BMC Medical Informatics and Decision Making最新文献

Background: There have been many efforts to expand existing data collection initiatives to include COVID-19 related data. One program that expanded is UK Biobank, a large-scale research and biomedical data collection resource that added several COVID-19 related data fields including questionnaires (exposures and symptoms), viral testing, and serological data. This study aimed to analyze this COVID-19 data to understand how COVID-19 data was collected and how it can be used to attribute COVID-19 and analyze differences in cohorts and time periods.

Methods: A cohort of COVID-19 infected individuals was defined from the UK Biobank population using viral testing, diagnosis, and self-reported data. Changes over time, from March 2020 to October 2021, in total case counts and changes in case counts by identification source (diagnosis from EHR, measurement from viral testing and self-reported from questionnaire) were also analyzed. For the questionnaires, an analysis of the structure and dynamics of the questionnaires was done which included the amount and type of questions asked, how often and how many individuals answered the questions and what responses were given. In addition, the amount of individuals who provided responses regarding different time segments covered by the questionnaire was calculated along with how often responses changed. The analysis included changes in population level responses over time. The analyses were repeated for COVID and non-COVID individuals and compared responses.

Results: There were 62 042 distinct participants who had COVID-19, with 49 120 identified through diagnosis, 30 553 identified through viral testing and 934 identified through self-reporting, with many identified in multiple methods. This included vast changes in overall cases and distribution of case data source over time. 6 899 of 9 952 participants completing the exposure questionnaire responded regarding every time period covered by the questionnaire including large changes in response over time. The most common change came for employment situation, which was changed by 74.78% of individuals from the first to last time of asking. On a population level, there were changes as face mask usage increased each successive time period. There were decreases in nearly every COVID-19 symptom from the first to the second questionnaire. When comparing COVID to non-COVID participants, COVID participants were more commonly keyworkers (COVID: 33.76%, non-COVID: 15.00%) and more often lived with young people attending school (61.70%, 45.32%).

Conclusion: To develop a robust cohort of COVID-19 participants from the UK Biobank population, multiple types of data were needed. The differences based on time and exposures show the important of comprehensive data capture and the utility of COVID-19 related questionnaire data.

Background: DNA microarrays provide informative data for transcriptional profiling and identifying gene expression signatures to help prevent progression of latent tuberculosis infection (LTBI) to active disease. However, constructing a prognostic model for distinguishing LTBI from active tuberculosis (ATB) is very challenging due to the noisy nature of data and lack of a generally stable analysis approach.

Methods: In the present study, we proposed an accurate predictive model with the help of data fusion at the decision level. In this regard, results of filter feature selection and wrapper feature selection techniques were combined with multiple-criteria decision-making (MCDM) methods to select 10 genes from six microarray datasets that can be the most discriminative genes for diagnosing tuberculosis cases. As the main contribution of this study, the final ranking function was constructed by combining protein-protein interaction (PPI) network with an MCDM method (called Decision-making Trial and Evaluation Laboratory or DEMATEL) to improve the feature ranking approach.

Results: By applying data fusion at the decision level on the 10 introduced genes in terms of fusion of classifiers of random forests (RF) and k-nearest neighbors (KNN) regarding Yager's theory, the proposed algorithm reached a sensitivity of 0.97, specificity of 0.90, and accuracy of 0.95. Finally, with the help of cumulative clustering, the genes involved in the diagnosis of latent and activated tuberculosis have been introduced.

Conclusions: The combination of MCDM methods and PPI networks can significantly improve the diagnosis different states of tuberculosis.

Clinical trial number: Not applicable.

Background: Ageing is one of the most important challenges in our society. Evaluating how one is ageing is important in many aspects, from giving personalized recommendations to providing insight for long-term care eligibility. Machine learning can be utilized for that purpose, however, user reservations towards "black-box" predictions call for increased transparency and explainability of results. This study aimed to explore the potential of developing a machine learning-based healthy ageing scale that provides explainable results that could be trusted and understood by informal carers.

Methods: In this study, we used data from 696 older adults collected via personal field interviews as part of independent research. Explanatory factor analysis was used to find candidate healthy ageing aspects. For visualization of key aspects, a web annotation application was developed. Key aspects were selected by gerontologists who later used web annotation applications to evaluate healthy ageing for each older adult on a Likert scale. Logistic Regression, Decision Tree Classifier, Random Forest, KNN, SVM and XGBoost were used for multi-classification machine learning. AUC OvO, AUC OvR, F1, Precision and Recall were used for evaluation. Finally, SHAP was applied to best model predictions to make them explainable.

Results: The experimental results show that human annotations of healthy ageing could be modelled using machine learning where among several algorithms XGBoost showed superior performance. The use of XGBoost resulted in 0.92 macro-averaged AuC OvO and 0.76 macro-averaged F1. SHAP was applied to generate local explanations for predictions and shows how each feature is influencing the prediction.

Conclusion: The resulting explainable predictions make a step toward practical scale implementation into decision support systems. The development of such a decision support system that would incorporate an explainable model could reduce user reluctance towards the utilization of AI in healthcare and provide explainable and trusted insights to informal carers or healthcare providers as a basis to shape tangible actions for improving ageing. Furthermore, the cooperation with gerontology specialists throughout the process also indicates expert knowledge as integrated into the model.

Background: Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease that results in death within a short time span (3-5 years). One of the major challenges in treating ALS is its highly heterogeneous disease progression and the lack of effective prognostic tools to forecast it. The main aim of this study was, then, to test the feasibility of predicting relevant clinical outcomes that characterize the progression of ALS with a two-year prediction horizon via artificial intelligence techniques using routine visits data.

Methods: Three classification problems were considered: predicting death (binary problem), predicting death or percutaneous endoscopic gastrostomy (PEG) (multiclass problem), and predicting death or non-invasive ventilation (NIV) (multiclass problem). Two supervised learning models, a logistic regression (LR) and a deep learning multilayer perceptron (MLP), were trained ensuring technical robustness and reproducibility. Moreover, to provide insights into model explainability and result interpretability, model coefficients for LR and Shapley values for both LR and MLP were considered to characterize the relationship between each variable and the outcome.

Results: On the one hand, predicting death was successful as both models yielded F1 scores and accuracy well above 0.7. The model explainability analysis performed for this outcome allowed for the understanding of how different methodological approaches consider the input variables when performing the prediction. On the other hand, predicting death alongside PEG or NIV proved to be much more challenging (F1 scores and accuracy in the 0.4-0.6 interval).

Conclusions: In conclusion, predicting death due to ALS proved to be feasible. However, predicting PEG or NIV in a multiclass fashion proved to be unfeasible with these data, regardless of the complexity of the methodological approach. The observed results suggest a potential ceiling on the amount of information extractable from the database, e.g., due to the intrinsic difficulty of the prediction tasks at hand, or to the absence of crucial predictors that are, however, not currently collected during routine practice.

Importance: Decision-making in trauma patients remains challenging and often results in deviation from guidelines. Machine-Learning (ML) enhanced decision-support could improve hemorrhage resuscitation.

Aim: To develop a ML enhanced decision support tool to predict Need for Hemorrhage Resuscitation (NHR) (part I) and test the collection of the predictor variables in real time in a smartphone app (part II).

Design, setting, and participants: Development of a ML model from a registry to predict NHR relying exclusively on prehospital predictors. Several models and imputation techniques were tested. Assess the feasibility to collect the predictors of the model in a customized smartphone app during prealert and generate a prediction in four level-1 trauma centers to compare the predictions to the gestalt of the trauma leader.

Main outcomes and measures: Part 1: Model output was NHR defined by 1) at least one RBC transfusion in resuscitation, 2) transfusion ≥ 4 RBC within 6 h, 3) any hemorrhage control procedure within 6 h or 4) death from hemorrhage within 24 h. The performance metric was the F4-score and compared to reference scores (RED FLAG, ABC). In part 2, the model and clinician prediction were compared with Likelihood Ratios (LR).

Results: From 36,325 eligible patients in the registry (Nov 2010-May 2022), 28,614 were included in the model development (Part 1). Median age was 36 [25-52], median ISS 13 [5-22], 3249/28614 (11%) corresponded to the definition of NHR. A XGBoost model with nine prehospital variables generated the best predictive performance for NHR according to the F4-score with a score of 0.76 [0.73-0.78]. Over a 3-month period (Aug-Oct 2022), 139 of 391 eligible patients were included in part II (38.5%), 22/139 with NHR. Clinician satisfaction was high, no workflow disruption observed and LRs comparable between the model and the clinicians.

Conclusions and relevance: The ShockMatrix pilot study developed a simple ML-enhanced NHR prediction tool demonstrating a comparable performance to clinical reference scores and clinicians. Collecting the predictor variables in real-time on prealert was feasible and caused no workflow disruption.

Background: Decision-making and problem-solving processes are powerful activities occurring daily across all healthcare settings. Their empowering potential is seldom fully exploited, and they may even be perceived as disempowering. We developed the EMPOWER-UP questionnaire to enable assessment of healthcare users' perception of empowerment across health conditions, healthcare settings, and healthcare providers' professional backgrounds. This article reports the initial development of EMPOWER-UP, including face and content validation.

Methods: Four grounded theories explaining barriers and enablers to empowerment in relational decision-making and problem-solving were reviewed to generate a preliminary item pool, which was subsequently reduced using constant comparison. Preliminary items were evaluated for face and content validity using an expert panel of seven researchers and cognitive interviews in Danish and English with 29 adults diagnosed with diabetes, cancer, or schizophrenia.

Results: A preliminary pool of 139 items was reduced to 46. Independent feedback from expert panel members resulted in further item reduction and modifications supporting content validity and strengthening the potential for generic use. Forty-one preliminary items were evaluated through 29 cognitive interviews, resulting in a 36-item draft questionnaire deemed to have good face and content validity and generic potential.

Conclusions: Face and content validation using an expert panel and cognitive interviews resulted in a 36-item draft questionnaire with a potential for evaluating empowerment in user-provider interactions regardless of health conditions, healthcare settings, and healthcare providers' professional backgrounds.

Breast cancer is the most common cancer in women. Previous studies have investigated estimating and predicting the proportional hazard rates and survival in breast cancer. This study deals with predicting accelerated hazards (AH) rate based on age categories in breast cancer patients using deep learning methods. The AH has a time-dependent structure whose rate changes according to time and variable effects. We have collected data related to 1225 female patients with breast cancer at the Mandarin University of Medical Sciences. The patients' demographic and clinical characteristics including family history, age, history of tobacco use, hysterectomy, first menstruation age, gravida, number of breastfeeding, disease grade, marital status, and survival status have been recorded. Initially, we dealt with predicting three age groups of patients: ≤ 40, 41-60, and ≥ 61 years. Then, the prediction of accelerated risk value based on age categories for each breast cancer patient through deep learning and the importance of variables using LightGBM is discussed. Improving clinical management and treatment of breast cancer requires advanced methods such as time-dependent AH calculation. When the behavioral effect is assumed as a time scale change between hazard functions, the AH model is more appropriate for randomized clinical trials. The study results demonstrate the proper performance of the proposed model for predicting AH by age categories based on breast cancer patients' demographic and clinical characteristics.

Introduction: Clinical notes, biomarkers, and neuroimaging have proven valuable in dementia prediction models. Whether commonly available structured clinical data can predict dementia is an emerging area of research. We aimed to predict gold-standard, research-based diagnoses of dementia including Alzheimer's disease (AD) and/or Alzheimer's disease related dementias (ADRD), in addition to ICD-based AD and/or ADRD diagnoses, in a well-phenotyped, population-based cohort using a machine learning approach.

Methods: Administrative healthcare data (k = 163 diagnostic features), in addition to census/vital record sociodemographic data (k = 6 features), were linked to the Cache County Study (CCS, 1995-2008).

Results: Among successfully linked UPDB-CCS participants (n = 4206), 522 (12.4%) had incident dementia (AD alone, AD comorbid with ADRD, or ADRD alone) as per the CCS "gold standard" assessments. Random Forest models, with a 1-year prediction window, achieved the best performance with an Area Under the Curve (AUC) of 0.67. Accuracy declined for dementia subtypes: AD/ADRD (AUC = 0.65); ADRD (AUC = 0.49). Accuracy improved when using ICD-based dementia diagnoses (AUC = 0.77).

Discussion: Commonly available structured clinical data (without labs, notes, or prescription information) demonstrate modest ability to predict "gold-standard" research-based AD/ADRD diagnoses, corroborated by prior research. Using ICD diagnostic codes to identify dementia as done in the majority of machine learning dementia prediction models, as compared to "gold-standard" dementia diagnoses, can result in higher accuracy, but whether these models are predicting true dementia warrants further research.

Massive transfusion of blood products poses challenges in determining the need for transfusion and the appropriate volume of blood products. This review explores the use of machine learning (ML) models to predict transfusion risk during surgical procedure, focusing on the methodology, variables, and software employed to predict transfusion. This scoping review investigates the development and current state of machine learning models for predicting transfusion risk during surgical procedure, aiming to inform physicians about the field's progress and potential directions.The review was conducted using the databases Cochrane, Embase, and PubMed. The search included keywords related to blood transfusion, statistical models, and surgical procedures. Peer-reviewed articles were included, while literature reviews, case reports, and non-human studies were excluded.A total of 40 studies met the inclusion criteria. The most frequently studied biological variables included haemoglobin, platelet count, international normalized ratio (INR), activated partial thromboplastin time (aPTT), fibrinogen, creatinine, white blood cells, and albumin. Clinical variables of importance included age, sex, surgery type, blood pressure, weight, surgery duration, american society of anesthesiology (ASA) status, blood loss, and body mass index (BMI). The software employed varied, with Python, R, SPSS, and SAS being the most commonly used. Logistic regression was the predominant methodology used in 20 studies.Our scoping review highlights the need for improved reporting and transparency in methodology, variables, and software used. Future research should focus on providing detailed descriptions and open access to codes of respective models, promoting reproducibility, and enhancing the clinical relevance of transfusion risk prediction models.

Several diverse fields including the healthcare system and drug development sectors have benefited immensely through the adoption of deep learning (DL), which is a subset of artificial intelligence (AI) and machine learning (ML). Cancer makes up a significant percentage of the illnesses that cause early human mortality across the globe, and this situation is likely to rise in the coming years, especially when non-communicable illnesses are not considered. As a result, cancer patients would greatly benefit from precise and timely diagnosis and prediction. Deep learning (DL) has become a common technique in healthcare due to the abundance of computational power. Gene expression datasets are frequently used in major DL-based applications for illness detection, notably in cancer therapy. The quantity of medical data, on the other hand, is often insufficient to fulfill deep learning requirements. Microarray gene expression datasets are used for training procedures despite their extreme dimensionality, limited volume of data samples, and sparsely available information. Data augmentation is commonly used to expand the training sample size for gene data. The Wasserstein Tabular Generative Adversarial Network (WT-GAN) model is used for the data augmentation process for generating synthetic data in this proposed work. The correlation-based feature selection technique selects the most relevant characteristics based on threshold values. Deep FNN and ML algorithms train and classify the gene expression samples. The augmented data give better classification results (> 97%) when using WT-GAN for cancer diagnosis.