| 1. |
- Agvall, B., et al.
(författare)
-
Characteristics, management and outcomes in patients with CKD in a healthcare region in Sweden: a population-based, observational study
- 2023
-
Ingår i: Bmj Open. - London : BMJ Publishing Group Ltd. - 2044-6055. ; 13:7
-
Tidskriftsartikel (refereegranskat)abstract
- ObjectivesTo describe chronic kidney disease (CKD) regarding treatment rates, comorbidities, usage of CKD International Classification of Diseases (ICD) diagnosis, mortality, hospitalisation, evaluate healthcare utilisation and screening for CKD in relation to new nationwide CKD guidelines. DesignPopulation-based observational study. SettingHealthcare registry data of patients in Southwest Sweden. ParticipantsA total cohort of 65 959 individuals aged >18 years of which 20 488 met the criteria for CKD (cohort 1) and 45 470 at risk of CKD (cohort 2). Primary and secondary outcome measuresData were analysed with regards to prevalence, screening rates of blood pressure, glucose, estimated glomerular filtration rate (eGFR), Urinary-albumin-creatinine ratio (UACR) and usage of ICD-codes for CKD. Mortality and hospitalisation were analysed with logistic regression models. ResultsOf the CKD cohort, 18% had CKD ICD-diagnosis and were followed annually for blood pressure (79%), glucose testing (76%), eGFR (65%), UACR (24%). UACR follow-up was two times as common in hypertensive and cardiovascular versus diabetes patients with CKD with a similar pattern in those at risk of CKD. Statin and renin-angiotensin-aldosterone inhibitor appeared in 34% and 43%, respectively. Mortality OR at CKD stage 5 was 1.23 (CI 0.68 to 0.87), diabetes 1.20 (CI 1.04 to 1.38), hypertension 1.63 (CI 1.42 to 1.88), atherosclerotic cardiovascular disease (ASCVD) 1.84 (CI 1.62 to 2.09) associated with highest mortality risk. Hospitalisation OR in CKD stage 5 was 1.96 (CI 1.40 to 2.76), diabetes 1.15 (CI 1.06 to 1.25), hypertension 1.23 (CI 1.13 to 1.33) and ASCVD 1.52 (CI 1.41 to 1.64). ConclusionsThe gap between patients with CKD by definition versus those diagnosed as such was large. Compared with recommendations patients with CKD have suboptimal follow-up and treatment with renin-angiotensin-aldosterone system inhibitor and statins. Hypertension, diabetes and ASCVD were associated with increased mortality and hospitalisation. Improved screening and diagnosis of CKD, identification and management of risk factors and kidney protective treatment could affect clinical and economic outcomes.
|
|
| 2. |
- Ashfaq, Awais, 1990-, et al.
(författare)
-
DEED : DEep Evidential Doctor
- 2023
-
Ingår i: Artificial Intelligence. - Amsterdam : Elsevier. - 0004-3702 .- 1872-7921. ; 325
-
Tidskriftsartikel (refereegranskat)abstract
- As Deep Neural Networks (DNN) make their way into safety-critical decision processes, it becomes imperative to have robust and reliable uncertainty estimates for their predictions for both in-distribution and out-of-distribution (OOD) examples. This is particularly important in real-life high-risk settings such as healthcare, where OOD examples (e.g., patients with previously unseen or rare labels, i.e., diagnoses) are frequent, and an incorrect clinical decision might put human life in danger, in addition to having severe ethical and financial costs. While evidential uncertainty estimates for deep learning have been studied for multi-class problems, research in multi-label settings remains untapped. In this paper, we propose a DEep Evidential Doctor (DEED), which is a novel deterministic approach to estimate multi-label targets along with uncertainty. We achieve this by placing evidential priors over the original likelihood functions and directly estimating the parameters of the evidential distribution using a novel loss function. Additionally, we build a redundancy layer (particularly for high uncertainty and OOD examples) to minimize the risk associated with erroneous decisions based on dubious predictions. We achieve this by learning the mapping between the evidential space and a continuous semantic label embedding space via a recurrent decoder. Thereby inferring, even in the case of OOD examples, reasonably close predictions to avoid catastrophic consequences. We demonstrate the effectiveness of DEED on a digit classification task based on a modified multi-label MNIST dataset, and further evaluate it on a diagnosis prediction task from a real-life electronic health record dataset. We highlight that in terms of prediction scores, our approach is on par with the existing state-of-the-art having a clear advantage of generating reliable, memory and time-efficient uncertainty estimates with minimal changes to any multi-label DNN classifier. © 2023 The Author(s)
|
|
| 3. |
- Bergfeldt, Lennart, 1950, et al.
(författare)
-
Spatial peak and mean QRS-T angles: A comparison of similar but different emerging risk factors for cardiac death.
- 2020
-
Ingår i: Journal of electrocardiology. - : Elsevier BV. - 1532-8430 .- 0022-0736. ; 61, s. 112-120
-
Tidskriftsartikel (refereegranskat)abstract
- The spatial peak and mean QRS-T angles are scientifically but not clinically established risk factors for cardiovascular events including cardiac death. The study aims were to compare these angles, assess their association with hypertension (HT) and diabetes mellitus (DM), and explore the relation between the mean QRS-T angle and the ventricular gradient (VG; reflecting electrical heterogeneity), which both are derived from the QRSarea and Tarea vectors.Altogether 1094 participants (aged 50-65years, 550 women) from the pilot of the population-based Swedish CArdioPulmonary bioImage Study with Frank vectorcardiographic recordings were included and divided into 5 subgroups: apparently healthy n=320; HT n=311; DM n=33; DM+HT n=53; miscellaneous conditions n=377. Abnormal peak and mean QRS-T angles were defined as >95th percentile.Peak QRS-T angles were generally narrower than the mean QRS-T angles; both were narrower in women than in men. Abnormal peak (>124°) and/or mean (>119°) QRS-T angles were found in 73 participants (6.7%). The concordance regarding abnormal versus normal-borderline QRS-T angles was good (Cohen's kappa 0.61). The prevalence of abnormal angles varied from 2.5% in healthy to 21.2% in DM. There was an inverse logarithmical relation between the mean QRS-T angle and the VG.The peak and mean QRS-T angles are not interchangeable but complementary. DM, HT, sex and absence of disease are important determinants of both QRS-T angles. The mean QRS-T angle and the VG relationship is complex. All three VCG derived measures reflect related but differing electrophysiological properties and have potential prognostic value vis-à-vis cardiovascular events.
|
|
| 4. |
- Galozy, Alexander, 1991-, et al.
(författare)
-
Pitfalls of medication adherence approximation through EHR and pharmacy records: Definitions, data and computation
- 2020
-
Ingår i: International Journal of Medical Informatics. - Shannon : Elsevier BV. - 1386-5056 .- 1872-8243. ; 136
-
Tidskriftsartikel (refereegranskat)abstract
- Background and purpose: Patients' adherence to medication is a complex, multidimensional phenomenon. Dispensation data and electronic health records are used to approximate medication-taking through refill adherence. In-depth discussions on the adverse effects of data quality and computational differences are rare. The purpose of this article is to evaluate the impact of common pitfalls when computing medication adherence using electronic health records. Procedures: We point out common pitfalls associated with the data and operationalization of adherence measures. We provide operational definitions of refill adherence and conduct experiments to determine the effect of the pitfalls on adherence estimations. We performed statistical significance testing on the impact of common pitfalls using a baseline scenario as reference. Findings: Slight changes in definition can significantly skew refill adherence estimates. Pickup patterns cause significant disagreement between measures and the commonly used proportion of days covered. Common data related issues had a small but statistically significant (p < 0.05) impact on population-level and significant effect on individual cases. Conclusion: Data-related issues encountered in real-world administrative databases, which affect various operational definitions of refill adherence differently, can significantly skew refill adherence values, leading to false conclusions about adherence, particularly when estimating adherence for individuals.
|
|
| 5. |
- Heyman, Ellen Tolestam, et al.
(författare)
-
Improving Machine Learning 30-Day Mortality Prediction by Discounting Surprising Deaths
- 2021
-
Ingår i: Journal of Emergency Medicine. - Philadelphia, PA : Elsevier BV. - 0736-4679 .- 1090-1280. ; 61:6, s. 763-773
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: Machine learning (ML) is an emerging tool for predicting need of end-of-life discussion and palliative care, by using mortality as a proxy. But deaths, unforeseen by emergency physicians at time of the emergency department (ED) visit, might have a weaker association with the ED visit.OBJECTIVES: To develop an ML algorithm that predicts unsurprising deaths within 30 days after ED discharge.METHODS: In this retrospective registry study, we included all ED attendances within the Swedish region of Halland in 2015 and 2016. All registered deaths within 30 days after ED discharge were classified as either "surprising" or "unsurprising" by an adjudicating committee with three senior specialists in emergency medicine. ML algorithms were developed for the death subclasses by using Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM).RESULTS: Of all 30-day deaths (n = 148), 76% (n = 113) were not surprising to the adjudicating committee. The most common diseases were advanced stage cancer, multidisease/frailty, and dementia. By using LR, RF, and SVM, mean area under the receiver operating characteristic curve (ROC-AUC) of unsurprising deaths in the test set were 0.950 (SD 0.008), 0.944 (SD 0.007), and 0.949 (SD 0.007), respectively. For all mortality, the ROC-AUCs for LR, RF, and SVM were 0.924 (SD 0.012), 0.922 (SD 0.009), and 0.931 (SD 0.008). The difference in prediction performance between all and unsurprising death was statistically significant (P < .001) for all three models.CONCLUSION: In patients discharged to home from the ED, three-quarters of all 30-day deaths did not surprise an adjudicating committee with emergency medicine specialists. When only unsurprising deaths were included, ML mortality prediction improved significantly.
|
|
| 6. |
- Löfström, Emma, et al.
(författare)
-
Dynamics of IgG-avidity and antibody levels after Covid-19
- 2021
-
Ingår i: Journal of Clinical Virology. - Amsterdam : Elsevier. - 1386-6532 .- 1873-5967. ; 144
-
Tidskriftsartikel (refereegranskat)abstract
- Background: A potentially important aspect of the humoral immune response to Covid-19 is avidity, the overall binding strength between antibody and antigen. As low avidity is associated with a risk of re- infection in several viral infections, avidity might be of value to predict risk for reinfection with covid-19. Objectives: The purpose of this study was to describe the maturation of IgG avidity and the antibody-levels over time in patients with PCR-confirmed non-severe covid-19. Study design: Prospective longitudinal cohort study including patients with RT-PCR confirmed covid-19. Blood samples were drawn 1, 3 and 6 months after infection. Antibody levels and IgG-avidity were analysed. Results: The majority had detectable s- and n-antibodies (88,1%, 89,1%, N = 75). The level of total n-antibodies significantly increased from 1 to 3 months (median value 28,3 vs 39,3 s/co, p<0.001) and significantly decreased from 3 to 6 months (median value 39,3 vs 17,1 s/co, p<0.001). A significant decrease in the IgG anti-spike levels (median value 37,6, 24,1 and 18,2 RU/ml, p<0.001) as well as a significant increase in the IgG-avidity index (median values 51,6, 66,0 and 71,0%, p<0.001) were seen from 1 to 3 to 6 months. Conclusion: We found a significant ongoing increase in avidity maturation after Covid-19 whilst the levels of antibodies were declining, suggesting a possible aspect of long-term immunity. © 2021 The Authors. Published by Elsevier B.V.
|
|
| 7. |
- Nikolentzos, G., et al.
(författare)
-
Synthetic electronic health records generated with variational graph autoencoders
- 2023
-
Ingår i: Npj Digital Medicine. - London : Springer Nature. - 2398-6352. ; 6:1
-
Tidskriftsartikel (refereegranskat)abstract
- Data-driven medical care delivery must always respect patient privacy-a requirement that is not easily met. This issue has impeded improvements to healthcare software and has delayed the long-predicted prevalence of artificial intelligence in healthcare. Until now, it has been very difficult to share data between healthcare organizations, resulting in poor statistical models due to unrepresentative patient cohorts. Synthetic data, i.e., artificial but realistic electronic health records, could overcome the drought that is troubling the healthcare sector. Deep neural network architectures, in particular, have shown an incredible ability to learn from complex data sets and generate large amounts of unseen data points with the same statistical properties as the training data. Here, we present a generative neural network model that can create synthetic health records with realistic timelines. These clinical trajectories are generated on a per-patient basis and are represented as linear-sequence graphs of clinical events over time. We use a variational graph autoencoder (VGAE) to generate synthetic samples from real-world electronic health records. Our approach generates health records not seen in the training data. We show that these artificial patient trajectories are realistic and preserve patient privacy and can therefore support the safe sharing of data across organizations.
|
|
| 8. |
- Soliman, Amira, 1980-, et al.
(författare)
-
The Price of Explainability in Machine Learning Models for 100-Day Readmission Prediction in Heart Failure : Retrospective, Comparative, Machine Learning Study
- 2023
-
Ingår i: Journal of Medical Internet Research. - Toronto : JMIR Publications. - 1438-8871. ; 25
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: Sensitive and interpretable machine learning (ML) models can provide valuable assistance to clinicians in managing patients with heart failure (HF) at discharge by identifying individual factors associated with a high risk of readmission. In this cohort study, we delve into the factors driving the potential utility of classification models as decision support tools for predicting readmissions in patients with HF. OBJECTIVE: The primary objective of this study is to assess the trade-off between using deep learning (DL) and traditional ML models to identify the risk of 100-day readmissions in patients with HF. Additionally, the study aims to provide explanations for the model predictions by highlighting important features both on a global scale across the patient cohort and on a local level for individual patients. METHODS: The retrospective data for this study were obtained from the Regional Health Care Information Platform in Region Halland, Sweden. The study cohort consisted of patients diagnosed with HF who were over 40 years old and had been hospitalized at least once between 2017 and 2019. Data analysis encompassed the period from January 1, 2017, to December 31, 2019. Two ML models were developed and validated to predict 100-day readmissions, with a focus on the explainability of the model's decisions. These models were built based on decision trees and recurrent neural architecture. Model explainability was obtained using an ML explainer. The predictive performance of these models was compared against 2 risk assessment tools using multiple performance metrics. RESULTS: The retrospective data set included a total of 15,612 admissions, and within these admissions, readmission occurred in 5597 cases, representing a readmission rate of 35.85%. It is noteworthy that a traditional and explainable model, informed by clinical knowledge, exhibited performance comparable to the DL model and surpassed conventional scoring methods in predicting readmission among patients with HF. The evaluation of predictive model performance was based on commonly used metrics, with an area under the precision-recall curve of 66% for the deep model and 68% for the traditional model on the holdout data set. Importantly, the explanations provided by the traditional model offer actionable insights that have the potential to enhance care planning. CONCLUSIONS: This study found that a widely used deep prediction model did not outperform an explainable ML model when predicting readmissions among patients with HF. The results suggest that model transparency does not necessarily compromise performance, which could facilitate the clinical adoption of such models. © Amira Soliman, Björn Agvall, Kobra Etminani, Omar Hamed, Markus Lingman. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 27.10.2023.
|
|
| 9. |
- Tolestam Heyman, Ellen, 1969, et al.
(författare)
-
Likelihood of admission to hospital from the emergency department is not universally associated with hospital bed occupancy at the time of admission
- 2021
-
Ingår i: International Journal of Health Planning and Management. - : Wiley. - 0749-6753 .- 1099-1751. ; 36:2, s. 353-363
-
Tidskriftsartikel (refereegranskat)abstract
- Background The decision to admit into the hospital from the emergency department (ED) is considered to be important and challenging. The aim was to assess whether previously published results suggesting an association between hospital bed occupancy and likelihood of hospital admission from the ED can be reproduced in a different study population. Methods A retrospective cohort study of attendances at two Swedish EDs in 2015 was performed. Admission to hospital was assessed in relation to hospital bed occupancy together with other clinically relevant variables. Hospital bed occupancy was categorized and univariate and multivariate logistic regression were performed. Results In total 89,503 patient attendances were included in the final analysis. Of those, 29.1% resulted in admission within 24 h. The mean hospital bed occupancy by the hour of the two hospitals was 87.1% (SD 7.6). In both the univariate and multivariate analysis, odds ratio for admission within 24 h from the ED did not decrease significantly with an increasing hospital bed occupancy. Conclusions A negative association between admission to hospital and occupancy level, as reported elsewhere, was not replicated. This suggests that the previously shown association might not be universal but may vary across sites due to setting specific circumstances.
|
|
| 10. |
- Wibring, Kristoffer, et al.
(författare)
-
Clinical presentation in EMS patients with acute chest pain in relation to sex, age and medical history: prospective cohort study
- 2022
-
Ingår i: Bmj Open. - : BMJ. - 2044-6055. ; 12:8
-
Tidskriftsartikel (refereegranskat)abstract
- Objective To assess symptom presentation related to age, sex and previous medical history in patients with chest pain. Design Prospective observational cohort study. Setting Two-centre study in a Swedish county emergency medical service (EMS) organisation. Participants Unselected inclusion of 2917 patients with chest pain cared for by the EMS during 2018. Data analysis Multivariate analysis on the association between symptom characteristics, patients' sex, age, previous acute coronary syndrome (ACS) or diabetes and the final outcome of acute myocardial infarction (AMI). Results Symptomology in patients assessed by the EMS due to acute chest pain varied with sex and age and also with previous ACS or diabetes. Women suffered more often from nausea (OR 1.6) and pain in throat (OR 2.1) or back (OR 2.1). Their pain was more often affected by palpation (1.7) or movement (OR 1.4). Older patients more often described pain onset while sleeping (OR 1.5) and that the onset of symptoms was slow, over hours rather than minutes (OR 1.4). They were less likely to report pain in other parts of their body than their chest (OR 1.4). They were to a lesser extent clammy (OR 0.6) or nauseous (OR 0.6). These differences were present regardless of whether the symptoms were caused by AMI or not. Conclusions A number of aspects of the symptom of chest pain appear to differ in unselected prehospital patients with chest pain in relation to age, sex and medical history, regardless of whether the chest pain was caused by a myocardial infarction or not. This complicates the possibility in prehospital care of using symptoms to predict the underlying aetiology of acute chest pain.
|
|
