Comparative study of ten machine learning algorithms for short-term forecasting in gas warning systems.

This research aims to explore more efficient machine learning (ML) algorithms with better performance for short-term forecasting. Up-to-date literature shows a lack of research on selecting practical ML algorithms for short-term forecasting in real-time industrial applications. This research uses a quantitative and qualitative mixed method combining two rounds of literature reviews, a case study, and a comparative analysis. Ten widely used ML algorithms are selected to conduct a comparative study of gas warning systems in a case study mine. We propose a new assessment visualization tool: a 2D space-based quadrant diagram can be used to visually map prediction error assessment and predictive performance assessment for tested algorithms. Overall, this visualization tool indicates that LR, RF, and SVM are more efficient ML algorithms with overall prediction performance for short-term forecasting. This research indicates ten tested algorithms can be visually mapped onto optimal (LR, RF, and SVM), efficient (ARIMA), suboptimal (BP-SOG, KNN, and Perceptron), and inefficient algorithms (RNN, BP_Resilient, and LSTM). The case study finds results that differ from previous studies regarding the ML efficiency of ARIMA, KNN, LR, LSTM, and SVM. This study finds different views on the prediction performance of a few paired algorithms compared with previous studies, including RF and LR, SVM and RF, KNN and ARIMA, KNN and SVM, RNN and ARIMA, and LSTM and SVM. This study also suggests that ARIMA, KNN, LR, and LSTM should be investigated further with additional prediction error assessments. Overall, no single algorithm can fit all applications. This study raises 20 valuable questions for further research.

Addition of contrast in ultrasound screening for hepatocellular carcinoma.

Objective: Screening ultrasound for hepatocellular carcinoma (HCC) identifies lesions which require further characterization by a contrast-enhanced exam to non-invasively diagnose HCC. While ultrasound is recommended in screening, some HCC can be occult on grayscale imaging. The purpose of this study was to determine if the addition of ultrasound contrast (sulfahexafluoride) to screening ultrasound for HCC can identify more HCC lesions than grayscale sonographic imaging alone. Methods: All HCC screening ultrasounds that also had contrast were evaluated in this retrospective study. Patients with a focal lesion seen only after administration of contrast (OAC) were noted, as well as any follow-up imaging or pathology results. Additional variables collected included patient demographics, cirrhosis type, and laboratory values. Results: 230 unique patients were included, of which 160 had imaging or pathology follow-up. 18 of these patients had an OAC lesion, of which 17 had follow-up. Among these OACs, there was one LIRADS M lesion (1/18, 5.6 %) and one bland portal vein thrombus identified, which were both confirmed on follow-up imaging. All LIRADS 4 OAC lesions were downgraded. No additional HCC were identified on follow-up imaging or pathology of these patients. Conclusion: Addition of contrast to screening ultrasound did identify additional lesions, portal vein thrombus, and high grade malignancy. However, as the incidence of OAC lesions was low (7.8 %, 18/230) and most of the lesions were not malignant, addition of post contrast sweeps through the liver is of low value in the low to medium at-risk cirrhotic population in identifying occult HCC.

Advance Directives Change Frequently in Nursing Home Residents.

OBJECTIVES: To describe the rate, timing, and pattern of changes in advance directives (ADs) of do not resuscitate (DNR) and do not hospitalize (DNH) orders among new admissions to nursing homes (NHs). DESIGN: A retrospective cohort study. SETTING AND PARTICIPANTS: Admissions to all publicly funded NHs in Ontario, Canada, between January 1, 2013, and December 31, 2017. METHODS: Residents were followed until discharged from incident NH stay, death, or were still present at the end of study (December 31, 2019). They were categorized into 3 mutually exclusive baseline composite AD groups: Full Code, DNR Only, and DNR+DNH. We used Poisson regression models to estimate the incidence rate ratios of AD change between different AD groups and different decision makers for personal care, adjusted for baseline clinical and sociodemographic variables. RESULTS: A total of 102,541 NH residents were eligible for inclusion. Residents with at least 1 AD change accounted for 46% of Full Code, 30% of DNR Only, and 25% of DNR+DNH group. Median time to first AD change ranged between 26 and 55 weeks. For Full Code and DNR Only residents, the most frequent change was to an AD 1 level lower in aggressiveness or intervention, whereas for DNR+DNH residents the most frequent change was to DNR Only. About 16% of residents had 2 or more AD changes during their stay. After controlling for covariates, residents with a DNR-only order or DNR+DNH orders at admission and those with a surrogate decision maker were associated with lower AD change rates. CONCLUSIONS AND IMPLICATIONS: Measuring AD adherence rates that are documented only at a particular time often underestimates the dynamics of AD changes during a resident's stay and results in an inaccurate measure of the effectiveness of AD on resident care. There should be more frequent reviews of ADs as they are quite dynamic. Mandatory review after an acute change in a resident's health would ensure that ADs are current.

eXplainable Artificial Intelligence (XAI) for improving organisational regility.

Since the pandemic started, organisations have been actively seeking ways to improve their organisational agility and resilience (regility) and turn to Artificial Intelligence (AI) to gain a deeper understanding and further enhance their agility and regility. Organisations are turning to AI as a critical enabler to achieve these goals. AI empowers organisations by analysing large data sets quickly and accurately, enabling faster decision-making and building agility and resilience. This strategic use of AI gives businesses a competitive advantage and allows them to adapt to rapidly changing environments. Failure to prioritise agility and responsiveness can result in increased costs, missed opportunities, competition and reputational damage, and ultimately, loss of customers, revenue, profitability, and market share. Prioritising can be achieved by utilising eXplainable Artificial Intelligence (XAI) techniques, illuminating how AI models make decisions and making them transparent, interpretable, and understandable. Based on previous research on using AI to predict organisational agility, this study focuses on integrating XAI techniques, such as Shapley Additive Explanations (SHAP), in organisational agility and resilience. By identifying the importance of different features that affect organisational agility prediction, this study aims to demystify the decision-making processes of the prediction model using XAI. This is essential for the ethical deployment of AI, fostering trust and transparency in these systems. Recognising key features in organisational agility prediction can guide companies in determining which areas to concentrate on in order to improve their agility and resilience.

Feasibility of a wearable self-management application for patients with COPD at home: a pilot study.

BACKGROUND: Among people with COPD, smartphone and wearable technology may provide an effective method to improve care at home by supporting, encouraging, and sustaining self-management. The current study was conducted to determine if patients with COPD will use a dedicated smartphone and smartwatch app to help manage their COPD and to determine the effects on their self-management. METHODS: We developed a COPD self-management application for smartphones and smartwatches. Participants were provided with the app on a smartphone and a smartwatch, as well as a cellular data plan and followed for 6 months. We measured usage of the different smartphone app functions. For the primary outcome, we examined the change in self-management from baseline to the end of follow up. Secondary outcomes include changes in self-efficacy, quality of life, and COPD disease control. RESULTS: Thirty-four patients were enrolled and followed. Mean age was 69.8 years, and half of the participants were women. The most used functions were recording steps through the smartwatch, entering a daily symptom questionnaire, checking oxygen saturation, and performing breathing exercises. There was no significant difference in the primary outcome of change in self-management after use of the app or in overall total scores of health-related quality of life, disease control or self-efficacy. CONCLUSION: We found older patients with COPD would engage with a COPD smartphone and smartwatch application, but this did not result in improved self-management. More research is needed to determine if a smartphone and smartwatch application can improve self-management in people with COPD. TRIAL REGISTRATION: ClinicalTrials.Gov NCT03857061, First Posted February 27, 2019.

Impressions and Perceptions of a Smartphone and Smartwatch Self-Management Tool for Patients With COPD: A Qualitative Study.

BACKGROUND: Patients with chronic obstructive pulmonary disease (COPD) often do not seek care until they experience an exacerbation. Improving self-management for these patients may increase health-related quality of life and reduce hospitalizations. Patients are willing to use wearable technology for real-time data reporting and perceive mobile technology as potentially helpful in COPD management, but there are many barriers to the uptake of these technologies. OBJECTIVE: We aimed to understand patients' experiences using a wearable and mobile app and identify areas for improvement. METHODS: We conducted semi-structured interviews as part of a larger prospective cohort study wherein patients used a wearable and app for 6 months. We asked which features patients found accessible, acceptable and useful. RESULTS: We completed 26 interviews. We summarized our research findings into four main themes: (1) information, support and reassurance, (2) barriers to adoption, (3) impact on communication with health care providers, and (4) opportunities for improvement. Most patients found the feedback received through the app to be reassuring and useful. Some patients experienced technical difficulties with the app and found the wearable to be uncomfortable. CONCLUSIONS: Patients found a wearable device and mobile application to be acceptable and useful for the management of COPD. We identified barriers to adoption and opportunities for improvement to the design of our app. Further research is needed to understand what people with COPD and their healthcare providers want and will use in a mobile app and wearable for COPD management.

Radially adjustable stent retriever for mechanical thrombectomy in acute ischemic stroke: Improved first-pass effect with rapid-inflation deflation technique.

INTRODUCTION: Evidence for improved first-pass effect with the novel radially adjustable radio-opaque stent retriever Tigertriever is lacking. OBJECTIVE: To compare improvement in first pass success with Tigertriever using two different techniques-rapid inflation deflation (RID) and suction thrombectomy (ST). METHODS: Retrospective analysis of patients with acute ischemic stroke who underwent mechanical thrombectomy with Tigertriever at a single comprehensive stroke center. RESULTS: Thirty patients were included. Mean age was 72.8 years. Twelve patients (48%) experienced successful first passes with Tigertriever. Successful revascularization (modified thrombolysis in cerebral infarction (mTICI) 2b/3) was achieved in all (100%) patients who received RID or ST technique for thrombectomy. Good clinical outcome (modified Rankin score = 0-2) was noted in 40% (n = 10). Total mortality in the cohort was 8% (n = 2). RID and ST groups comprised of 10 and 15 patients, respectively. Five patients underwent MT with Tigertriever as a rescue device. RID VS ST: No difference was noted in mean age (p = 0.27), gender (p = 0.29), location of occlusion (p = 0.46), and device used for first pass (p = 0.57). A 70% first-pass success rate in RID group and 37.5% in ST group was noticed (p = 0.06). Mean time from groin puncture to reperfusion (TICI 2b//3) was statistically similar (p = 0.29, RID: 19.9 min vs ST: 25 min). Both groups noted a 100% complete recanalization rate. The rate of mortality between the two groups were not statistically different (p = 0.46). CONCLUSION: The preliminary first-pass success rates of RID technique with Tigertriever compared to ST technique, are encouraging. Longitudinal studies with longer follow up are needed to elucidate the smaller learning curve with this device.

Predictors of Readmission, for Patients with Chronic Obstructive Pulmonary Disease (COPD) - A Systematic Review.

Introduction: Chronic obstructive pulmonary disease (COPD) is the third-leading cause of death globally and is responsible for over 3 million deaths annually. One of the factors contributing to the significant healthcare burden for these patients is readmission. The aim of this review is to describe significant predictors and prediction scores for all-cause and COPD-related readmission among patients with COPD. Methods: A search was conducted in Ovid MEDLINE, Ovid Embase, Cochrane Database of Systematic Reviews, and Cochrane Central Register of Controlled Trials, from database inception to June 7, 2022. Studies were included if they reported on patients at least 40 years old with COPD, readmission data within 1 year, and predictors of readmission. Study quality was assessed. Significant predictors of readmission and the degree of significance, as noted by the p-value, were extracted for each study. This review was registered on PROSPERO (CRD42022337035). Results: In total, 242 articles reporting on 16,471,096 patients were included. There was a low risk of bias across the literature. Of these, 153 studies were observational, reporting on predictors; 57 studies were observational studies reporting on interventions; and 32 were randomized controlled trials of interventions. Sixty-four significant predictors for all-cause readmission and 23 for COPD-related readmission were reported across the literature. Significant predictors included 1) pre-admission patient characteristics, such as male sex, prior hospitalization, poor performance status, number and type of comorbidities, and use of long-term oxygen; 2) hospitalization details, such as length of stay, use of corticosteroids, and use of ventilatory support; 3) results of investigations, including anemia, lower FEV1, and higher eosinophil count; and 4) discharge characteristics, including use of home oxygen and discharge to long-term care or a skilled nursing facility. Conclusion: The findings from this review may enable better predictive modeling and can be used by clinicians to better inform their clinical gestalt of readmission risk.

Development of Multivariable Prediction Models for the Identification of Patients Admitted to Hospital with an Exacerbation of COPD and the Prediction of Risk of Readmission: A Retrospective Cohort Study using Electronic Medical Record Data.

BACKGROUND: Approximately 20% of patients who are discharged from hospital for an acute exacerbation of COPD (AECOPD) are readmitted within 30 days. To reduce this, it is important both to identify all individuals admitted with AECOPD and to predict those who are at higher risk for readmission. OBJECTIVES: To develop two clinical prediction models using data available in electronic medical records: 1) identifying patients admitted with AECOPD and 2) predicting 30-day readmission in patients discharged after AECOPD. METHODS: Two datasets were created using all admissions to General Internal Medicine from 2012 to 2018 at two hospitals: one cohort to identify AECOPD and a second cohort to predict 30-day readmissions. We fit and internally validated models with four algorithms. RESULTS: Of the 64,609 admissions, 3,620 (5.6%) were diagnosed with an AECOPD. Of those discharged, 518 (15.4%) had a readmission to hospital within 30 days. For identification of patients with a diagnosis of an AECOPD, the top-performing models were LASSO and a four-variable regression model that consisted of specific medications ordered within the first 72 hours of admission. For 30-day readmission prediction, a two-variable regression model was the top performing model consisting of number of COPD admissions in the previous year and the number of non-COPD admissions in the previous year. CONCLUSION: We generated clinical prediction models to identify AECOPDs during hospitalization and to predict 30-day readmissions after an acute exacerbation from a dataset derived from available EMR data. Further work is needed to improve and externally validate these models.