| 1. |
- El-Merhi, Ali, et al.
(författare)
-
Towards Trustworthy Cross-patient Model Development
- 2021
-
Ingår i: ArXiv:2112.1044.
-
Konferensbidrag (refereegranskat)abstract
- Machine learning is used in medicine to support physicians in examination, diagnosis, and predicting outcomes. One of the most dynamic area is the usage of patient generated health data from intensive care units. The goal of this paper is to demonstrate how we advance cross-patient ML model development by combining the patient’s demographics data with their physiological data. We used a population of patients undergoing Carotid Enderarterectomy (CEA), where we studied differences in model performance and explainability when trained for all patients and one patient at a time. The results show that patients’ demographics has a large impact on the performance and explainability and thus trustworthiness. We conclude that we can increase trust in ML models in a cross-patient context, by careful selection of models and patients based on their demographics and the surgical procedure.
|
|
| 2. |
- El-Merhi, Ali, et al.
(författare)
-
Trusting Machine Learning Results from Medical Procedures in the Operating Room
- 2022
-
Ingår i: ArXiv 2022.
-
Konferensbidrag (refereegranskat)abstract
- Machine learning can be used to analyse physiological data for several purposes. Detection of cerebral ischemia is an achievement that would have high impact on patient care. We attempted to study if collection of continous physiological data from non-invasive monitors, and analysis with machine learning could detect cerebral ischemia in tho different setting, during surgery for carotid endarterectomy and during endovascular thrombectomy in acute stroke. We compare the results from the two different group and one patient from each group in details. While results from CEA-patients are consistent, those from thrombectomy patients are not and frequently contain extreme values such as 1.0 in accuracy. We conlcude that this is a result of short duration of the procedure and abundance of data with bad quality resulting in small data sets. These results can therefore not be trusted.
|
|
| 3. |
- Erlandsson, Karin, 1973, et al.
(författare)
-
Positive end-expiratory pressure optimization using electric impedance tomography in morbidly obese patients during laparoscopic gastric bypass surgery.
- 2006
-
Ingår i: Acta anaesthesiologica Scandinavica. - : Wiley. - 0001-5172 .- 1399-6576. ; 50:7, s. 833-9
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: Morbidly obese patients have an increased risk for peri-operative lung complications and develop a decrease in functional residual capacity (FRC). Electric impedance tomography (EIT) can be used for continuous, fast-response measurement of lung volume changes. This method was used to optimize positive end-expiratory pressure (PEEP) to maintain FRC. METHODS: Fifteen patients with a body mass index of 49 +/- 8 kg/m(2) were studied during anaesthesia for laparoscopic gastric bypass surgery. Before induction, 16 electrodes were placed around the thorax to monitor ventilation-induced impedance changes. Calibration of the electric impedance tomograph against lung volume changes was made by increasing the tidal volume in steps of 200 ml. PEEP was titrated stepwise to maintain a horizontal baseline of the EIT curve, corresponding to a stable FRC. Absolute FRC was measured with a nitrogen wash-out/wash-in technique. Cardiac output was measured with an oesophageal Doppler method. Volume expanders, 1 +/- 0.5 l, were given to prevent PEEP-induced haemodynamic impairment. RESULTS: Impedance changes closely followed tidal volume changes (R(2) > 0.95). The optimal PEEP level was 15 +/- 1 cmH(2)O, and FRC at this PEEP level was 1706 +/- 447 ml before and 2210 +/- 540 ml after surgery (P < 0.01). The cardiac index increased significantly from 2.6 +/- 0.5 before to 3.1 +/- 0.8 l/min/m(2) after surgery, and the alveolar dead space decreased. P(a)O2/F(i)O2, shunt and compliance remained unchanged. CONCLUSION: EIT enables rapid assessment of lung volume changes in morbidly obese patients, and optimization of PEEP. High PEEP levels need to be used to maintain a normal FRC and to minimize shunt. Volume loading prevents circulatory depression in spite of a high PEEP level.
|
|
| 4. |
- Lindgren, Sophie, 1971, et al.
(författare)
-
Regional lung derecruitment after endotracheal suction during volume- or pressure-controlled ventilation: a study using electric impedance tomography
- 2007
-
Ingår i: Intensive care medicine. - 0342-4642. ; 33:1, s. 172-80
-
Tidskriftsartikel (refereegranskat)abstract
- OBJECTIVE: To assess lung volume and compliance changes during open- and closed-system suctioning using electric impedance tomography (EIT) during volume- or pressure-controlled ventilation. DESIGN AND SETTING: Experimental study in a university research laboratory. SUBJECTS: Nine bronchoalveolar saline-lavaged pigs. INTERVENTIONS: Open and closed suctioning using a 14-F catheter in volume- or pressure-controlled ventilation at tidal volume 10 ml/kg, respiratory rate 20 breaths/min, and positive end-expiratory pressure 10 cmH2O. MEASUREMENTS AND RESULTS: Lung volume was monitored by EIT and a modified N2 washout/-in technique. Airway pressure was measured via a pressure line in the endotracheal tube. In four ventral-to-dorsal regions of interest regional ventilation and compliance were calculated at baseline and 30 s and 1, 2, and 10 min after suctioning. Blood gases were followed. At disconnection functional residual capacity (FRC) decreased by 58+/-24% of baseline and by a further 22+/-10% during open suctioning. Arterial oxygen tension decreased to 59+/-14% of baseline value 1 min after open suctioning. Regional compliance deteriorated most in the dorsal parts of the lung. Restitution of lung volume and compliance was significantly slower during pressure-controlled than volume-controlled ventilation. CONCLUSIONS: EIT can be used to monitor rapid lung volume changes. The two dorsal regions of the lavaged lungs are most affected by disconnection and suctioning with marked decreases in compliance. Volume-controlled ventilation can be used to rapidly restitute lung aeration and oxygenation after lung collapse induced by open suctioning.
|
|
| 5. |
- Lowhagen, Karin, 1973, et al.
(författare)
-
A new non-radiological method to assess potential lung recruitability: a pilot study in ALI patients.
- 2011
-
Ingår i: Acta anaesthesiologica Scandinavica. - : Wiley. - 1399-6576 .- 0001-5172. ; 55:2, s. 165-74
-
Tidskriftsartikel (refereegranskat)abstract
- Potentially recruitable lung has been assessed previously in patients with acute lung injury (ALI) by computed tomography. A large variability in lung recruitability was observed between patients. In this study, we assess whether a new non-radiological bedside technique could determine potentially recruitable lung volume (PRLV) in ALI patients.
|
|
| 6. |
- Lowhagen, Karin, 1973, et al.
(författare)
-
Prolonged moderate pressure recruitment manoeuvre results in lower optimal positive end-expiratory pressure and plateau pressure.
- 2011
-
Ingår i: Acta anaesthesiologica Scandinavica. - : Wiley. - 1399-6576 .- 0001-5172. ; 55:2, s. 175-84
-
Tidskriftsartikel (refereegranskat)abstract
- In acute lung injury (ALI)/acute respiratory distress syndrome (ARDS), recruitment manoeuvres (RMs) are used frequently. In pigs with induced ALI, superior effects have been found using a slow moderate-pressure recruitment manoeuvre (SLRM) compared with a vital capacity recruitment manoeuvre (VICM). We hypothesized that the positive recruitment effects of SLRM could also be achieved in ALI/ARDS patients. Our primary research question was whether the same compliance could be obtained using lower RM pressure and subsequent positive end-expiratory pressure (PEEP). Secondly, optimal PEEP levels following the RMs were compared, and the use of volume-dependent compliance (VDC) to identify successful lung recruitment and optimal PEEP was evaluated.
|
|
| 7. |
- Norberg, Erik, et al.
(författare)
-
Impact of Acute Cardiac Complications After Subarachnoid Hemorrhage on Long-Term Mortality and Cardiovascular Events.
- 2018
-
Ingår i: Neurocritical care. - : Springer Science and Business Media LLC. - 1556-0961 .- 1541-6933. ; 29:3, s. 404-412
-
Tidskriftsartikel (refereegranskat)abstract
- Cardiac complications frequently occur after subarachnoid hemorrhage (SAH) and are associated with an increased risk of neurological complications and poor outcomes. The aim of this study was to evaluate the impact of acute cardiac complications after SAH on long-term mortality and cardiovascular events.All patients admitted to our Neuro intensive care unit with verified SAH from January 2010 to April 2015, and electrocardiogram, echocardiogram, and troponin T or NTproBNP data obtained within 72h of admission were included in the study. Mortality data were obtained from the Swedish population register. Data regarding cause of death and hospitalization for cardiovascular events were obtained from the Swedish Board of Health and Welfare.A total of 455 patients were included in the study analysis. There were 102 deaths during the study period. Cardiac troponin release (HR 1.08, CI 1.02-1.15 per 100ng/l, p=0.019), NTproBNP (HR 1.05, CI 1.01-1.09 per 1000ng/l, p=0.018), and ST-T abnormalities (HR 1.53, CI 1.02-2.29, p=0.040) were independently associated with an increased risk of death. However, these associations were significant only during the first 3months after the hemorrhage. Cardiac events were observed in 25 patients, and cerebrovascular events were observed in 62 patients during the study period. ST-T abnormalities were independently associated with an increased risk of cardiac events (HR 5.52, CI 2.07-14.7, p<0.001), and stress cardiomyopathy was independently associated with an increased risk of cerebrovascular events (HR 3.65, CI 1.55-8.58, p=0.003).Cardiac complications after SAH are associated with an increased risk of short-term death. Patients with electrocardiogram abnormalities and stress cardiomyopathy need appropriate follow-up for the identification of cardiac disease or risk factors for cardiovascular disease.
|
|
| 8. |
- Nordenskjöld Syrous, Alma, 1979, et al.
(författare)
-
End-of-life decision-making in critically ill old patients with and without coronavirus disease 2019.
- 2023
-
Ingår i: Acta anaesthesiologica Scandinavica. - 0001-5172 .- 1399-6576. ; 68:1, s. 63-70
-
Tidskriftsartikel (refereegranskat)abstract
- There are few studies on the differences in end-of-life decisions making in critically ill patients with and without coronavirus disease 2019 (COVID-19). This study aimed to investigate the independent factors that predicted the decision to withdraw or withhold life-sustaining treatments (LST) in critically ill patients and if these decisions were based on different variables for critically ill patients with COVID-19 compared to those for critically ill patients with other diagnoses in a Swedish intensive care unit.This observational pilot study was performed at Sahlgrenska University Hospital, Gothenburg, Sweden. Patients ≥65years were included from 1 March 2020 to 30 April 2021. The association between a decision to limit LST and a priori selected variables including sex, age, Simplified Acute Physiology Score 3 (SAPS 3), Clinical Frailty Scale ≥4, Charlson Comorbidity Index, Body Mass Index, living at home, invasive and non-invasive mechanical ventilation was assessed using a univariate and multivariable logistic regression model and presented as odds ratio with corresponding 95% confidence intervals.There were 394 patients included in this study, 131 in the non-COVID-19 group and 263 in the COVID-19 group. For the non-COVID-19 cohort, the univariate analysis demonstrated that age and SAPS 3 were significantly associated with the decision to withdraw or withhold life-sustaining treatments, and this association remained in the multivariable analysis, with odds ratios of 1.10 (1.03-1.19) p=.009 and 1.06 (1.03-1.10) p<.001, respectively. For the COVID-19 cohort, the univariate analysis indicated that age, SAPS 3, and Charlson comorbidity index were significantly associated with the decision to withdraw or withhold life-sustaining treatments. However, in multivariable analysis, only the Charlson comorbidity index remained independently associated with the decision to withdraw or withhold life-sustaining treatments, with an odds ratio of 1.26 (1.07-1.49), p=.006.Decisions to withdraw or withhold life-sustaining treatments were based on other variables for the critically ill COVID-19 cohort compared to those for the critically ill non-COVID-19 cohort. Further studies are warranted to forge a common path for ethical end-of-life decision-making in critically ill patients.
|
|
| 9. |
- Odenstedt, Helena, 1968, et al.
(författare)
-
Acute hemodynamic changes during lung recruitment in lavage and endotoxin-induced ALI.
- 2005
-
Ingår i: Intensive care medicine. - : Springer Science and Business Media LLC. - 0342-4642 .- 1432-1238. ; 31:1, s. 112-20
-
Tidskriftsartikel (refereegranskat)abstract
- OBJECTIVE: To assess acute cardiorespiratory effects of recruitment manoeuvres in experimental acute lung injury. DESIGN: Experimental study in animal models of acute lung injury. SETTING: Experimental laboratory at a University Medical Centre. ANIMALS: Ten pigs with bronchoalveolar lavage and eight pigs with endotoxin-induced ALI. INTERVENTIONS: Two kinds of recruitment manoeuvres during 1 min; a) vital capacity manoeuvres (ViCM) consisting in a sustained inflation at 30 cmH(2)O and 40 cmH(2)O; b) manoeuvres obtained during ongoing pressure-controlled ventilation (PCRM) with peak airway pressure 30 cmH(2)O, positive end-expiratory pressure (PEEP) 15 and peak airway pressure 40, PEEP 20. Recruitment manoeuvres were repeated after volume expansion (dextran 8 ml/kg). Oxygenation, mean arterial, and pulmonary artery pressures, aortic, mesenteric, and renal blood flow were monitored. MEASUREMENTS AND RESULTS: Lower pressure recruitment manoeuvres (ViCM30 and PCRM30/15) did not significantly improve oxygenation. With ViCM and PCRM at peak airway pressure 40 cmH(2)O, PaO(2) increased to similar levels in both lavage and endotoxin groups. Aortic blood flow was reduced from baseline during PCRM40/20 and ViCM40 by 57+/-3% and 61+/-6% in the lavage group and by 57+/-8% and 82+/-7% (P<0.05 vs PCRM40/20) in endotoxin group. The decrease in blood pressure was less pronounced. Prior volume expansion attenuated circulatory impairment. After cessation of recruitment hemodynamic parameters were restored within 3 min. CONCLUSION: Effective recruitment resulted in systemic hypotension, pulmonary hypertension, and decrease in aortic blood flow especially in endotoxinemic animals. Circulatory depression may be attenuated using recruitment manoeuvres during ongoing pressure-controlled ventilation and by prior volume expansion.
|
|
| 10. |
- Odenstedt, Helena, 1968
(författare)
-
Minimally invasive cardiorespiratory monitoring and lung recruitment manoeuvres. Studies in animals and humans
- 2005
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Background: The maintenance of adequate oxygen delivery is essential to preserve organ function. The circulatory and respiratory systems interact closely to accomplish this. Optimising the function of these systems is of fundamental importance in critically ill patients. This requires adequate monitoring techniques. Availability of non-invasive, bedside, continuous techniques for monitoring the effects of circulatory and respiratory interventions are limited. The aim of this thesis was to evaluate alternative techniques to monitor central hemodynamics and lung volume changes and to use these techniques to study the effects of lung recruitment manoeuvres.Methods: A transoesophageal echo-Doppler technique measuring descending aortic blood flow and a partial CO2 rebreathing technique for cardiac output determination were evaluated in unstable hemodynamic conditions and compared with cardiac output measured by thermodilution using pulmonary artery catheter. Electric impedance tomography was used to monitor global and regional lung volume changes and ventilation distribution during different recruitment manoeuvres in experimental acute lung injury (ALI).Results: The transoesophageal echo-Doppler rapidly and accurately followed marked changes in central blood flow during liver transplantation and experimental cardiac tamponade. Partial CO2 rebreathing technique showed good agreement with the reference technique, despite several possible sources of error. By continuously monitoring aortic blood flow, pronounced circulatory depression was detected during lung recruitment using rapid high airway pressure manoeuvres especially in endotoxin- induced ALI. Monitoring blood pressure underestimated the true circulatory impairment. These negative circulatory effects could be attenuated by prior volume expansion or by using a slow, lower pressure recruitment manoeuvre. This recruitment manoeuvre increased lung volume, compliance and improved gas exchange equally well as high-pressure manoeuvres. Electric impedance tomography revealed the changes in ventilation distribution from non-dependent to dependent areas induced by lung recruitment.Conclusions: The investigated minimally invasive monitoring techniques are useful for monitoring changes in central hemodynamics in critically ill patients. Adequate monitoring techniques are required to detect rapid changes in blood flow like those seen during recruitment manoeuvres. Lung recruitment can be achieved by a slow, lower pressure manoeuvre with less circulatory side-effects. Electric impedance tomography is a promising technique for bedside monitoring of global and regional lung volume changes.
|
|
