| 1. |
- Böhmer, Jens, 1981, et al.
(author)
-
Absolute Quantification of Donor-Derived Cell-Free DNA in Pediatric and Adult Patients After Heart Transplantation: A Prospective Study.
- 2023
-
In: Transplant international : official journal of the European Society for Organ Transplantation. - 0934-0874 .- 1432-2277. ; 36
-
Journal article (peer-reviewed)abstract
- In this prospective study we investigated a cohort after heart transplantation with a novel PCR-based approach with focus on treated rejection. Blood samples were collected coincidentally to biopsies, and both absolute levels of dd-cfDNA and donor fraction were reported using digital PCR. 52 patients (11 children and 41 adults) were enrolled (NCT03477383, clinicaltrials.gov), and 557 plasma samples were analyzed. 13 treated rejection episodes >14 days after transplantation were observed in 7 patients. Donor fraction showed a median of 0.08% in the cohort and was significantly elevated during rejection (median 0.19%, p < 0.0001), using a cut-off of 0.1%, the sensitivity/specificity were 92%/56% (AUC ROC-curve: 0.78). Absolute levels of dd-cfDNA showed a median of 8.8 copies/mL and were significantly elevated during rejection (median 23, p = 0.0001). Using a cut-off of 7.5 copies/mL, the sensitivity/specificity were 92%/43% for donor fraction (AUC ROC-curve: 0.75). The results support the feasibility of this approach in analyzing dd-cfDNA after heart transplantation. The obtained values are well aligned with results from other trials. The possibility to quantify absolute levels adds important value to the differentiation between ongoing graft damage and quiescent situations.
|
|
| 2. |
- Hagberg, Eva, et al.
(author)
-
Semi-supervised learning with natural language processing for right ventricle classification in echocardiography—a scalable approach
- 2022
-
In: Computers in Biology and Medicine. - : Elsevier BV. - 0010-4825 .- 1879-0534. ; 143
-
Journal article (peer-reviewed)abstract
- We created a deep learning model, trained on text classified by natural language processing (NLP), to assess right ventricular (RV) size and function from echocardiographic images. We included 12,684 examinations with corresponding written reports for text classification. After manual annotation of 1489 reports, we trained an NLP model to classify the remaining 10,651 reports. A view classifier was developed to select the 4-chamber or RV-focused view from an echocardiographic examination (n = 539). The final models were two image classification models trained on the predicted labels from the combined manual annotation and NLP models and the corresponding echocardiographic view to assess RV function (training set n = 11,008) and size (training set n = 9951. The text classifier identified impaired RV function with 99% sensitivity and 98% specificity and RV enlargement with 98% sensitivity and 98% specificity. The view classification model identified the 4-chamber view with 92% accuracy and the RV-focused view with 73% accuracy. The image classification models identified impaired RV function with 93% sensitivity and 72% specificity and an enlarged RV with 80% sensitivity and 85% specificity; agreement with the written reports was substantial (both κ = 0.65). Our findings show that models for automatic image assessment can be trained to classify RV size and function by using model-annotated data from written echocardiography reports. This pipeline for auto-annotation of the echocardiographic images, using a NLP model with medical reports as input, can be used to train an image-assessment model without manual annotation of images and enables fast and inexpensive expansion of the training dataset when needed. © 2022
|
|
| 3. |
- Javeed, Ashir, 1989-, et al.
(author)
-
Predictive Power of XGBoost_BiLSTM Model : A Machine-Learning Approach for Accurate Sleep Apnea Detection Using Electronic Health Data
- 2023
-
In: International Journal of Computational Intelligence Systems. - : Springer Nature. - 1875-6891 .- 1875-6883. ; 16:1
-
Journal article (peer-reviewed)abstract
- Sleep apnea is a common disorder that can cause pauses in breathing and can last from a few seconds to several minutes, as well as shallow breathing or complete cessation of breathing. Obstructive sleep apnea is strongly associated with the risk of developing several heart diseases, including coronary heart disease, heart attack, heart failure, and stroke. In addition, obstructive sleep apnea increases the risk of developing irregular heartbeats (arrhythmias), which can lead to low blood pressure. To prevent these conditions, this study presents a novel machine-learning (ML) model for predicting sleep apnea based on electronic health data that provides accurate predictions and helps in identifying the risk factors that contribute to the development of sleep apnea. The dataset used in the study includes 75 features and 10,765 samples from the Swedish National Study on Aging and Care (SNAC). The proposed model is based on two modules: the XGBoost module assesses the most important features from feature space, while the Bidirectional Long Short-Term Memory Networks (BiLSTM) module classifies the probability of sleep apnea. Using a cross-validation scheme, the proposed XGBoost_BiLSTM algorithm achieves an accuracy of 97% while using only the six most significant features from the dataset. The model’s performance is also compared with conventional long-short-term memory networks (LSTM) and other state-of-the-art ML models. The results of the study suggest that the proposed model improved the diagnosis and treatment of sleep apnea by identifying the risk factors.
|
|
| 4. |
- Gonzalez, Manuel, et al.
(author)
-
Effect of a lifestyle-focused electronic patient support application for improving risk factor management, self-rated health, and prognosis in post-myocardial infarction patients : study protocol for a multi-center randomized controlled trial
- 2019
-
In: Trials. - : Springer Science and Business Media LLC. - 1745-6215. ; 20:1
-
Journal article (peer-reviewed)abstract
- BACKGROUND: Cardiac rehabilitation (CR) programs addressing risk factor management, educational interventions, and exercise contribute to reduce mortality after myocardial infarction (MI). However, the fulfillment of guideline-recommended CR targets is currently unsatisfactory. eHealth, i.e., the use of electronic communication for healthcare, including the use of mobile smartphone applications combined with different sensors and interactive computerized programs, offers a new array of possibilities to provide clinical care. The present study aims to assess the efficacy of a web-based application (app) designed to support persons in adhering to lifestyle advice and medication as a complement to traditional CR programs for improvement of risk factors and clinical outcomes in patients with MI compared with usual care. METHODS/DESIGN: An open-label multi-center randomized controlled trial is being conducted at different CR centers from three Swedish University Hospitals. The aim is to include 150 patients with MI < 75 years of age who are confident smartphone and/or Internet users. In addition to participation in CR programs according to the usual routine at each center, patients randomized to the intervention arm will receive access to the web-based app. A CR nurse reviews the patients' self-reported data twice weekly through a medical interface at the clinic. The primary outcome of the study will be change in submaximal exercise capacity (in watts) between 2 and 4 weeks after discharge and when the patient has completed his/her exercise program at the CR center, usually around 3-6 months post-discharge. Secondary outcomes include changes in self-reported physical activity, objectively assessed physical activity by accelerometry, self-rated health, dietary, and smoking habits, body mass index, blood pressure, blood lipids, and glucose/HbA1c levels between inclusion and follow-up visits during the first year post-MI. Additionally, we will assess uptake and adherence to the application, the number of CR staff contacts, and the incidence of cardiovascular events at 1 and 3 years after the MI. Patient recruitment started in 2016, and the first study results are expected in the beginning of 2019. DISCUSSION: The present study will add evidence to whether electronic communication can be used to improve traditional CR programs for patients after MI. TRIAL REGISTRATION: ClinicalTrials.gov, NCT03260582 . Retrospectively registered on 24 August 2017.
|
|
| 5. |
- Javeed, Ashir, 1989-, et al.
(author)
-
Decision Support System for Predicting Mortality in Cardiac Patients Based on Machine Learning
- 2023
-
In: Applied Sciences. - : MDPI. - 2076-3417. ; 13:8
-
Journal article (peer-reviewed)abstract
- Researchers have proposed several automated diagnostic systems based on machine learning and data mining techniques to predict heart failure. However, researchers have not paid close attention to predicting cardiac patient mortality. We developed a clinical decision support system for predicting mortality in cardiac patients to address this problem. The dataset collected for the experimental purposes of the proposed model consisted of 55 features with a total of 368 samples. We found that the classes in the dataset were highly imbalanced. To avoid the problem of bias in the machine learning model, we used the synthetic minority oversampling technique (SMOTE). After balancing the classes in the dataset, the newly proposed system employed a (Formula presented.) statistical model to rank the features from the dataset. The highest-ranked features were fed into an optimized random forest (RF) model for classification. The hyperparameters of the RF classifier were optimized using a grid search algorithm. The performance of the newly proposed model ((Formula presented.) _RF) was validated using several evaluation measures, including accuracy, sensitivity, specificity, F1 score, and a receiver operating characteristic (ROC) curve. With only 10 features from the dataset, the proposed model (Formula presented.) _RF achieved the highest accuracy of 94.59%. The proposed model (Formula presented.) _RF improved the performance of the standard RF model by 5.5%. Moreover, the proposed model (Formula presented.) _RF was compared with other state-of-the-art machine learning models. The experimental results show that the newly proposed decision support system outperforms the other machine learning systems using the same feature selection module ((Formula presented.)).
|
|
| 6. |
- Liu, Xixi, 1995, et al.
(author)
-
Deep Nearest Neighbors for Anomaly Detection in Chest X-Rays
- 2024
-
In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). - 1611-3349 .- 0302-9743. ; 14349 LNCS, s. 293-302
-
Conference paper (peer-reviewed)abstract
- Identifying medically abnormal images is crucial to the diagnosis procedure in medical imaging. Due to the scarcity of annotated abnormal images, most reconstruction-based approaches for anomaly detection are trained only with normal images. At test time, images with large reconstruction errors are declared abnormal. In this work, we propose a novel feature-based method for anomaly detection in chest x-rays in a setting where only normal images are provided during training. The model consists of lightweight adaptor and predictor networks on top of a pre-trained feature extractor. The parameters of the pre-trained feature extractor are frozen, and training only involves fine-tuning the proposed adaptor and predictor layers using Siamese representation learning. During inference, multiple augmentations are applied to the test image, and our proposed anomaly score is simply the geometric mean of the k-nearest neighbor distances between the augmented test image features and the training image features. Our method achieves state-of-the-art results on two challenging benchmark datasets, the RSNA Pneumonia Detection Challenge dataset, and the VinBigData Chest X-ray Abnormalities Detection dataset. Furthermore, we empirically show that our method is robust to different amounts of anomalies among the normal images in the training dataset. The code is available at: https://github.com/XixiLiu95/deep-kNN-anomaly-detection.
|
|
| 7. |
- Abdellah, Tebani, et al.
(author)
-
Integration of molecular profiles in a longitudinal wellness profiling cohort.
- 2020
-
In: Nature communications. - : Springer Science and Business Media LLC. - 2041-1723. ; 11:1
-
Journal article (peer-reviewed)abstract
- An important aspect of precision medicine is to probe the stability in molecular profiles among healthy individuals over time. Here, we sample a longitudinal wellness cohort with 100 healthy individuals and analyze blood molecular profiles including proteomics, transcriptomics, lipidomics, metabolomics, autoantibodies and immune cell profiling, complemented with gut microbiota composition and routine clinical chemistry. Overall, our results show high variation between individuals across different molecular readouts, while the intra-individual baseline variation is low. The analyses show that each individual has a unique and stable plasma protein profile throughout the study period and that many individuals also show distinct profiles with regards to the other omics datasets, with strong underlying connections between the blood proteome and the clinical chemistry parameters. In conclusion, the results support an individual-based definition of health and show that comprehensive omics profiling in a longitudinal manner is a path forward for precision medicine.
|
|
| 8. |
- Andersson, Eva M., 1968, et al.
(author)
-
Road traffic noise, air pollution and cardiovascular events in a Swedish cohort
- 2020
-
In: Environmental Research. - : Elsevier BV. - 0013-9351. ; 185
-
Journal article (peer-reviewed)abstract
- Urbanization and increasing road traffic cause exposure to both noise and air pollution. While the levels of air pollutants such as nitrogen oxides (NOx) have decreased in Sweden during the past decades, exposure to traffic noise has increased. The association with cardiovascular morbidity is less well established for noise than for air pollution, and most studies have only studied one of the two highly spatially correlated exposures. The Swedish Primary Prevention Study cohort consists of men aged 47 to 55 when first examined in 1970-1973. The cohort members were linked to the Swedish patient registry through their personal identity number and followed until first cardiovascular event 1970-2011. The address history during the entire study period was used to assign annual modelled residential exposure to road traffic noise and NOx. The Cox proportional hazards model with age on the time axis and time-varying exposures were used in the analysis. The results for 6304 men showed a non-significant increased risk of cardiovascular disease for long-term road traffic noise at the home address, after adjusting for air pollution. The hazard ratios were 1.08 (95% CI 0.90-1.28) for cardiovascular mortality, 1.14 (95% CI 0.96-1.36) for ischemic heart disease incidence and 1.07 (95% CI 0.85-1.36) for stroke incidence, for noise above 60 dB, compared to below 50 dB. This study found some support for cardiovascular health effects of long-term exposure to road traffic noise above 60 dB, after having accounted for exposure to air pollution.
|
|
| 9. |
- Baldanzi, Gabriel, et al.
(author)
-
OSA Is Associated With the Human Gut Microbiota Composition and Functional Potential in the Population-Based Swedish CardioPulmonary bioImage Study
- 2023
-
In: Chest. - : Elsevier. - 0012-3692 .- 1931-3543. ; 164:2, s. 503-516
-
Journal article (peer-reviewed)abstract
- BACKGROUND: Obstructive sleep apnea (OSA) is a common sleep-breathing disorder linked to increased risk of cardiovascular disease. Intermittent hypoxia and intermittent airway obstruction, hallmarks of OSA, have been shown in animal models to induce substantial changes to the gut microbiota composition and subsequent transplantation of fecal matter to other animals induced changes in blood pressure and glucose metabolism.RESEARCH QUESTION: Does obstructive sleep apnea in adults associate with the composition and metabolic potential of the human gut microbiota?STUDY DESIGN AND METHODS: We used respiratory polygraphy data from up to 3,570 individuals aged 50-64 from the population-based Swedish CardioPulmonary bioImage Study combined with deep shotgun metagenomics of fecal samples to identify cross-sectional associations between three OSA parameters covering apneas and hypopneas, cumulative sleep time in hypoxia and number of oxygen desaturation events with gut microbiota composition. Data collection about potential confounders was based on questionnaires, on-site anthropometric measurements, plasma metabolomics, and linkage with the Swedish Prescribed Drug Register.RESULTS: We found that all three OSA parameters were associated with lower diversity of species in the gut. Further, the OSA-related hypoxia parameters were in multivariable-adjusted analysis associated with the relative abundance of 128 gut bacterial species, including higher abundance of Blautia obeum and Collinsela aerofaciens. The latter species was also independently associated with increased systolic blood pressure. Further, the cumulative time in hypoxia during sleep was associated with the abundance of genes involved in nine gut microbiota metabolic pathways, including propionate production from lactate. Lastly, we observed two heterogeneous sets of plasma metabolites with opposite association with species positively and negatively associated with hypoxia parameters, respectively.INTERPRETATION: OSA-related hypoxia, but not the number of apneas/hypopneas, is associated with specific gut microbiota species and functions. Our findings lay the foundation for future research on the gut microbiota-mediated health effects of OSA.
|
|
| 10. |
- Belda, E., et al.
(author)
-
Impairment of gut microbial biotin metabolism and host biotin status in severe obesity: effect of biotin and prebiotic supplementation on improved metabolism
- 2022
-
In: Gut. - : BMJ. - 0017-5749 .- 1468-3288. ; 71:12, s. 2463-2480.
-
Journal article (peer-reviewed)abstract
- Objectives Gut microbiota is a key component in obesity and type 2 diabetes, yet mechanisms and metabolites central to this interaction remain unclear. We examined the human gut microbiome's functional composition in healthy metabolic state and the most severe states of obesity and type 2 diabetes within the MetaCardis cohort. We focused on the role of B vitamins and B7/B8 biotin for regulation of host metabolic state, as these vitamins influence both microbial function and host metabolism and inflammation. Design We performed metagenomic analyses in 1545 subjects from the MetaCardis cohorts and different murine experiments, including germ-free and antibiotic treated animals, faecal microbiota transfer, bariatric surgery and supplementation with biotin and prebiotics in mice. Results Severe obesity is associated with an absolute deficiency in bacterial biotin producers and transporters, whose abundances correlate with host metabolic and inflammatory phenotypes. We found suboptimal circulating biotin levels in severe obesity and altered expression of biotin-associated genes in human adipose tissue. In mice, the absence or depletion of gut microbiota by antibiotics confirmed the microbial contribution to host biotin levels. Bariatric surgery, which improves metabolism and inflammation, associates with increased bacterial biotin producers and improved host systemic biotin in humans and mice. Finally, supplementing high-fat diet-fed mice with fructo-oligosaccharides and biotin improves not only the microbiome diversity, but also the potential of bacterial production of biotin and B vitamins, while limiting weight gain and glycaemic deterioration. Conclusion Strategies combining biotin and prebiotic supplementation could help prevent the deterioration of metabolic states in severe obesity.
|
|
