| 1. |
|
|
| 2. |
|
|
| 3. |
|
|
| 4. |
- Kuo, NIH, et al.
(author)
-
The Health Gym: synthetic health-related datasets for the development of reinforcement learning algorithms
- 2022
-
In: Scientific data. - : Springer Science and Business Media LLC. - 2052-4463. ; 9:1, s. 693-
-
Journal article (peer-reviewed)abstract
- In recent years, the machine learning research community has benefited tremendously from the availability of openly accessible benchmark datasets. Clinical data are usually not openly available due to their confidential nature. This has hampered the development of reproducible and generalisable machine learning applications in health care. Here we introduce the Health Gym - a growing collection of highly realistic synthetic medical datasets that can be freely accessed to prototype, evaluate, and compare machine learning algorithms, with a specific focus on reinforcement learning. The three synthetic datasets described in this paper present patient cohorts with acute hypotension and sepsis in the intensive care unit, and people with human immunodeficiency virus (HIV) receiving antiretroviral therapy. The datasets were created using a novel generative adversarial network (GAN). The distributions of variables, and correlations between variables and trends in variables over time in the synthetic datasets mirror those in the real datasets. Furthermore, the risk of sensitive information disclosure associated with the public distribution of the synthetic datasets is estimated to be very low.
|
|
| 5. |
- O'Brien, Z, et al.
(author)
-
Higher versus Lower Continuous Renal Replacement Therapy Intensity in Critically ill Patients with Liver Dysfunction
- 2018
-
In: Blood purification. - : S. Karger AG. - 1421-9735 .- 0253-5068. ; 45:1-3, s. 36-43
-
Journal article (peer-reviewed)abstract
- <b><i>Aims:</i></b> To study the association between higher versus lower continuous renal replacement therapy (CRRT) intensity and mortality in critically ill patients with combined acute kidney injury and liver dysfunction. <b><i>Methods:</i></b> Post-hoc analysis of patients with liver dysfunction (Sequential Organ Failure Assessment liver score ≥2 or diagnosis of liver failure/transplant) included in the Randomized Evaluation of Normal versus Augmented Level renal replacement therapy (RENAL) trial. <b><i>Results:</i></b> Of 444 patients, 210 (47.3%) were randomized to higher intensity (effluent flow 40 mL/kg/h) and 234 (52.7%) to lower intensity (effluent flow 25 mL/kg/h) therapy. Overall, 79 and 86% of prescribed effluent flow was delivered in the higher-intensity and lower-intensity groups, respectively (<i>p</i> < 0.001). In total, 113 (54.1%) and 120 (51.3%) patients died in each group. On multivariable Cox regression analysis, we found no independent association between higher CRRT intensity and mortality (HR 0.93, 95% CI 0.70-1.24; <i>p</i> = 0.642). <b><i>Conclusions:</i></b> In RENAL patients with liver dysfunction, higher CRRT intensity was not associated with reduced mortality.
|
|
| 6. |
|
|
| 7. |
- Ranieri, V. Marco, et al.
(author)
-
Drotrecogin Alfa (Activated) in Adults with Septic Shock
- 2012
-
In: New England Journal of Medicine. - 0028-4793 .- 1533-4406. ; 366:22, s. 2055-2064
-
Journal article (peer-reviewed)abstract
- BACKGROUND There have been conflicting reports on the efficacy of recombinant human activated protein C, or drotrecogin alfa (activated) (DrotAA), for the treatment of patients with septic shock. METHODS In this randomized, double-blind, placebo-controlled, multicenter trial, we assigned 1697 patients with infection, systemic inflammation, and shock who were receiving fluids and vasopressors above a threshold dose for 4 hours to receive either DrotAA (at a dose of 24 mu g per kilogram of body weight per hour) or placebo for 96 hours. The primary outcome was death from any cause 28 days after randomization. RESULTS At 28 days, 223 of 846 patients (26.4%) in the DrotAA group and 202 of 834 (24.2%) in the placebo group had died (relative risk in the DrotAA group, 1.09; 95% confidence interval [CI], 0.92 to 1.28; P = 0.31). At 90 days, 287 of 842 patients (34.1%) in the DrotAA group and 269 of 822 (32.7%) in the placebo group had died (relative risk, 1.04; 95% CI, 0.90 to 1.19; P = 0.56). Among patients with severe protein C deficiency at baseline, 98 of 342 (28.7%) in the DrotAA group had died at 28 days, as compared with 102 of 331 (30.8%) in the placebo group (risk ratio, 0.93; 95% CI, 0.74 to 1.17; P = 0.54). Similarly, rates of death at 28 and 90 days were not significantly different in other predefined subgroups, including patients at increased risk for death. Serious bleeding during the treatment period occurred in 10 patients in the DrotAA group and 8 in the placebo group (P = 0.81). CONCLUSIONS DrotAA did not significantly reduce mortality at 28 or 90 days, as compared with placebo, in patients with septic shock.
|
|
| 8. |
|
|
| 9. |
|
|
| 10. |
|
|
