk of re-admission arises due to non-uniform data and lack of knowledge in health system variables. The knowledge of the impact of risk factors will provide clinicians better decision-making and in reducing the number of patients admitted to the hospital. Traditional approaches are not capable to account for the uncertain nature of risk of hospital re-admissions. More problems arise due to large amount of uncertain information. Patients can be at high, medium or low risk of re-admission, and these strata have ill-defined boundaries. We believe that our model that adapts fuzzy regression method will start a novel approach to handle uncertain data, uncertain relationships between health system variables and the risk of re-admission. Because of nature of ill-defined boundaries of risk bands, this approach does allow the clinicians to target individuals at boundaries. Targeting individuals at boundaries and providing them proper care may provide some ability to move patients from high risk to low risk band. In developing this algorithm, we aimed to help potential users to assess the patients for various risk score thresholds and avoid readmission of high risk patients with proper interventions. A model for predicting patients at high risk of re-admission will enable interventions to be targeted before costs have been incurred and health status have deteriorated. A risk score cut off level would flag patients and result in net savings where intervention costs are much higher per patient. Preventing hospital re-admissions is important for patients, and our algorithm may also impact hospital income.

Bit-index sort: A fast non-comparison integer sorting algorithm for permutations

This paper describes a fast integer sorting algorithm, herein referred to as Bit-index sort, which does not use comparisons and is intended to sort partial permutations. Experimental results exhibit linear complexity order in execution time. Bit-index sort uses a bit-array to classify input sequences of distinct integers, and exploits built-in bit functions in C compilers, supported by machine hardware, to retrieve the ordered output sequence. Results show that Bit-index sort outperforms quicksort and counting sort algorithms when compared in their execution time. A parallel approach for Bit-index sort using two simultaneous threads is also included, which obtains further speedups of up to 1.6 compared to its sequential case.