| 1. |
- Budu, Emmanuella, 1995-, et al.
(författare)
-
A Framework for Evaluating Synthetic Electronic Health Records
- 2023
-
Ingår i: Caring is Sharing – Exploiting the Value in Data for Health and Innovation. - Amsterdam : IOS Press. - 9781643683881 - 9781643683898 ; , s. 378-379
-
Konferensbidrag (refereegranskat)abstract
- Synthetic data generation can be applied to Electronic Health Records (EHRs) to obtain synthetic versions that do not compromise patients' privacy. However, the proliferation of synthetic data generation techniques has led to the introduction of a wide variety of methods for evaluating the quality of generated data. This makes the task of evaluating generated data from different models challenging as there is no consensus on the methods used. Hence the need for standard ways of evaluating the generated data. In addition, the available methods do not assess whether dependencies between different variables are maintained in the synthetic data. Furthermore, synthetic time series EHRs (patient encounters) are not well investigated, as the available methods do not consider the temporality of patient encounters. In this work, we present an overview of evaluation methods and propose an evaluation framework to guide the evaluation of synthetic EHRs. © 2023 European Federation for Medical Informatics (EFMI) and IOS Press.
|
|
| 2. |
- Etminani, Kobra, 1984-, et al.
(författare)
-
A 3D deep learning model to predict the diagnosis of dementia with Lewy bodies, Alzheimers disease, and mild cognitive impairment using brain 18F-FDG PET
- 2022
-
Ingår i: European Journal of Nuclear Medicine and Molecular Imaging. - New York : Springer. - 1619-7070 .- 1619-7089. ; 49, s. 563-584
-
Tidskriftsartikel (refereegranskat)abstract
- Purpose The purpose of this study is to develop and validate a 3D deep learning model that predicts the final clinical diagnosis of Alzheimers disease (AD), dementia with Lewy bodies (DLB), mild cognitive impairment due to Alzheimers disease (MCI-AD), and cognitively normal (CN) using fluorine 18 fluorodeoxyglucose PET (18F-FDG PET) and compare models performance to that of multiple expert nuclear medicine physicians readers. Materials and methods Retrospective 18F-FDG PET scans for AD, MCI-AD, and CN were collected from Alzheimers disease neuroimaging initiative (556 patients from 2005 to 2020), and CN and DLB cases were from European DLB Consortium (201 patients from 2005 to 2018). The introduced 3D convolutional neural network was trained using 90% of the data and externally tested using 10% as well as comparison to human readers on the same independent test set. The models performance was analyzed with sensitivity, specificity, precision, F1 score, receiver operating characteristic (ROC). The regional metabolic changes driving classification were visualized using uniform manifold approximation and projection (UMAP) and network attention. Results The proposed model achieved area under the ROC curve of 96.2% (95% confidence interval: 90.6-100) on predicting the final diagnosis of DLB in the independent test set, 96.4% (92.7-100) in AD, 71.4% (51.6-91.2) in MCI-AD, and 94.7% (90-99.5) in CN, which in ROC space outperformed human readers performance. The network attention depicted the posterior cingulate cortex is important for each neurodegenerative disease, and the UMAP visualization of the extracted features by the proposed model demonstrates the reality of development of the given disorders. Conclusion Using only 18F-FDG PET of the brain, a 3D deep learning model could predict the final diagnosis of the most common neurodegenerative disorders which achieved a competitive performance compared to the human readers as well as their consensus.
|
|
| 3. |
- Etminani, Kobra, 1984-, et al.
(författare)
-
Peeking inside the box : Transfer Learning vs 3D convolutional neural networks applied in neurodegenerative diseases
- 2021
-
Ingår i: Proceedings of CIBB 2021.
-
Konferensbidrag (refereegranskat)abstract
- Convolutional Neural Networks (CNNs) have shown their effectiveness in a variety of imaging applications including medical imaging diagnostics. However, these deep learning models are data-hungry and need enough labeled samples for the training phase which is limited in the medical domain. Transfer learning is one possible solution to this challenge with training a new model. Assessing model performance should be done not only based on criteria like accuracy, and area under the ROC curve, but also it is important to investigate what regions were of most interest for the classification decisions, especially for medical applications. We performed a case study on neurodegenerative disorders, in specific Alzheimer’s disease, mild cognitive im- pairment, dementia with lewy bodies and cognitively normal brains using 3D 18F-FDG-PET brain scans. Two transfer learning models, InceptionV3 and ResNet50, as well as a custom 3D-CNN that is trained from scratch are compared. Two XAI methods, occlusion and Grad-CAM are chosen to visualize the important brain regions using correctly classified cases. We found that the TL models learn significantly different decision surfaces than the 3D-CNN model. The 3D spatial structure of the brain regions are better kept in the 3D-CNN model, and that might explain the higher performance of this model over 2D-TL models. Moreover, we found out the two XAI methods provide different results, where occlusion method focused more on specific brain regions.
|
|
| 4. |
- Hamed, Omar, 1979-, et al.
(författare)
-
Temporal Context Matters : An Explainable Model for Medical Resource Utilization in Chronic Kidney Disease
- 2023
-
Ingår i: Caring is Sharing – Exploiting the Value in Data for Health and Innovation. - Amsterdam : IOS Press. - 9781643683881 - 9781643683898 ; , s. 613-614
-
Konferensbidrag (refereegranskat)abstract
- The prediction of medical resource utilization is beneficial for effective healthcare resource planning and allocation. Previous work in resource utilization prediction can be categorized into two main classes, count-based and trajectory-based. Both of these classes have some challenges, in this work we propose a hybrid approach to overcome these challenges. Our initial results promote the value of temporal context in resource utilization prediction and highlight the importance of model explainability in understanding the main important variables. © 2023 European Federation for Medical Informatics (EFMI) and IOS Press.
|
|
| 5. |
- Hashemi, Atiye Sadat, 1991-, et al.
(författare)
-
Domain Knowledge-Driven Generation of Synthetic Healthcare Data
- 2023
-
Ingår i: Caring is Sharing – Exploiting the Value in Data for Health and Innovation. - Amsterdam : IOS Press. - 9781643683898 ; , s. 352-353
-
Konferensbidrag (refereegranskat)abstract
- Healthcare longitudinal data collected around patients' life cycles, today offer a multitude of opportunities for healthcare transformation utilizing artificial intelligence algorithms. However, access to "real" healthcare data is a big challenge due to ethical and legal reasons. There is also a need to deal with challenges around electronic health records (EHRs) including biased, heterogeneity, imbalanced data, and small sample sizes. In this study, we introduce a domain knowledge-driven framework for generating synthetic EHRs, as an alternative to methods only using EHR data or expert knowledge. By leveraging external medical knowledge sources in the training algorithm, the suggested framework is designed to maintain data utility, fidelity, and clinical validity while preserving patient privacy. © 2023 European Federation for Medical Informatics (EFMI) and IOS Press.
|
|
| 6. |
- Hashemi, Atiye Sadat, 1991-, et al.
(författare)
-
Time-series Anonymization of Tabular Health Data using Generative Adversarial Network
- 2023
-
Ingår i: 2023 International Joint Conference on Neural Networks (IJCNN). - Piscataway, NJ : IEEE. - 9781665488679 - 9781665488686
-
Konferensbidrag (refereegranskat)abstract
- Data anonymization has been used as a fundamental tool in various domains, e.g. healthcare, to alter personal data such that individuals can no longer be identified directly or indirectly in a way to enable broader sharing of data. For example, data perturbation techniques add noise to original data allowing individual record confidentiality while maintaining high-quality data for analytical purposes. In this paper, we propose a perturbation technique for anonymizing longitudinal tabular data such as electronic health records (EHRs). Our model starts by learning a latent space of original data to better capture temporal trends, then employs a generative adversarial network together to train a perturbation generator. During model training, a time-supervised loss function for handling sequence-dependent noise, together with the adversarial unsupervised, anonymization, and reconstruction loss functions are utilized. To evaluate our model quantitatively, we use multiple evaluation metrics for the fidelity, utility, and identifiability of generated data, in addition, the model is evaluated qualitatively by visualizing generated and original data. The results confirm that our model preserves the privacy of the original data and generates a perturbed version with high fidelity and utility compared to some state-of-the-art techniques. © 2023 IEEE.
|
|
| 7. |
- Nair, Monika, 1985-, et al.
(författare)
-
Machine Learning Model for Readmission Prediction of Patients With Heart Failure Based on Electronic Health Records : Protocol for a Quasi-Experimental Study for Impact Assessment
- 2024
-
Ingår i: JMIR Research Protocols. - Toronto, ON : JMIR Publications. - 1929-0748. ; 13:1
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Care for patients with heart failure (HF) causes a substantial load on health care systems where a prominent challenge is the elevated rate of readmissions within 30 days following initial discharge. Clinical professionals face high levels of uncertainty and subjectivity in the decision-making process on the optimal timing of discharge. Unwanted hospital stays generate costs and cause stress to patients and potentially have an impact on care outcomes. Recent studies have aimed to mitigate the uncertainty by developing and testing risk assessment tools and predictive models to identify patients at risk of readmission, often using novel methods such as machine learning (ML).Objective: This study aims to investigate how a developed clinical decision support (CDS) tool alters the decision-making processes of health care professionals in the specific context of discharging patients with HF, and if so, in which ways. Additionally, the aim is to capture the experiences of health care practitioners as they engage with the system’s outputs to analyze usability aspects and obtain insights related to future implementation.Methods: A quasi-experimental design with randomized crossover assessment will be conducted with health care professionals on HF patients’ scenarios in a region located in the South of Sweden. In total, 12 physicians and nurses will be randomized into control and test groups. The groups shall be provided with 20 scenarios of purposefully sampled patients. The clinicians will be asked to take decisions on the next action regarding a patient. The test group will be provided with the 10 scenarios containing patient data from electronic health records and an outcome from an ML-based CDS model on the risk level for readmission of the same patients. The control group will have 10 other scenarios without the CDS model output and containing only the patients’ data from electronic medical records. The groups will switch roles for the next 10 scenarios. This study will collect data through interviews and observations. The key outcome measures are decision consistency, decision quality, work efficiency, perceived benefits of using the CDS model, reliability, validity, and confidence in the CDS model outcome, integrability in the routine workflow, ease of use, and intention to use. This study will be carried out in collaboration with Cambio Healthcare Systems.Results: The project is part of the Center for Applied Intelligent Systems Research Health research profile, funded by the Knowledge Foundation (2021-2028). Ethical approval for this study was granted by the Swedish ethical review authority (2022-07287-02). The recruitment process of the clinicians and the patient scenario selection will start in September 2023 and last till March 2024.Conclusions: This study protocol will contribute to the development of future formative evaluation studies to test ML models with clinical professionals. © 2024 JMIR Publications Inc.. All rights reserved.
|
|
| 8. |
- Oss Boll, Heloísa, et al.
(författare)
-
Graph neural networks for clinical risk prediction based on electronic health records : A survey
- 2024
-
Ingår i: Journal of Biomedical Informatics. - Maryland Heights, MO : Academic Press. - 1532-0464 .- 1532-0480. ; 151
-
Forskningsöversikt (refereegranskat)abstract
- Objective: This study aims to comprehensively review the use of graph neural networks (GNNs) for clinical risk prediction based on electronic health records (EHRs). The primary goal is to provide an overview of the state-of-the-art of this subject, highlighting ongoing research efforts and identifying existing challenges in developing effective GNNs for improved prediction of clinical risks. Methods: A search was conducted in the Scopus, PubMed, ACM Digital Library, and Embase databases to identify relevant English-language papers that used GNNs for clinical risk prediction based on EHR data. The study includes original research papers published between January 2009 and May 2023. Results: Following the initial screening process, 50 articles were included in the data collection. A significant increase in publications from 2020 was observed, with most selected papers focusing on diagnosis prediction (n = 36). The study revealed that the graph attention network (GAT) (n = 19) was the most prevalent architecture, and MIMIC-III (n = 23) was the most common data resource. Conclusion: GNNs are relevant tools for predicting clinical risk by accounting for the relational aspects among medical events and entities and managing large volumes of EHR data. Future studies in this area may address challenges such as EHR data heterogeneity, multimodality, and model interpretability, aiming to develop more holistic GNN models that can produce more accurate predictions, be effectively implemented in clinical settings, and ultimately improve patient care. © 2024 The Authors
|
|
| 9. |
- Sjöström, Jonas, 1974-, et al.
(författare)
-
Design Principles for Machine Learning Based Clinical Decision Support Systems : A Design Science Study
- 2024
-
Ingår i: Design Science Research for a Resilient Future. - Cham : Springer. - 9783031611742 - 9783031611759 ; , s. 109-122
-
Konferensbidrag (refereegranskat)abstract
- Employing a design science research approach building on four modes of inquiry, this study presents a Clinical Decision Support System for predicting heart failure readmissions, combining machine learning, inpatient care process analysis, and user experience design. It introduces three key design principles: contextual integration, actionable insights, and adaptive explanation levels, to support the design of decision support in clinical settings. The research, while focused on a specific healthcare context, offers a model for integrating technical precision and user-centric design in inpatient care processes, suggesting broader applications and future research directions in diverse healthcare environments. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
|
|
| 10. |
- Soliman, Amira, 1980-, et al.
(författare)
-
Adopting transfer learning for neuroimaging : a comparative analysis with a custom 3D convolution neural network model
- 2022
-
Ingår i: BMC Medical Informatics and Decision Making. - London : BioMed Central (BMC). - 1472-6947. ; 22, s. 1-15
-
Tidskriftsartikel (refereegranskat)abstract
- Background: In recent years, neuroimaging with deep learning (DL) algorithms have made remarkable advances in the diagnosis of neurodegenerative disorders. However, applying DL in different medical domains is usually challenged by lack of labeled data. To address this challenge, transfer learning (TL) has been applied to use state-of-the-art convolution neural networks pre-trained on natural images. Yet, there are differences in characteristics between medical and natural images, also image classification and targeted medical diagnosis tasks. The purpose of this study is to investigate the performance of specialized and TL in the classification of neurodegenerative disorders using 3D volumes of 18F-FDG-PET brain scans. Results: Results show that TL models are suboptimal for classification of neurodegenerative disorders, especially when the objective is to separate more than two disorders. Additionally, specialized CNN model provides better interpretations of predicted diagnosis. Conclusions: TL can indeed lead to superior performance on binary classification in timely and data efficient manner, yet for detecting more than a single disorder, TL models do not perform well. Additionally, custom 3D model performs comparably to TL models for binary classification, and interestingly perform better for diagnosis of multiple disorders. The results confirm the superiority of the custom 3D-CNN in providing better explainable model compared to TL adopted ones. © 2022, The Author(s).
|
|
