Newborn Screening for High-Risk Congenital Heart Disease by Dried Blood Spot Biomarker Analysis.

Importance: Congenital heart disease (CHD) is the most common human organ malformation, affecting approximately 1 of 125 newborns globally. Objectives: Assessing the performance of 2 diagnostic tests using minimal amounts of dried blood spots (DBS) to identify high-risk CHD compared with controls in a Swedish cohort of neonates. Design, Setting, and Participants: This diagnostic study took place in Sweden between 2019 and 2023 and enrolled full-term babies born between 2005 and 2023. All cases were identified through centralized pediatric cardiothoracic surgical services in Lund and Gothenburg, Sweden. Controls were followed up for 1 year to ensure no late presentations of high-risk CHD occurred. Cases were verified through surgical records and echocardiography. Exposure: High-risk CHD, defined as cases requiring cardiac surgical management during infancy due to evolving signs of heart failure or types in which the postnatal circulation depends on patency of the arterial duct. Using 3-µL DBS samples, automated quantitative tests for NT-proBNP and interleukin 1 receptor-like 1 (IL-1 RL1; formerly known as soluble ST2) were compared against established CHD screening methods. Main Outcomes and Measures: Performance of DBS tests to detect high-risk CHD using receiver operating characteristic curves; Bland-Altman and Pearson correlation analyses to compare IL-1 RL1 DBS with plasma blood levels. Results: A total of 313 newborns were included (mean [SD] gestational age, 39.4 [1.3] weeks; 181 [57.8%] male). Mean (SD) birthweight was 3495 (483) grams. Analyzed DBS samples included 217 CHD cases and 96 controls. Among the CHD cases, 188 participants (89.3%) were high-risk types, of which 73 (38.8%) were suspected prenatally. Of the 188 high-risk cases, 94 (50.0%) passed pulse oximetry screening and 36 (19.1%) were initially discharged after birth without diagnoses. Combining NT-proBNP and IL-1 RL1 tests performed well in comparison with existing screening methods and enabled additional identification of asymptomatic babies with receiver operating characteristic area under the curve 0.95 (95% CI, 0.93-0.98). Conclusions and relevance: In this diagnostic study, NT-proBNP and IL-1 RL1 DBS assays identified high-risk CHD in a timely manner, including in asymptomatic newborns, and improved overall screening performance in this cohort from Sweden. Prospective evaluation of this novel approach is warranted.

Adaptive risk prediction system with incremental and transfer learning.

Currently, popular methods for prenatal risk assessment of fetal aneuploidies are based on multivariate probabilistic modelling, that are built on decades of scientific research and large-scale multi-center clinical studies. These static models that are deployed to screening labs are rarely updated or adapted to local population characteristics. In this article, we propose an adaptive risk prediction system or ARPS, which considers these changing characteristics and automatically deploys updated risk models. 8 years of real-life Down syndrome screening data was used to firstly develop a distribution shift detection method that captures significant changes in the patient population and secondly a probabilistic risk modelling system that adapts to new data when these changes are detected. Various candidate systems that utilize transfer -and incremental learning that implement different levels of plasticity were tested. Distribution shift detection using a windowed approach provides a computationally less expensive alternative to fitting models at every data block step while not sacrificing performance. This was possible when utilizing transfer learning. Deploying an ARPS to a lab requires careful consideration of the parameters regarding the distribution shift detection and model updating, as they are affected by lab throughput and the incidence of the screened rare disorder. When this is done, ARPS could be also utilized for other population screening problems. We demonstrate with a large real-life dataset that our best performing novel Incremental-Learning-Population-to-Population-Transfer-Learning design can achieve on par prediction performance without human intervention, when compared to a deployed risk screening algorithm that has been manually updated over several years.

Evaluation of Circulating Cardiovascular Biomarker Levels for Early Detection of Congenital Heart Disease in Newborns in Sweden.

Importance: Congenital heart disease (CHD) is the most common congenital malformation in humans worldwide. Circulating cardiovascular biomarkers could potentially improve the early detection of CHD, even in asymptomatic newborns. Objectives: To assess the performance of a dried blood spot (DBS) test to measure the cardiovascular biomarker amino terminal fragment of the prohormone brain-type natriuretic peptide (NT-proBNP) levels in newborns and to compare DBS with standard EDTA analysis in control newborns during the first week of life. Design, Setting, and Participants: This diagnostic study was conducted in a single regional pediatric service in southern Sweden. Healthy, term neonates born between July 1, 2018, and May 31, 2019, were prospectively enrolled and compared against retrospectively identified newborns with CHD born between September 1, 2003, and September 30, 2019. Neonates who required inpatient treatment beyond the standard postnatal care were excluded. Exposure: New DBS test for NT-proBNP quantification in newborns that used 3 µL of blood vs the current screening standard. Main Outcomes and Measures: Performance of the new test and when combined with pulse oximetry screening was measured by receiver operating characteristic curve analysis. Performance of the new test and EDTA screening was compared using Pearson linear correlation analysis. Results: The DBS samples of 115 neonates (81 control newborns and 34 newborns with CHD, of whom 63 were boys [55%] and the mean [SD] gestational age was 39.6 [1.4] weeks) were analyzed. The new NT-proBNP test alone identified 71% (n = 24 of 34) of all CHD cases and 68% (n = 13 of 19) of critical CHD cases as soon as 2 days after birth. Detection of any CHD type improved to 82% (n = 28 of 34 newborns) and detection of critical CHD improved to 89% (n = 17 of 19 newborns) when combined pulse oximetry screening and NT-proBNP test results were used. Performance of the NT-proBNP test was excellent when control newborns were matched to newborns with CHD born between July 1, 2018, and May 31, 2019 (area under the curve, 0.96; SE, 0.027; 95% CI, 0.908-1.0; asymptotic P < .05). Conclusions and Relevance: This study found that NT-proBNP assay using minimal DBS samples appears to be timely and accurate in detecting CHD in newborns and to discriminate well between healthy newborns and newborns with various types of CHD. This finding warrants further studies in larger cohorts and highlights the potential of NT-proBNP to improve neonatal CHD screening.

Synthetic minority oversampling of vital statistics data with generative adversarial networks.

OBJECTIVE: Minority oversampling is a standard approach used for adjusting the ratio between the classes on imbalanced data. However, established methods often provide modest improvements in classification performance when applied to data with extremely imbalanced class distribution and to mixed-type data. This is usual for vital statistics data, in which the outcome incidence dictates the amount of positive observations. In this article, we developed a novel neural network-based oversampling method called actGAN (activation-specific generative adversarial network) that can derive useful synthetic observations in terms of increasing prediction performance in this context. MATERIALS AND METHODS: From vital statistics data, the outcome of early stillbirth was chosen to be predicted based on demographics, pregnancy history, and infections. The data contained 363 560 live births and 139 early stillbirths, resulting in class imbalance of 99.96% and 0.04%. The hyperparameters of actGAN and a baseline method SMOTE-NC (Synthetic Minority Over-sampling Technique-Nominal Continuous) were tuned with Bayesian optimization, and both were compared against a cost-sensitive learning-only approach. RESULTS: While SMOTE-NC provided mixed results, actGAN was able to improve true positive rate at a clinically significant false positive rate and area under the curve from the receiver-operating characteristic curve consistently. DISCUSSION: Including an activation-specific output layer to a generator network of actGAN enables the addition of information about the underlying data structure, which overperforms the nominal mechanism of SMOTE-NC. CONCLUSIONS: actGAN provides an improvement to the prediction performance for our learning task. Our developed method could be applied to other mixed-type data prediction tasks that are known to be afflicted by class imbalance and limited data availability.

Predicting risk of stillbirth and preterm pregnancies with machine learning.

Modelling the risk of abnormal pregnancy-related outcomes such as stillbirth and preterm birth have been proposed in the past. Commonly they utilize maternal demographic and medical history information as predictors, and they are based on conventional statistical modelling techniques. In this study, we utilize state-of-the-art machine learning methods in the task of predicting early stillbirth, late stillbirth and preterm birth pregnancies. The aim of this experimentation is to discover novel risk models that could be utilized in a clinical setting. A CDC data set of almost sixteen million observations was used conduct feature selection, parameter optimization and verification of proposed models. An additional NYC data set was used for external validation. Algorithms such as logistic regression, artificial neural network and gradient boosting decision tree were used to construct individual classifiers. Ensemble learning strategies of these classifiers were also experimented with. The best performing machine learning models achieved 0.76 AUC for early stillbirth, 0.63 for late stillbirth and 0.64 for preterm birth while using a external NYC test data. The repeatable performance of our models demonstrates robustness that is required in this context. Our proposed novel models provide a solid foundation for risk prediction and could be further improved with the addition of biochemical and/or biophysical markers.

Evaluation of machine learning algorithms for improved risk assessment for Down's syndrome.

Prenatal screening generates a great amount of data that is used for predicting risk of various disorders. Prenatal risk assessment is based on multiple clinical variables and overall performance is defined by how well the risk algorithm is optimized for the population in question. This article evaluates machine learning algorithms to improve performance of first trimester screening of Down syndrome. Machine learning algorithms pose an adaptive alternative to develop better risk assessment models using the existing clinical variables. Two real-world data sets were used to experiment with multiple classification algorithms. Implemented models were tested with a third, real-world, data set and performance was compared to a predicate method, a commercial risk assessment software. Best performing deep neural network model gave an area under the curve of 0.96 and detection rate of 78% with 1% false positive rate with the test data. Support vector machine model gave area under the curve of 0.95 and detection rate of 61% with 1% false positive rate with the same test data. When compared with the predicate method, the best support vector machine model was slightly inferior, but an optimized deep neural network model was able to give higher detection rates with same false positive rate or similar detection rate but with markedly lower false positive rate. This finding could further improve the first trimester screening for Down syndrome, by using existing clinical variables and a large training data derived from a specific population.