Machine learning approach to dynamic risk modeling of mortality in COVID-19: a UK Biobank study

Unlocking the potential of big data and AI in medicine: insights from biobanking

Big data and artificial intelligence are key elements in the medical field as they are expected to improve accuracy and efficiency in diagnosis and treatment, particularly in identifying biomedically relevant patterns, facilitating progress towards individually tailored preventative and therapeutic interventions. These applications belong to current research practice that is data-intensive. While the combination of imaging, pathological, genomic, and clinical data is needed to train algorithms to realize the full potential of these technologies, biobanks often serve as crucial infrastructures for data-sharing and data flows. In this paper, we argue that the 'data turn' in the life sciences has increasingly re-structured major infrastructures, which often were created for biological samples and associated data, as predominantly data infrastructures. These have evolved and diversified over time in terms of tackling relevant issues such as harmonization and standardization, but also consent practices and risk assessment. In line with the datafication, an increased use of AI-based technologies marks the current developments at the forefront of the big data research in life science and medicine that engender new issues and concerns along with opportunities. At a time when secure health data environments, such as European Health Data Space, are in the making, we argue that such meta-infrastructures can benefit both from the experience and evolution of biobanking, but also the current state of affairs in AI in medicine, regarding good governance, the social aspects and practices, as well as critical thinking about data practices, which can contribute to trustworthiness of such meta-infrastructures.

Healthcare data quality assessment for improving the quality of the Korea Biobank Network

Numerous studies make extensive use of healthcare data, including human materials and clinical information, and acknowledge its significance. However, limitations in data collection methods can impact the quality of healthcare data obtained from multiple institutions. In order to secure high-quality data related to human materials, research focused on data quality is necessary. This study validated the quality of data collected in 2020 from 16 institutions constituting the Korea Biobank Network using 104 validation rules. The validation rules were developed based on the DQ4HEALTH model and were divided into four dimensions: completeness, validity, accuracy, and uniqueness. Korea Biobank Network collects and manages human materials and clinical information from multiple biobanks, and is in the process of developing a common data model for data integration. The results of the data quality verification revealed an error rate of 0.74%. Furthermore, an analysis of the data from each institution was performed to examine the relationship between the institution's characteristics and error count. The results from a chi-square test indicated that there was an independent correlation between each institution and its error count. To confirm this correlation between error counts and the characteristics of each institution, a correlation analysis was conducted. The results, shown in a graph, revealed the relationship between factors that had high correlation coefficients and the error count. The findings suggest that the data quality was impacted by biases in the evaluation system, including the institution's IT environment, infrastructure, and the number of collected samples. These results highlight the need to consider the scalability of research quality when evaluating clinical epidemiological information linked to human materials in future validation studies of data quality.

Feasibility of using intermittent active monitoring of vital signs by smartphone users to predict SARS-CoV-2 PCR positivity

Early detection of highly infectious respiratory diseases, such as COVID-19, can help curb their transmission. Consequently, there is demand for easy-to-use population-based screening tools, such as mobile health applications. Here, we describe a proof-of-concept development of a machine learning classifier for the prediction of a symptomatic respiratory disease, such as COVID-19, using smartphone-collected vital sign measurements. The Fenland App study followed 2199 UK participants that provided measurements of blood oxygen saturation, body temperature, and resting heart rate. Total of 77 positive and 6339 negative SARS-CoV-2 PCR tests were recorded. An optimal classifier to identify these positive cases was selected using an automated hyperparameter optimisation. The optimised model achieved an ROC AUC of 0.695 ± 0.045. The data collection window for determining each participant's vital sign baseline was increased from 4 to 8 or 12 weeks with no significant difference in model performance (F(2) = 0.80, p = 0.472). We demonstrate that 4 weeks of intermittently collected vital sign measurements could be used to predict SARS-CoV-2 PCR positivity, with applicability to other diseases causing similar vital sign changes. This is the first example of an accessible, smartphone-based remote monitoring tool deployable in a public health setting to screen for potential infections.

Artificial intelligence-driven prediction of COVID-19-related hospitalization and death: a systematic review

Aim

To perform a systematic review on the use of Artificial Intelligence (AI) techniques for predicting COVID-19 hospitalization and mortality using primary and secondary data sources.

Study eligibility criteria

Cohort, clinical trials, meta-analyses, and observational studies investigating COVID-19 hospitalization or mortality using artificial intelligence techniques were eligible. Articles without a full text available in the English language were excluded.

Data sources

Articles recorded in Ovid MEDLINE from 01/01/2019 to 22/08/2022 were screened.

Data extraction

We extracted information on data sources, AI models, and epidemiological aspects of retrieved studies.

Bias assessment

A bias assessment of AI models was done using PROBAST.

Participants

Patients tested positive for COVID-19.

Results

We included 39 studies related to AI-based prediction of hospitalization and death related to COVID-19. The articles were published in the period 2019-2022, and mostly used Random Forest as the model with the best performance. AI models were trained using cohorts of individuals sampled from populations of European and non-European countries, mostly with cohort sample size <5,000. Data collection generally included information on demographics, clinical records, laboratory results, and pharmacological treatments (i.e., high-dimensional datasets). In most studies, the models were internally validated with cross-validation, but the majority of studies lacked external validation and calibration. Covariates were not prioritized using ensemble approaches in most of the studies, however, models still showed moderately good performances with Area under the Receiver operating characteristic Curve (AUC) values >0.7. According to the assessment with PROBAST, all models had a high risk of bias and/or concern regarding applicability.

Conclusions

A broad range of AI techniques have been used to predict COVID-19 hospitalization and mortality. The studies reported good prediction performance of AI models, however, high risk of bias and/or concern regarding applicability were detected.

Diagnosis of COVID-19 Using Chest X-ray Images and Disease Symptoms Based on Stacking Ensemble Deep Learning

The COVID-19 virus is one of the most devastating illnesses humanity has ever faced. COVID-19 is an infection that is hard to diagnose until it has caused lung damage or blood clots. As a result, it is one of the most insidious diseases due to the lack of knowledge of its symptoms. Artificial intelligence (AI) technologies are being investigated for the early detection of COVID-19 using symptoms and chest X-ray images. Therefore, this work proposes stacking ensemble models using two types of COVID-19 datasets, symptoms and chest X-ray scans, to identify COVID-19. The first proposed model is a stacking ensemble model that is merged from the outputs of pre-trained models in the stacking: multi-layer perceptron (MLP), recurrent neural network (RNN), long short-term memory (LSTM), and gated recurrent unit (GRU). Stacking trains and evaluates the meta-learner as a support vector machine (SVM) to predict the final decision. Two datasets of COVID-19 symptoms are used to compare the first proposed model with MLP, RNN, LSTM, and GRU models. The second proposed model is a stacking ensemble model that is merged from the outputs of pre-trained DL models in the stacking: VGG16, InceptionV3, Resnet50, and DenseNet121; it uses stacking to train and evaluate the meta-learner (SVM) to identify the final prediction. Two datasets of COVID-19 chest X-ray images are used to compare the second proposed model with other DL models. The result has shown that the proposed models achieve the highest performance compared to other models for each dataset.

Comorbid asthma decreased the risk for COVID-19 mortality in the United Kingdom: Evidence based on a meta-analysis

The study aimed to investigate the influence of comorbid asthma on the risk for mortality among patients with coronavirus disease 2019 (COVID-19) in the United Kingdom (UK) by utilizing a quantitative meta-analysis. The pooled odds ratio (OR) with 95% confidence interval (CI) was estimated by conducting a random-effects model. Sensitivity analysis, I² statistic, meta-regression, subgroup analysis, Begg's analysis and Egger's analysis were all implemented. Our results presented that comorbid asthma was significantly related to a decreased risk for COVID-19 mortality in the UK based on 24 eligible studies with 1,209,675 COVID-19 patients (pooled OR = 0.81, 95% CI: 0.71-0.93; I² = 89.2%, P < 0.01). Coming through further meta-regression to seek the possible cause of heterogeneity, none of elements might be responsible for heterogeneity. A sensitivity analysis proved the stability and reliability of the overall results. Both Begg's analysis (P = 1.000) and Egger's analysis (P = 0.271) manifested that publication bias did not exist. In conclusion, our data demonstrated that COVID-19 patients with comorbid asthma might bear a lower risk for mortality in the UK. Furthermore, routine intervention and treatment of asthma patients with severe acute respiratory syndrome coronavirus 2 infection should be continued in the UK.

Data sharing to advance gene-targeted therapies in rare diseases

Recent advancements in gene-targeted therapies have highlighted the critical role data sharing plays in successful translational drug development for people with rare diseases. To scale these efforts, we need to systematize these sharing principles, creating opportunities for more rapid, efficient, and scalable drug discovery/testing including long-term and transparent assessment of clinical safety and efficacy. A number of challenges will need to be addressed, including the logistical difficulties of studying rare diseases affecting individuals who may be scattered across the globe, scientific, technical, regulatory, and ethical complexities of data collection, and harmonization and integration across multiple platforms and contexts. The NCATS/NIH Gene-Targeted Therapies: Early Diagnosis and Equitable Delivery meeting series held during June 2021 included data sharing models that address these issues and framed discussions of areas that require improvement. This article describes these discussions and provides a series of considerations for future data sharing.

Generalizable machine learning approach for COVID-19 mortality risk prediction using on-admission clinical and laboratory features

We aimed to propose a mortality risk prediction model using on-admission clinical and laboratory predictors. We used a dataset of confirmed COVID-19 patients admitted to three general hospitals in Tehran. Clinical and laboratory values were gathered on admission. Six different machine learning models and two feature selection methods were used to assess the risk of in-hospital mortality. The proposed model was selected using the area under the receiver operator curve (AUC). Furthermore, a dataset from an additional hospital was used for external validation. 5320 hospitalized COVID-19 patients were enrolled in the study, with a mortality rate of 17.24% (N = 917). Among 82 features, ten laboratories and 27 clinical features were selected by LASSO. All methods showed acceptable performance (AUC > 80%), except for K-nearest neighbor. Our proposed deep neural network on features selected by LASSO showed AUC scores of 83.4% and 82.8% in internal and external validation, respectively. Furthermore, our imputer worked efficiently when two out of ten laboratory parameters were missing (AUC = 81.8%). We worked intimately with healthcare professionals to provide a tool that can solve real-world needs. Our model confirmed the potential of machine learning methods for use in clinical practice as a decision-support system.

A predictive model for hospitalization and survival to COVID-19 in a retrospective population-based study

The development of tools that provide early triage of COVID-19 patients with minimal use of diagnostic tests, based on easily accessible data, can be of vital importance in reducing COVID-19 mortality rates during high-incidence scenarios. This work proposes a machine learning model to predict mortality and risk of hospitalization using both 2 simple demographic features and 19 comorbidities obtained from 86,867 electronic medical records of COVID-19 patients, and a new method (LR-IPIP) designed to deal with data imbalance problems. The model was able to predict with high accuracy (90-93%, ROC-AUC = 0.94) the patient's final status (deceased or discharged), while its accuracy was medium (71-73%, ROC-AUC = 0.75) with respect to the risk of hospitalization. The most relevant characteristics for these models were age, sex, number of comorbidities, osteoarthritis, obesity, depression, and renal failure. Finally, to facilitate its use by clinicians, a user-friendly website has been developed ( https://alejandrocisterna.shinyapps.io/PROVIA ).

A Survey on the Role of Artificial Intelligence in Biobanking Studies: A Systematic Review

Introduction: In biobanks, participants' biological samples are stored for future research. The application of artificial intelligence (AI) involves the analysis of data and the prediction of any pathological outcomes. In AI, models are used to diagnose diseases as well as classify and predict disease risks. Our research analyzed AI's role in the development of biobanks in the healthcare industry, systematically. Methods: The literature search was conducted using three digital reference databases, namely PubMed, CINAHL, and WoS. Guidelines for preferred reporting elements for systematic reviews and meta-analyses (PRISMA)-2020 in conducting the systematic review were followed. The search terms included "biobanks", "AI", "machine learning", and "deep learning", as well as combinations such as "biobanks with AI", "deep learning in the biobanking field", and "recent advances in biobanking". Only English-language papers were included in the study, and to assess the quality of selected works, the Newcastle-Ottawa scale (NOS) was used. The good quality range (NOS ≥ 7) is only considered for further review. Results: A literature analysis of the above entries resulted in 239 studies. Based on their relevance to the study's goal, research characteristics, and NOS criteria, we included 18 articles for reviewing. In the last decade, biobanks and artificial intelligence have had a relatively large impact on the medical system. Interestingly, UK biobanks account for the highest percentage of high-quality works, followed by Qatar, South Korea, Singapore, Japan, and Denmark. Conclusions: Translational bioinformatics probably represent a future leader in precision medicine. AI and machine learning applications to biobanking research may contribute to the development of biobanks for the utility of health services and citizens.

On Comparing Cross-Validated Forecasting Models with a Novel Fuzzy-TOPSIS Metric: A COVID-19 Case Study

Time series cross-validation is a technique to select forecasting models. Despite the sophistication of cross-validation over single test/training splits, traditional and independent metrics, such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), are commonly used to assess the model’s accuracy. However, what if decision-makers have different models fitting expectations to each moment of a time series? What if the precision of the forecasted values is also important? This is the case of predicting COVID-19 in Amapá, a Brazilian state in the Amazon rainforest. Due to the lack of hospital capacities, a model that promptly and precisely responds to notable ups and downs in the number of cases may be more desired than average models that only have good performances in more frequent and calm circumstances. In line with this, this paper proposes a hybridization of the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and fuzzy sets to create a similarity metric, the closeness coefficient (CC), that enables relative comparisons of forecasting models under heterogeneous fitting expectations and also considers volatility in the predictions. We present a case study using three parametric and three machine learning models commonly used to forecast COVID-19 numbers. The results indicate that the introduced fuzzy similarity metric is a more informative performance assessment metric, especially when using time series cross-validation.