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

Unraveling Down Syndrome: From Genetic Anomaly to Artificial Intelligence-Enhanced Diagnosis

Down syndrome arises from chromosomal non-disjunction during gametogenesis, resulting in an additional chromosome. This anomaly presents with intellectual impairment, growth limitations, and distinct facial features. Positive correlation exists between maternal age, particularly in advanced cases, and the global annual incidence is over 200,000 cases. Early interventions, including first and second-trimester screenings, have improved DS diagnosis and care. The manifestations of Down syndrome result from complex interactions between genetic factors linked to various health concerns. To explore recent advancements in Down syndrome research, we focus on the integration of artificial intelligence (AI) and machine learning (ML) technologies for improved diagnosis and management. Recent developments leverage AI and ML algorithms to detect subtle Down syndrome indicators across various data sources, including biological markers, facial traits, and medical images. These technologies offer potential enhancements in accuracy, particularly in cases complicated by cognitive impairments. Integration of AI and ML in Down syndrome diagnosis signifies a significant advancement in medical science. These tools hold promise for early detection, personalized treatment, and a deeper comprehension of the complex interplay between genetics and environmental factors. This review provides a comprehensive overview of neurodevelopmental and cognitive profiles, comorbidities, diagnosis, and management within the Down syndrome context. The utilization of AI and ML represents a transformative step toward enhancing early identification and tailored interventions for individuals with Down syndrome, ultimately improving their quality of life.

Machine learning based analysis for intellectual disability in Down syndrome

Down syndrome (DS) or trisomy 21 is the most common genetic cause of intellectual disability (ID), but a pathogenic mechanism has not been identified yet. Studying a complex and not monogenic condition such as DS, a clear correlation between cause and effect might be difficult to find through classical analysis methods, thus different approaches need to be used. The increased availability of big data has made the use of artificial intelligence (AI) and in particular machine learning (ML) in the medical field possible. The purpose of this work is the application of ML techniques to provide an analysis of clinical records obtained from subjects with DS and study their association with ID. We have applied two tree-based ML models (random forest and gradient boosting machine) to the research question: how to identify key features likely associated with ID in DS. We analyzed 109 features (or variables) in 106 DS subjects. The outcome of the analysis was the age equivalent (AE) score as indicator of intellectual functioning, impaired in ID. We applied several methods to configure the models: feature selection through Boruta framework to minimize random correlation; data augmentation to overcome the issue of a small dataset; age effect mitigation to take into account the chronological age of the subjects. The results show that ML algorithms can be applied with good accuracy to identify variables likely involved in cognitive impairment in DS. In particular, we show how random forest and gradient boosting machine produce results with low error (MSE <0.12) and an acceptable R2 (0.70 and 0.93). Interestingly, the ranking of the variables point to several features of interest related to hearing, gastrointestinal alterations, thyroid state, immune system and vitamin B12 that can be considered with particular attention for improving care pathways for people with DS. In conclusion, ML-based model may assist researchers in identifying key features likely correlated with ID in DS, and ultimately, may improve research efforts focused on the identification of possible therapeutic targets and new care pathways. We believe this study can be the basis for further testing/validating of our algorithms with multiple and larger datasets.

Pregnant women's knowledge of birth defects and their associated factors among antenatal care attendees in referral hospitals of Amhara regional state, Ethiopia, in 2019

Background

Birth defects (BDs) are structural, behavioral, functional, and metabolic disorders present at birth. Due to lack of knowledge, families and communities stigmatized pregnant women following the birth of a child with birth defects. In Ethiopia, there was limited evidence to assess the level of knowledge among pregnant women despite increasing magnitude of birth defects.

Objectives

This study aims to assess pregnant women's knowledge of birth defects and its associated factors among antenatal care (ANC) attendees in referral hospitals of Amhara regional state in 2019.

Materials and methods

Between 1 June and 30 June 2019, 636 pregnant women receiving prenatal care participated in an institution-based cross-sectional study. The approach for sampling was multistage. A semi-structured pretested interviewer-administered questionnaire was used to collect data. Data were entered in EpiData version 4.6 and analyzed using SPSS version 25 software. A bivariable and multivariable logistic regression model was used. Odds ratio with 95% confidence interval and p-value of ≤0.05 declared statistical significance association.

Results

A total of 636 pregnant women were included in the analysis. Accordingly, pregnant women's knowledge of birth defects was found to be 49.2% (95% CI: 45.4-53.1). Age group of <25 years (AOR = 0.16, 95% CI: 0.04-0.61), urban residence (AOR = 6.06, 95% CI: 2.17-16.94), ANC booked before 20 weeks of gestational age (AOR = 3.42, 95% CI: 1.37-8.54), and ever heard on birth defects (AOR = 5.00, 95% CI: 1.87-13.43) were significantly associated factors with pregnant women's knowledge of birth defects.

Conclusions

Approximately half of the pregnant mothers were aware of birth defects. Addressing pre-pregnancy and pregnancy health information and education particularly on the prevention of birth defects is recommended.

Bringing machine learning to research on intellectual and developmental disabilities: taking inspiration from neurological diseases

Intellectual and Developmental Disabilities (IDDs), such as Down syndrome, Fragile X syndrome, Rett syndrome, and autism spectrum disorder, usually manifest at birth or early childhood. IDDs are characterized by significant impairment in intellectual and adaptive functioning, and both genetic and environmental factors underpin IDD biology. Molecular and genetic stratification of IDDs remain challenging mainly due to overlapping factors and comorbidity. Advances in high throughput sequencing, imaging, and tools to record behavioral data at scale have greatly enhanced our understanding of the molecular, cellular, structural, and environmental basis of some IDDs. Fueled by the "big data" revolution, artificial intelligence (AI) and machine learning (ML) technologies have brought a whole new paradigm shift in computational biology. Evidently, the ML-driven approach to clinical diagnoses has the potential to augment classical methods that use symptoms and external observations, hoping to push the personalized treatment plan forward. Therefore, integrative analyses and applications of ML technology have a direct bearing on discoveries in IDDs. The application of ML to IDDs can potentially improve screening and early diagnosis, advance our understanding of the complexity of comorbidity, and accelerate the identification of biomarkers for clinical research and drug development. For more than five decades, the IDDRC network has supported a nexus of investigators at centers across the USA, all striving to understand the interplay between various factors underlying IDDs. In this review, we introduced fast-increasing multi-modal data types, highlighted example studies that employed ML technologies to illuminate factors and biological mechanisms underlying IDDs, as well as recent advances in ML technologies and their applications to IDDs and other neurological diseases. We discussed various molecular, clinical, and environmental data collection modes, including genetic, imaging, phenotypical, and behavioral data types, along with multiple repositories that store and share such data. Furthermore, we outlined some fundamental concepts of machine learning algorithms and presented our opinion on specific gaps that will need to be filled to accomplish, for example, reliable implementation of ML-based diagnosis technology in IDD clinics. We anticipate that this review will guide researchers to formulate AI and ML-based approaches to investigate IDDs and related conditions.

Machine learning-based evaluation of application value of the USM combined with NIPT in the diagnosis of fetal chromosomal abnormalities

OBJECTIVE

To explore the soft ultrasound marker (USM) combined with non-invasive prenatal testing (NIPT) in diagnosing fetal chromosomal abnormalities based on machine learning and data mining techniques.

METHODS

To analyze the data of ultrasonic examination from 856 cases with high-risk single pregnancy during early and middle pregnancy stage. NIPT was applied in 642 patients. All 856 patients accepted amniocentesis and chromosome karyotype analysis to determine the efficacy of USM, Down's syndrome screening, and NIPT in detecting fetal chromosomal abnormalities.

RESULTS

Among the 856 fetuses, 129 fetuses (15.07%) with single positive USM and 36 fetuses (4.21%) with two or more positive USM. There were 81 fetuses (9.46%) with chromosomal abnormalities. In the group with multiple USM, chromosomal abnormalities were found in 36.11% of them. It was higher than the group without USM, which was 6.22% (P < 0.01), and the group with just a single USM (19.38%, P < 0.05). The sensitivity, specificity and accuracy were 96.72%, 98.45% and 98.29% when the combination of USM, Down's syndrome screening and NIPT was used to diagnose fetal chromosomal abnormalities further evaluating the accuracy and effectiveness of the above diagnostic criteria and methods with mainstream Classifiers based evaluation indicators of accuracy, f1 score, AUC.

CONCLUSIONS

The combination of USM, Down's syndrome screening and NIPT is valuable for the diagnosis of fetal chromosomal abnormalities.

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.

A machine learning model for the prediction of down syndrome in second trimester antenatal screening

BACKGROUND

Down syndrome (DS) is the most common human chromosomal abnormality. About 1200 laboratories carry out antenatal screening for DS in second trimester pregnancies in China. Their prenatal assessment of DS pregnancy risk is based on biometric calculations conducted on maternal serum biochemical markers and ultrasonic markers of fetal growth. However, the performance of this triple test for DS in second trimester pregnancies has a false positive rate of 5%, and a detection rate of about 60%∼65%.

METHOD

A total of 58,972 pregnant women, including 49 DS cases, who had undergone DS screening in the second trimester were retrospectively included and a machine learning (ML) model based on random forest was built to predict DS. In addition, the model was applied to another hospital data set of 27,170 pregnant women, including 27 DS cases, to verify the predictive efficiency of the model.

RESULTS

The ML model gave a DS detection rate of 66.7%, with a 5% false positive rate in the model data set. In the external verification data set, the ML model achieved a DS detection rate of 85.2%, with a 5% false positive rate . In comparison with the current laboratory risk model, the ML model improves the DS detection rate with the same false positive rate, while the difference has no significance.

CONCLUSIONS

The ML model for DS detection described here has a comparable detection rate with the same false positive rate as the DS risk screening software currently used in China. Our ML model exhibited robust performance and good extrapolation, and could function as an alternative tool for DS risk assessment in second trimester maternal serum.

Diagnostic Accuracy of Machine Learning Models to Identify Congenital Heart Disease: A Meta-Analysis

Background: With the dearth of trained care providers to diagnose congenital heart disease (CHD) and a surge in machine learning (ML) models, this review aims to estimate the diagnostic accuracy of such models for detecting CHD. Methods: A comprehensive literature search in the PubMed, CINAHL, Wiley Cochrane Library, and Web of Science databases was performed. Studies that reported the diagnostic ability of ML for the detection of CHD compared to the reference standard were included. Risk of bias assessment was performed using Quality Assessment for Diagnostic Accuracy Studies-2 tool. The sensitivity and specificity results from the studies were used to generate the hierarchical Summary ROC (HSROC) curve. Results: We included 16 studies (1217 participants) that used ML algorithm to diagnose CHD. Neural networks were used in seven studies with overall sensitivity of 90.9% (95% CI 85.2-94.5%) and specificity was 92.7% (95% CI 86.4-96.2%). Other ML models included ensemble methods, deep learning and clustering techniques but did not have sufficient number of studies for a meta-analysis. Majority (n=11, 69%) of studies had a high risk of patient selection bias, unclear bias on index test (n=9, 56%) and flow and timing (n=12, 75%) while low risk of bias was reported for the reference standard (n=10, 62%). Conclusion: ML models such as neural networks have the potential to diagnose CHD accurately without the need for trained personnel. The heterogeneity of the diagnostic modalities used to train these models and the heterogeneity of the CHD diagnoses included between the studies is a major limitation.

Application of intelligent algorithms in Down syndrome screening during second trimester pregnancy

BACKGROUND

Down syndrome (DS) is one of the most common chromosomal aneuploidy diseases. Prenatal screening and diagnostic tests can aid the early diagnosis, appropriate management of these fetuses, and give parents an informed choice about whether or not to terminate a pregnancy. In recent years, investigations have been conducted to achieve a high detection rate (DR) and reduce the false positive rate (FPR). Hospitals have accumulated large numbers of screened cases. However, artificial intelligence methods are rarely used in the risk assessment of prenatal screening for DS.

AIM

To use a support vector machine algorithm, classification and regression tree algorithm, and AdaBoost algorithm in machine learning for modeling and analysis of prenatal DS screening.

METHODS

The dataset was from the Center for Prenatal Diagnosis at the First Hospital of Jilin University. We designed and developed intelligent algorithms based on the synthetic minority over-sampling technique (SMOTE)-Tomek and adaptive synthetic sampling over-sampling techniques to preprocess the dataset of prenatal screening information. The machine learning model was then established. Finally, the feasibility of artificial intelligence algorithms in DS screening evaluation is discussed.

RESULTS

The database contained 31 DS diagnosed cases, accounting for 0.03% of all patients. The dataset showed a large difference between the numbers of DS affected and non-affected cases. A combination of over-sampling and under-sampling techniques can greatly increase the performance of the algorithm at processing non-balanced datasets. As the number of iterations increases, the combination of the classification and regression tree algorithm and the SMOTE-Tomek over-sampling technique can obtain a high DR while keeping the FPR to a minimum.

CONCLUSION

The support vector machine algorithm and the classification and regression tree algorithm achieved good results on the DS screening dataset. When the T21 risk cutoff value was set to 270, machine learning methods had a higher DR and a lower FPR than statistical methods.

Impact of replacing or adding placental growth factor on Down syndrome screening: A prospective cohort study

OBJECTIVES

To assess whether adding placental growth factor (PlGF) or replacing pregnancy-associated plasma protein-A (PAPP-A) improves the first trimester combined test performance for trisomy 21.

METHODS

A total of 11,518 women with a singleton pregnancy who underwent the first trimester combined test between December 2016 and December 2019 were included. PlGF was measured and estimated term risk for trisomy 21 was calculated by (1) adding PlGF to the combined test and (2) replacing PAPP-A with PlGF.

RESULTS

Twenty-nine pregnancies had trisomy 21. The combined tests detection rate (DR), false positive rate (FPR) and screen positive rate (SPR) were 89.7%, 5.7% and 6% respectively. DR when adding PlGF to the combined test or replacing PAPP-A remained unchanged. Replacing PAPP-A by PlGF increased FPR and SPR to 6.2% and 6.4% respectively. Adding PlGF to the combined test gave FPR and SPR rates of 5.5% and 5.7% respectively. Change in FPR and SPR was not significant (p > 0.1 for all).

CONCLUSION

Adding PlGF to the combined test or replacing PAPP-A with PlGF did not improve trisomy 21 DR and resulted in a non-significant marginal change in FPR and SPR.

Towards deep phenotyping pregnancy: a systematic review on artificial intelligence and machine learning methods to improve pregnancy outcomes

Objective

Development of novel informatics methods focused on improving pregnancy outcomes remains an active area of research. The purpose of this study is to systematically review the ways that artificial intelligence (AI) and machine learning (ML), including deep learning (DL), methodologies can inform patient care during pregnancy and improve outcomes.

Materials and methods

We searched English articles on EMBASE, PubMed and SCOPUS. Search terms included ML, AI, pregnancy and informatics. We included research articles and book chapters, excluding conference papers, editorials and notes.

Results

We identified 127 distinct studies from our queries that were relevant to our topic and included in the review. We found that supervised learning methods were more popular (n = 69) than unsupervised methods (n = 9). Popular methods included support vector machines (n = 30), artificial neural networks (n = 22), regression analysis (n = 17) and random forests (n = 16). Methods such as DL are beginning to gain traction (n = 13). Common areas within the pregnancy domain where AI and ML methods were used the most include prenatal care (e.g. fetal anomalies, placental functioning) (n = 73); perinatal care, birth and delivery (n = 20); and preterm birth (n = 13). Efforts to translate AI into clinical care include clinical decision support systems (n = 24) and mobile health applications (n = 9).

Conclusions

Overall, we found that ML and AI methods are being employed to optimize pregnancy outcomes, including modern DL methods (n = 13). Future research should focus on less-studied pregnancy domain areas, including postnatal and postpartum care (n = 2). Also, more work on clinical adoption of AI methods and the ethical implications of such adoption is needed.

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.

Diagnosis support systems for rare diseases: a scoping review

INTRODUCTION

Rare diseases affect approximately 350 million people worldwide. Delayed diagnosis is frequent due to lack of knowledge of most clinicians and a small number of expert centers. Consequently, computerized diagnosis support systems have been developed to address these issues, with many relying on rare disease expertise and taking advantage of the increasing volume of generated and accessible health-related data. Our objective is to perform a review of all initiatives aiming to support the diagnosis of rare diseases.

METHODS

A scoping review was conducted based on methods proposed by Arksey and O'Malley. A charting form for relevant study analysis was developed and used to categorize data.

RESULTS

Sixty-eight studies were retained at the end of the charting process. Diagnosis targets varied from 1 rare disease to all rare diseases. Material used for diagnosis support consisted mostly of phenotype concepts, images or fluids. Fifty-seven percent of the studies used expert knowledge. Two-thirds of the studies relied on machine learning algorithms, and one-third used simple similarities. Manual algorithms were encountered as well. Most of the studies presented satisfying performance of evaluation by comparison with references or with external validation. Fourteen studies provided online tools, most of which aimed to support the diagnosis of all rare diseases by considering queries based on phenotype concepts.

CONCLUSION

Numerous solutions relying on different materials and use of various methodologies are emerging with satisfying preliminary results. However, the variability of approaches and evaluation processes complicates the comparison of results. Efforts should be made to adequately validate these tools and guarantee reproducibility and explicability.