Machine learning classification of medication adherence in patients with movement disorders using non-wearable sensors.

Medication non-adherence is a major concern in the healthcare industry and has led to increases in health risks and medical costs. For many neurological diseases, adherence to medication regimens can be assessed by observing movement patterns. However, physician observations are typically assessed based on visual inspection of movement and are limited to clinical testing procedures. Consequently, medication adherence is difficult to measure when patients are away from the clinical setting. The authors propose a data mining driven methodology that uses low cost, non-wearable multimodal sensors to model and predict patients' adherence to medication protocols, based on variations in their gait. The authors conduct a study involving Parkinson's disease patients that are "on" and "off" their medication in order to determine the statistical validity of the methodology. The data acquired can then be used to quantify patients' adherence while away from the clinic. Accordingly, this data-driven system may allow for early warnings regarding patient safety. Using whole-body movement data readings from the patients, the authors were able to discriminate between PD patients on and off medication, with accuracies greater than 97% for some patients using an individually customized model and accuracies of 78% for a generalized model containing multiple patient gait data. The proposed methodology and study demonstrate the potential and effectiveness of using low cost, non-wearable hardware and data mining models to monitor medication adherence outside of the traditional healthcare facility. These innovations may allow for cost effective, remote monitoring of treatment of neurological diseases.

A data mining methodology for predicting early stage Parkinson's disease using non-invasive, high-dimensional gait sensor data.

Parkinson's disease (PD) is the second most common neurological disorder after Alzheimer's disease. Key clinical features of PD are motor-related and are typically assessed by healthcare providers based on qualitative visual inspection of a patient's movement/gait/posture. More advanced diagnostic techniques such as computed tomography scans that measure brain function, can be cost prohibitive and may expose patients to radiation and other harmful effects. To mitigate these challenges, and open a pathway to remote patient-physician assessment, the authors of this work propose a data mining driven methodology that uses low cost, non-invasive sensors to model and predict the presence (or lack therefore) of PD movement abnormalities and model clinical subtypes. The study presented here evaluates the discriminative ability of non-invasive hardware and data mining algorithms to classify PD cases and controls. A 10-fold cross validation approach is used to compare several data mining algorithms in order to determine that which provides the most consistent results when varying the subject gait data. Next, the predictive accuracy of the data mining model is quantified by testing it against unseen data captured from a test pool of subjects. The proposed methodology demonstrates the feasibility of using non-invasive, low cost, hardware and data mining models to monitor the progression of gait features outside of the traditional healthcare facility, which may ultimately lead to earlier diagnosis of emerging neurological diseases.

Quantifying complexity in translational research: an integrated approach.

PURPOSE: The purpose of this paper is to quantify complexity in translational research. The impact of major operational steps and technical requirements is calculated with respect to their ability to accelerate moving new discoveries into clinical practice. DESIGN/METHODOLOGY/APPROACH: A three-phase integrated quality function deployment (QFD) and analytic hierarchy process (AHP) method was used to quantify complexity in translational research. A case study in obesity was used to usability. FINDINGS: Generally, the evidence generated was valuable for understanding various components in translational research. Particularly, the authors found that collaboration networks, multidisciplinary team capacity and community engagement are crucial for translating new discoveries into practice. RESEARCH LIMITATIONS/IMPLICATIONS: As the method is mainly based on subjective opinion, some argue that the results may be biased. However, a consistency ratio is calculated and used as a guide to subjectivity. Alternatively, a larger sample may be incorporated to reduce bias. PRACTICAL IMPLICATIONS: The integrated QFD-AHP framework provides evidence that could be helpful to generate agreement, develop guidelines, allocate resources wisely, identify benchmarks and enhance collaboration among similar projects. ORIGINALITY/VALUE: Current conceptual models in translational research provide little or no clue to assess complexity. The proposed method aimed to fill this gap. Additionally, the literature review includes various features that have not been explored in translational research.

Patient flow in the emergency department: a classification and analysis of admission process policies.

STUDY OBJECTIVE: We investigate the effect of admission process policies on patient flow in the emergency department (ED). METHODS: We surveyed an advisory panel group to determine approaches to admission process policies and classified them as admission decision is made by the team of providers (attending physicians, residents, physician extenders) (type 1) or attending physicians (type 2) on the admitting service, team of providers (type 3), or attending physicians (type 4) in the ED. We developed discrete-event simulation models of patient flow to evaluate the potential effect of the 4 basic policy types and 2 hybrid types, referred to as triage attending physician consultation and remote collaborative consultation on key performance measures. RESULTS: Compared with the current admission process policy (type 1), the alternatives were all effective in reducing the length of stay of admitted patients by 14% to 26%. In other words, patients may spend 1.4 to 2.5 hours fewer on average in the ED before being admitted to internal medicine under a new admission process policy. The improved flow of admitted patients decreased both the ED length of stay of discharged patients and the overall length of stay by up to 5% and 6.4%, respectively. These results are framed in context of teaching mission and physician experience. CONCLUSION: An efficient admission process can reduce waiting times for both admitted and discharged ED patients. This study contributed to demonstrating the potential value of leveraging admission process policies and developing a framework for pursuing these policies.

Interdigitated silver-polymer-based antibacterial surface system activated by oligodynamic iontophoresis - an empirical characterization study.

There is a pressing need to control the occurrences of nosocomial infections due to their detrimental effects on patient well-being and the rising treatment costs. To prevent the contact transmission of such infections via health-critical surfaces, a prophylactic surface system that consists of an interdigitated array of oppositely charged silver electrodes with polymer separations and utilizes oligodynamic iontophoresis has been recently developed. This paper presents a systematic study that empirically characterizes the effects of the surface system parameters on its antibacterial efficacy, and validates the system's effectiveness. In the first part of the study, a fractional factorial design of experiments (DOE) was conducted to identify the statistically significant system parameters. The data were used to develop a first-order response surface model to predict the system's antibacterial efficacy based on the input parameters. In the second part of the study, the effectiveness of the surface system was validated by evaluating it against four bacterial species responsible for several nosocomial infections - Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, and Enterococcus faecalis - alongside non-antibacterial polymer (acrylic) control surfaces. The system demonstrated statistically significant efficacy against all four bacteria. The results indicate that given a constant total effective surface area, the system designed with micro-scale features (minimum feature width: 20 µm) and activated by 15 µA direct current will provide the most effective antibacterial prophylaxis.

Yield improvement for lost mould rapid infiltration forming process by a multistage fractional factorial split plot design.

Statistical design of experiments is widely used among scientists and engineers to understand influential factors in a laboratory or manufacturing process. One of the underlying principles of using the statistical design of experiments method is randomisation, each run of experimental settings will be determined completely unsystematically. In practice, especially in a complicated process that consists of multiple stages, randomisation may pose too high a burden on time and cost.In this study, the multistage fraction factorial split plot design is proposed for green yield improvement in a lost mould rapid infiltration process that has been developed to fabricate zirconia ceramic parts. This design allows a relaxation of the randomisation principle so that certain experimental runs can be carried out in convenient groups. The results indicate that the type of immersion chemical and mould coating play a role in improving process yield. Additionally, the results suggest that a mould infiltration machine should be used to improve the reproducibility of the process.