scholarly journals Does the Estimation of the Propensity Score by Machine Learning Improve Matching Estimation? The Case of Germany's Programmes for Long Term Unemployed

2019 ◽  
Author(s):  
Daniel Goller ◽  
Michael Lechner ◽  
Andreas Moczall ◽  
Joachim Wolff
Keyword(s):  
Machine Learning ◽  
Propensity Score ◽  
Matching Estimation

scholarly journals Global soil moisture data derived through machine learning trained with in-situ measurements

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Sungmin O. ◽  
Rene Orth
Keyword(s):  
Machine Learning ◽  
Soil Moisture ◽  
Large Scale ◽  
Short Term Memory ◽  
Temporal Dynamics ◽  
Soil Moisture Data ◽  
Wide Range ◽  
Global Soil

AbstractWhile soil moisture information is essential for a wide range of hydrologic and climate applications, spatially-continuous soil moisture data is only available from satellite observations or model simulations. Here we present a global, long-term dataset of soil moisture derived through machine learning trained with in-situ measurements, SoMo.ml. We train a Long Short-Term Memory (LSTM) model to extrapolate daily soil moisture dynamics in space and in time, based on in-situ data collected from more than 1,000 stations across the globe. SoMo.ml provides multi-layer soil moisture data (0–10 cm, 10–30 cm, and 30–50 cm) at 0.25° spatial and daily temporal resolution over the period 2000–2019. The performance of the resulting dataset is evaluated through cross validation and inter-comparison with existing soil moisture datasets. SoMo.ml performs especially well in terms of temporal dynamics, making it particularly useful for applications requiring time-varying soil moisture, such as anomaly detection and memory analyses. SoMo.ml complements the existing suite of modelled and satellite-based datasets given its distinct derivation, to support large-scale hydrological, meteorological, and ecological analyses.

scholarly journals Female Sex and Outcomes after Endovascular Aneurysm Repair for Abdominal Aortic Aneurysm: A Propensity Score Matched Cohort Analysis

2021 ◽  
Vol 10 (1) ◽  
pp. 162
Author(s):  
Christian-Alexander Behrendt ◽  
Thea Kreutzburg ◽  
Jenny Kuchenbecker ◽  
Giuseppe Panuccio ◽  
Mark Dankhoff ◽  
...  
Keyword(s):  
Abdominal Aortic Aneurysm ◽  
Aortic Aneurysm ◽  
Propensity Score ◽  
Acute Limb Ischemia ◽  
Insurance Claims ◽  
Long Term Outcomes ◽  
Term Survival ◽  
Abdominal Aortic ◽  
Significant Difference

Objective: Previous studies have showed a potential disadvantage of female patients who underwent abdominal aortic aneurysm (AAA) repair. The current study aims to determine sex-specific perioperative and long-term outcomes using propensity score matched unselected nationwide health insurance claims data. Methods: Insurance claims from a large German fund were used, covering around 8% of the insured German population. Patients who underwent endovascular aortic repair (EVAR) for intact AAA from 1 January 2011 to 30 April 2017 were included in the cohort. A 1:2 female to male propensity score matching was applied to adjust for confounding variables. Perioperative and long-term outcomes after 5 years were determined using matching and regression methods. Results: Among a total of 3736 patients (19.3% females, mean 75 years) undergoing EVAR for intact AAA, we identified 1863 matched patients. Before matching, females were more likely to be previously diagnosed with hypothyroidism, electrolyte disorders, rheumatoid disorders, and depression, while males were more often diabetics. In the matched sample, 23.4% of the females and 25.8% of the males died during a median follow-up of 776 and 792 days, respectively. Perioperatively, females were more likely to exhibit acute limb ischemia (5.3% vs. 3.2%, p = 0.031) and major bleeding (22.0% vs. 15.9%, p = 0.001) before they were discharged to rehabilitation (5.5% vs. 1.5%, p < 0.001) when compared to males. No statistically significant difference in perioperative (odds ratio 1.12, 95% CI 0.54–2.16) or long-term mortality (hazard ratio 0.91, 95% CI 0.76–1.08) was observed between sexes. This was also true regarding aortic reintervention rates after 1 year (2.0% vs. 2.9%) and 5 years (10.9% vs. 8.1%). Conclusion: The current retrospective matched analysis of insurance claims revealed high early access-related morbidity in females when compared to their male counterparts. Short-term or long-term survival and reintervention outcomes were similar between sexes.

scholarly journals Outcomes of Incisional Hernia Repair Surgery After Multiple Re-recurrences: A Propensity Score Matched Analysis

2021 ◽  
Author(s):  
Dimitri Sneiders ◽  
Gijs H. J. de Smet ◽  
Floris den Hartog ◽  
Yagmur Yurtkap ◽  
Anand G. Menon ◽  
...  
Keyword(s):  
Propensity Score ◽  
Incisional Hernia ◽  
Incisional Hernia Repair ◽  
Recurrent Hernia ◽  
Control Group ◽  
Baseline Characteristics ◽  
Biological Situation ◽  
Similar Risk

Abstract Background Patients with a re-recurrent hernia may account for up to 20% of all incisional hernia (IH) patients. IH repair in this population may be complex due to an altered anatomical and biological situation as a result of previous procedures and outcomes of IH repair in this population have not been thoroughly assessed. This study aims to assess outcomes of IH repair by dedicated hernia surgeons in patients who have already had two or more re-recurrences. Methods A propensity score matched analysis was performed using a registry-based, prospective cohort. Patients who underwent IH repair after ≥ 2 re-recurrences operated between 2011 and 2018 and who fulfilled 1 year follow-up visit were included. Patients with similar follow-up who underwent primary IH repair were propensity score matched (1:3) and served as control group. Patient baseline characteristics, surgical and functional outcomes were analyzed and compared between both groups. Results Seventy-three patients operated on after ≥ 2 IH re-recurrences were matched to 219 patients undergoing primary IH repair. After propensity score matching, no significant differences in patient baseline characteristics were present between groups. The incidence of re-recurrence was similar between groups (≥ 2 re-recurrences: 25% versus control 24%, p = 0.811). The incidence of complications, as well as long-term pain, was similar between both groups. Conclusion IH repair in patients who have experienced multiple re-recurrences results in outcomes comparable to patients operated for a primary IH with a similar risk profile. Further surgery in patients who have already experienced multiple hernia re-recurrences is justifiable when performed by a dedicated hernia surgeon.

scholarly journals Role of diuretics on long-term mortality may differ in volume status in patients with acute myocardial infarction

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
T Kawai ◽  
D Nakatani ◽  
T Yamada ◽  
T Watanabe ◽  
T Morita ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
Propensity Score ◽  
Plasma Volume ◽  
Cox Regression ◽  
Volume Status ◽  
Increased Risk ◽  
Long Term Mortality

Abstract Background Diuretics has been reported to have a potential for an activation of the renin-angiotensin-aldosterone system and the sympathetic nervous system, leading to a possibility of poor clinical outcome in patients with cardiovascular disease. However, few data are available on clinical impact of diuretics on long-term outcome in patients with acute myocardial infarction (AMI) based on plasma volume status. Methods To address the issue, a total of 3,416 survived patients with AMI who were registered to a large database of the Osaka Acute Coronary Insufficiency Study (OACIS) were studied. Plasma volume status was assessed with the estimated plasma volume status (ePVS) that was calculated at discharge as follows: actual PV = (1 − hematocrit) × [a + (b × body weight)] (a=1530 in males and a=864 in females, b=41.0 in males and b=47.9 in females); ideal PV = c × body weight (c=39 in males and c=40 in females), and ePVS = [(actual PV − ideal PV)/ideal PV] × 100 (%). Multivariable Cox regression analysis and propensity score matching were performed to account for imbalances in covariates. The endpoint was all-cause of death (ACD) within 5 years. Results During a median follow-up period of 855±656 days, 193 patients had ACD. In whole population, there was no significant difference in long-term mortality risk between patients with and without diuretics in both multivariate cox regression model and propensity score matching population. When patients were divided into 2 groups according to ePVS with a median value of 4.2%, 46 and 147 patients had ACD in groups with low ePVS and high ePVS, respectively. Multivariate Cox analysis showed that use of diuretics was independently associated with an increased risk of ACD in low ePVS group, (HR: 2.63, 95% confidence interval [CI]: 1.22–5.63, p=0.01), but not in high ePVS group (HR: 0.70, 95% CI: 0.44–1.10, p=0.12). These observations were consistent in the propensity-score matched cohorts; the 5-year mortality rate was significantly higher in patients with diuretics than those without among low ePVS group (4.7% vs 1.7%, p=0.041), but not among high ePVS group (8.0% vs 10.3%, p=0.247). Conclusion Prescription of diuretics at discharge was associated with increased risk of 5-year mortality in patients with AMI without PV expansion, but not with PV expansion. The role of diuretics on long-term mortality may differ in plasma volume status. Therefore, prescription of diuretics after AMI may be considered based on plasma volume status. Funding Acknowledgement Type of funding source: None

Sign in / Sign up

Export Citation Format

Share Document