Targeted serum proteomics of longitudinal samples from newly diagnosed youth with type 1 diabetes distinguishes markers of disease and C-peptide trajectory.

AIMS/HYPOTHESIS: There is a growing need for markers that could help indicate the decline in beta cell function and recognise the need and efficacy of intervention in type 1 diabetes. Measurements of suitably selected serum markers could potentially provide a non-invasive and easily applicable solution to this challenge. Accordingly, we evaluated a broad panel of proteins previously associated with type 1 diabetes in serum from newly diagnosed individuals during the first year from diagnosis. To uncover associations with beta cell function, comparisons were made between these targeted proteomics measurements and changes in fasting C-peptide levels. To further distinguish proteins linked with the disease status, comparisons were made with measurements of the protein targets in age- and sex-matched autoantibody-negative unaffected family members (UFMs). METHODS: Selected reaction monitoring (SRM) mass spectrometry analyses of serum, targeting 85 type 1 diabetes-associated proteins, were made. Sera from individuals diagnosed under 18 years (n=86) were drawn within 6 weeks of diagnosis and at 3, 6 and 12 months afterwards (288 samples in total). The SRM data were compared with fasting C-peptide/glucose data, which was interpreted as a measure of beta cell function. The protein data were further compared with cross-sectional SRM measurements from UFMs (n=194). RESULTS: Eleven proteins had statistically significant associations with fasting C-peptide/glucose. Of these, apolipoprotein L1 and glutathione peroxidase 3 (GPX3) displayed the strongest positive and inverse associations, respectively. Changes in GPX3 levels during the first year after diagnosis indicated future fasting C-peptide/glucose levels. In addition, differences in the levels of 13 proteins were observed between the individuals with type 1 diabetes and the matched UFMs. These included GPX3, transthyretin, prothrombin, apolipoprotein C1 and members of the IGF family. CONCLUSIONS/INTERPRETATION: The association of several targeted proteins with fasting C-peptide/glucose levels in the first year after diagnosis suggests their connection with the underlying changes accompanying alterations in beta cell function in type 1 diabetes. Moreover, the direction of change in GPX3 during the first year was indicative of subsequent fasting C-peptide/glucose levels, and supports further investigation of this and other serum protein measurements in future studies of beta cell function in type 1 diabetes.

A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories.

Pancreatic cancer is an aggressive disease that typically presents late with poor outcomes, indicating a pronounced need for early detection. In this study, we applied artificial intelligence methods to clinical data from 6 million patients (24,000 pancreatic cancer cases) in Denmark (Danish National Patient Registry (DNPR)) and from 3 million patients (3,900 cases) in the United States (US Veterans Affairs (US-VA)). We trained machine learning models on the sequence of disease codes in clinical histories and tested prediction of cancer occurrence within incremental time windows (CancerRiskNet). For cancer occurrence within 36 months, the performance of the best DNPR model has area under the receiver operating characteristic (AUROC) curve = 0.88 and decreases to AUROC (3m) = 0.83 when disease events within 3 months before cancer diagnosis are excluded from training, with an estimated relative risk of 59 for 1,000 highest-risk patients older than age 50 years. Cross-application of the Danish model to US-VA data had lower performance (AUROC = 0.71), and retraining was needed to improve performance (AUROC = 0.78, AUROC (3m) = 0.76). These results improve the ability to design realistic surveillance programs for patients at elevated risk, potentially benefiting lifespan and quality of life by early detection of this aggressive cancer.

DanMAC5: a browser of aggregated sequence variants from 8,671 whole genome sequenced Danish individuals.

OBJECTIVES: Allele counts of sequence variants obtained by whole genome sequencing (WGS) often play a central role in interpreting the results of genetic and genomic research. However, such variant counts are not readily available for individuals in the Danish population. Here, we present a dataset with allele counts for sequence variants (single nucleotide variants (SNVs) and indels) identified from WGS of 8,671 (5,418 females) individuals from the Danish population. The data resource is based on WGS data from three independent research projects aimed at assessing genetic risk factors for cardiovascular, psychiatric, and headache disorders. To enable the sharing of information on sequence variation in Danish individuals, we created summarized statistics on allele counts from anonymized data and made them available through the European Genome-phenome Archive (EGA, https://identifiers.org/ega. DATASET: EGAD00001009756 ) and in a dedicated browser, DanMAC5 (available at www.danmac5.dk ). The summary level data and the DanMAC5 browser provide insight into the allelic spectrum of sequence variants segregating in the Danish population, which is important in variant interpretation. DATA DESCRIPTION: Three WGS datasets with an average coverage of 30x were processed independently using the same quality control pipeline. Subsequently, we summarized, filtered, and merged allele counts to create a high-quality summary level dataset of sequence variants.

Discrete-time survival analysis in the critically ill: a deep learning approach using heterogeneous data.

Prediction of survival for patients in intensive care units (ICUs) has been subject to intense research. However, no models exist that embrace the multiverse of data in ICUs. It is an open question whether deep learning methods using automated data integration with minimal pre-processing of mixed data domains such as free text, medical history and high-frequency data can provide discrete-time survival estimates for individual ICU patients. We trained a deep learning model on data from patients admitted to ten ICUs in the Capital Region of Denmark and the Region of Southern Denmark between 2011 and 2018. Inspired by natural language processing we mapped the electronic patient record data to an embedded representation and fed the data to a recurrent neural network with a multi-label output layer representing the chance of survival at different follow-up times. We evaluated the performance using the time-dependent concordance index. In addition, we quantified and visualized the drivers of survival predictions using the SHAP methodology. We included 37,355 admissions of 29,417 patients in our study. Our deep learning models outperformed traditional Cox proportional-hazard models with concordance index in the ranges 0.72-0.73, 0.71-0.72, 0.71, and 0.69-0.70, for models applied at baseline 0, 24, 48, and 72 h, respectively. Deep learning models based on a combination of entity embeddings and survival modelling is a feasible approach to obtain individualized survival estimates in data-rich settings such as the ICU. The interpretable nature of the models enables us to understand the impact of the different data domains.

Dynamic and explainable machine learning prediction of mortality in patients in the intensive care unit: a retrospective study of high-frequency data in electronic patient records.

BACKGROUND: Many mortality prediction models have been developed for patients in intensive care units (ICUs); most are based on data available at ICU admission. We investigated whether machine learning methods using analyses of time-series data improved mortality prognostication for patients in the ICU by providing real-time predictions of 90-day mortality. In addition, we examined to what extent such a dynamic model could be made interpretable by quantifying and visualising the features that drive the predictions at different timepoints. METHODS: Based on the Simplified Acute Physiology Score (SAPS) III variables, we trained a machine learning model on longitudinal data from patients admitted to four ICUs in the Capital Region, Denmark, between 2011 and 2016. We included all patients older than 16 years of age, with an ICU stay lasting more than 1 h, and who had a Danish civil registration number to enable 90-day follow-up. We leveraged static data and physiological time-series data from electronic health records and the Danish National Patient Registry. A recurrent neural network was trained with a temporal resolution of 1 h. The model was internally validated using the holdout method with 20% of the training dataset and externally validated using previously unseen data from a fifth hospital in Denmark. Its performance was assessed with the Matthews correlation coefficient (MCC) and area under the receiver operating characteristic curve (AUROC) as metrics, using bootstrapping with 1000 samples with replacement to construct 95% CIs. A Shapley additive explanations algorithm was applied to the prediction model to obtain explanations of the features that drive patient-specific predictions, and the contributions of each of the 44 features in the model were analysed and compared with the variables in the original SAPS III model. FINDINGS: From a dataset containing 15 615 ICU admissions of 12 616 patients, we included 14 190 admissions of 11 492 patients in our analysis. Overall, 90-day mortality was 33·1% (3802 patients). The deep learning model showed a predictive performance on the holdout testing dataset that improved over the timecourse of an ICU stay: MCC 0·29 (95% CI 0·25-0·33) and AUROC 0·73 (0·71-0·74) at admission, 0·43 (0·40-0·47) and 0·82 (0·80-0·84) after 24 h, 0·50 (0·46-0·53) and 0·85 (0·84-0·87) after 72 h, and 0·57 (0·54-0·60) and 0·88 (0·87-0·89) at the time of discharge. The model exhibited good calibration properties. These results were validated in an external validation cohort of 5827 patients with 6748 admissions: MCC 0·29 (95% CI 0·27-0·32) and AUROC 0·75 (0·73-0·76) at admission, 0·41 (0·39-0·44) and 0·80 (0·79-0·81) after 24 h, 0·46 (0·43-0·48) and 0·82 (0·81-0·83) after 72 h, and 0·47 (0·44-0·49) and 0·83 (0·82-0·84) at the time of discharge. INTERPRETATION: The prediction of 90-day mortality improved with 1-h sampling intervals during the ICU stay. The dynamic risk prediction can also be explained for an individual patient, visualising the features contributing to the prediction at any point in time. This explanation allows the clinician to determine whether there are elements in the current patient state and care that are potentially actionable, thus making the model suitable for further validation as a clinical tool. FUNDING: Novo Nordisk Foundation and the Innovation Fund Denmark.

Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records.

BACKGROUND: Intensive-care units (ICUs) treat the most critically ill patients, which is complicated by the heterogeneity of the diseases that they encounter. Severity scores based mainly on acute physiology measures collected at ICU admission are used to predict mortality, but are non-specific, and predictions for individual patients can be inaccurate. We investigated whether inclusion of long-term disease history before ICU admission improves mortality predictions. METHODS: Registry data for long-term disease histories for more than 230 000 Danish ICU patients were used in a neural network to develop an ICU mortality prediction model. Long-term disease histories and acute physiology measures were aggregated to predict mortality risk for patients for whom both registry and ICU electronic patient record data were available. We compared mortality predictions with admission scores on the Simplified Acute Physiology Score (SAPS) II, the Acute Physiologic Assessment and Chronic Health Evaluation (APACHE) II, and the best available multimorbidity score, the Multimorbidity Index. An external validation set from an additional hospital was acquired after model construction to confirm the validity of our model. During initial model development data were split into a training set (85%) and an independent test set (15%), and a five-fold cross-validation was done during training to avoid overfitting. Neural networks were trained for datasets with disease history of 1 month, 3 months, 6 months, 1 year, 2·5 years, 5 years, 7·5 years, 10 years, and 23 years before ICU admission. FINDINGS: Mortality predictions with a model based solely on disease history outperformed the Multimorbidity Index (Matthews correlation coefficient 0·265 vs 0·065), and performed similarly to SAPS II and APACHE II (Matthews correlation coefficient with disease history, age, and sex 0·326 vs 0·347 and 0·300 for SAPS II and APACHE II, respectively). Diagnoses up to 10 years before ICU admission affected current mortality prediction. Aggregation of previous disease history and acute physiology measures in a neural network yielded the most precise predictions of in-hospital mortality (Matthews correlation coefficient 0·391 for in-hospital mortality compared with 0·347 with SAPS II and 0·300 with APACHE II). These results for the aggregated model were validated in an external independent dataset of 1528 patients (Matthews correlation coefficient for prediction of in-hospital mortality 0·341). INTERPRETATION: Longitudinal disease-spectrum-wide data available before ICU admission are useful for mortality prediction. Disease history can be used to differentiate mortality risk between patients with similar vital signs with more precision than SAPS II and APACHE II scores. Machine learning models can be deconvoluted to generate novel understandings of how ICU patient features from long-term and short-term events interact with each other. Explainable machine learning models are key in clinical settings, and our results emphasise how to progress towards the transformation of advanced models into actionable, transparent, and trustworthy clinical tools. FUNDING: Novo Nordisk Foundation and Innovation Fund Denmark.

The governance structure for data access in the DIRECT consortium: an innovative medicines initiative (IMI) project.

Biomedical research projects involving multiple partners from public and private sectors require coherent internal governance mechanisms to engender good working relationships. The DIRECT project is an example of such a venture, funded by the Innovative Medicines Initiative Joint Undertaking (IMI JU). This paper describes the data access policy that was developed within DIRECT to support data access and sharing, via the establishment of a 3-tiered Data Access Committee. The process was intended to allow quick access to data, whilst enabling strong oversight of how data were being accessed and by whom, and any subsequent analyses, to contribute to the overall objectives of the consortium.