Identification of newborns at risk for autism using electronic medical records and machine learning

Predicting autism traits from baby wellness records: A machine learning approach

LAY ABSTRACT

Timely identification of autism spectrum conditions is a necessity to enable children to receive the most benefit from early interventions. Emerging technological advancements provide avenues for detecting subtle, early indicators of autism from routinely collected health information. This study tested a model that provides a likelihood score for autism diagnosis from baby wellness visit records collected during the first 2 years of life. It included records of 591,989 non-autistic children and 12,846 children with autism. The model identified two-thirds of the autism spectrum condition group (boys 63% and girls 66%). Sex-specific models had several predictive features in common. These included language development, fine motor skills, and social milestones from visits at 12-24 months, mother's age, and lower initial growth but higher last growth measurements. Parental concerns about development or hearing impairment were other predictors. The models differed in other growth measurements and birth parameters. These models can support the detection of early signs of autism in girls and boys by using information routinely recorded during the first 2 years of life.

Exploring sensory alterations and repetitive behaviors in children with autism spectrum disorder from the perspective of artificial neural networks

BACKGROUND

Restrictive repetitive behaviors (RRBs) and sensory processing disorders are core symptoms of autism spectrum disorder (ASD). Their relationship is reported, but existing data are conflicting as to whether they are related but distinct, or different aspects of the same phenomenon.

AIMS

This study investigates this relationship using artificial neural networks (ANN) analysis and an innovative data mining analysis known as Auto Contractive Map (Auto-CM), which allows to discover hidden trends and associations among complex networks of variables (e.g. biological systems).

METHODS AND PROCEDURES

The Short Sensory Profile and the Repetitive Behavior Scale-Revised were administered to 45 ASD children's caregivers (M 78 %; F 22 %; mean age 6 years). Questionnaires' scores, clinical and demographic data were collected and analyzed applying Auto-CM, and a connectivity map was drawn.

OUTCOMES AND RESULTS

The main associations shown by the resulting maps confirm the known relationship between RBBs and sensory abnormalities, and support the existence of sensory phenotypes, and important links between RRBs and sleep disturbance in ASD.

CONCLUSIONS AND IMPLICATIONS

Our study demonstrates the usefulness of ANNs application and its easy handling to research RBBs and sensory abnormalities in ASD, with the aim to achieve better individualized rehabilitation technique and improve early diagnosis.

PAPER'S CONTRIBUTION

Restricted, repetitive patterns of behaviors and interests and alteration of sensory elaboration are core symptoms of ASD; their impact on patients' quality of life is known. This study introduces two main novelties: 1) the simultaneous and comparative use of two parent questionnaires (SSP and RBS-R) utilized for RRBs and alteration of sensory profile; 2) the application of ANNs to this kind of research. ANNs are adaptive models particularly suited for solving non-linear problems. While they have been widely used in the medical field, they have not been applied yet to the analysis of RRBs and sensory abnormalities in general, much less in children with ASD. The application of Auto Contractive Map (Auto-CM), a fourth generation ANNs analysis, to a dataset previously explored using classical statistical models, confirmed and expanded the associations emerged between SSP and RBS-R subscales and demographic-clinical variables. In particular, the Low Energy subscale has proven to be the central hub of the system; interesting links have emerged between the subscale Self-Injurious Behaviors and the variable intellectual disability and between sleep disturbance and various RRBs. Expanding research in this area aims to guide and modulate an emerging targeted and personalized rehabilitation therapy.

Predicting neurodevelopmental disorders using machine learning models and electronic health records - status of the field

Machine learning (ML) is increasingly used to identify patterns that could predict neurodevelopmental disorders (NDDs), such as autism spectrum disorder (ASD) and attention-deficit hyperactivity disorder (ADHD). One key source of multilevel data for ML prediction models includes population-based registers and electronic health records. These can contain rich information on individual and familial medical histories and socio-demographics. This review summarizes studies published between 2010-2022 that used ML algorithms to develop predictive models for NDDs using population-based registers and electronic health records. A literature search identified 1191 articles, of which 32 were retained. Of these, 47% developed ASD prediction models and 25% ADHD models. Classical ML methods were used in 82% of studies and in particular tree-based prediction models performed well. The sensitivity of the models was lower than 75% for most studies, while the area under the curve (AUC) was greater than 75%. The most important predictors were patient and familial medical history and sociodemographic factors. Using private in-house datasets makes comparing and validating model generalizability across studies difficult. The ML model development and reporting guidelines were adopted only in a few recently reported studies. More work is needed to harness the power of data for detecting NDDs early.

A deep learning model for prediction of autism status using whole-exome sequencing data

Autism is a developmental disability. Research demonstrated that children with autism benefit from early diagnosis and early intervention. Genetic factors are considered major contributors to the development of autism. Machine learning (ML), including deep learning (DL), has been evaluated in phenotype prediction, but this method has been limited in its application to autism. We developed a DL model, the Separate Translated Autism Research Neural Network (STAR-NN) model to predict autism status. The model was trained and tested using whole exome sequencing data from 43,203 individuals (16,809 individuals with autism and 26,394 non-autistic controls). Polygenic scores from common variants and the aggregated count of rare variants on genes were used as input. In STAR-NN, protein truncating variants, possibly damaging missense variants and mild effect missense variants on the same gene were separated at the input level and merged to one gene node. In this way, rare variants with different level of pathogenic effects were treated separately. We further validated the performance of STAR-NN using an independent dataset, including 13,827 individuals with autism and 14,052 non-autistic controls. STAR-NN achieved a modest ROC-AUC of 0.7319 on the testing dataset and 0.7302 on the independent dataset. STAR-NN outperformed other traditional ML models. Gene Ontology analysis on the selected gene features showed an enrichment for potentially informative pathways including calcium ion transport.

Machine Learning Differentiation of Autism Spectrum Sub-Classifications

PURPOSE

Disorders on the autism spectrum have characteristics that can manifest as difficulties with communication, executive functioning, daily living, and more. These challenges can be mitigated with early identification. However, diagnostic criteria has changed from DSM-IV to DSM-5, which can make diagnosing a disorder on the autism spectrum complex. We evaluated machine learning to classify individuals as having one of three disorders of the autism spectrum under DSM-IV, or as non-spectrum.

METHODS

We employed machine learning to analyze retrospective data from 38,560 individuals. Inputs encompassed clinical, demographic, and assessment data.

RESULTS

The algorithm achieved AUROCs ranging from 0.863 to 0.980. The model correctly classified 80.5% individuals; 12.6% of individuals from this dataset were misclassified with another disorder on the autism spectrum.

CONCLUSION

Machine learning can classify individuals as having a disorder on the autism spectrum or as non-spectrum using minimal data inputs.

A Prediction Model of Autism Spectrum Diagnosis from Well-Baby Electronic Data Using Machine Learning

Early detection of autism spectrum disorder (ASD) is crucial for timely intervention, yet diagnosis typically occurs after age three. This study aimed to develop a machine learning model to predict ASD diagnosis using infants' electronic health records obtained through a national screening program and evaluate its accuracy. A retrospective cohort study analyzed health records of 780,610 children, including 1163 with ASD diagnoses. Data encompassed birth parameters, growth metrics, developmental milestones, and familial and post-natal variables from routine wellness visits within the first two years. Using a gradient boosting model with 3-fold cross-validation, 100 parameters predicted ASD diagnosis with an average area under the ROC curve of 0.86 (SD < 0.002). Feature importance was quantified using the Shapley Additive explanation tool. The model identified a high-risk group with a 4.3-fold higher ASD incidence (0.006) compared to the cohort (0.001). Key predictors included failing six milestones in language, social, and fine motor domains during the second year, male gender, parental developmental concerns, non-nursing, older maternal age, lower gestational age, and atypical growth percentiles. Machine learning algorithms capitalizing on preventative care electronic health records can facilitate ASD screening considering complex relations between familial and birth factors, post-natal growth, developmental parameters, and parent concern.

Artificial intelligence for prenatal chromosome analysis

This review article delves into the rapidly advancing domain of prenatal diagnostics, with a primary focus on the detection and management of chromosomal abnormalities such as trisomy 13 ("Patau syndrome)", "trisomy 18 (Edwards syndrome)", and "trisomy 21 (Down syndrome)". The objective of the study is to examine the utilization and effectiveness of novel computational methodologies, such as "machine learning (ML)", "deep learning (DL)", and data analysis, in enhancing the detection rates and accuracy of these prenatal conditions. The contribution of the article lies in its comprehensive examination of advancements in "Non-Invasive Prenatal Testing (NIPT)", prenatal screening, genomics, and medical imaging. It highlights the potential of these techniques for prenatal diagnosis and the contributions of ML and DL to these advancements. It highlights the application of ensemble models and transfer learning to improving model performance, especially with limited datasets. This also delves into optimal feature selection and fusion of high-dimensional features, underscoring the need for future research in these areas. The review finds that ML and DL have substantially improved the detection and management of prenatal conditions, despite limitations such as small sample sizes and issues related to model generalizability. It recognizes the promising results achieved through the use of ensemble models and transfer learning in prenatal diagnostics. The review also notes the increased importance of feature selection and high-dimensional feature fusion in the development and training of predictive models. The findings underline the crucial role of AI and machine learning techniques in early detection and improved therapeutic strategies in prenatal diagnostics, highlighting a pressing need for further research in this area.

Assessment of autonomic symptom scales in patients with neurodevelopmental diagnoses using electronic health record data

Background

Sleep disturbances, gastrointestinal problems, and atypical heart rate are commonly observed in patients with autism spectrum disorder (ASD) and may relate to underlying function of the autonomic nervous system (ANS). The overall objective of the current study was to quantitatively characterize features of ANS function using symptom scales and available electronic health record (EHR) data in a clinically and genetically characterized pediatric cohort.

Methods

We assessed features of ANS function via chart review of patient records adapted from items drawn from a clinical research questionnaire of autonomic symptoms. This procedure coded for the presence and/or absence of targeted symptoms and was completed in 3 groups of patients, including patients with a clinical neurodevelopmental diagnosis and identified genetic etiology (NPD, n=244), those with an ASD diagnosis with no known genetic cause (ASD, n=159), and age and sex matched controls (MC, n=213). Symptoms were assessed across four main categories: (1) Mood, Behavior, and Emotion; (2) Secretomotor, Sensory Integration; (3) Urinary, Gastrointestinal, and Digestion; and (4) Circulation, Thermoregulation, Circadian function, and Sleep/Wake cycles.

Results

Chart review scores indicate an increased rate of autonomic symptoms across all four sections in our NPD group as compared to scores with ASD and/or MC. Additionally, we note several significant relationships between individual differences in autonomic symptoms and quantitative ASD traits.

Conclusion

These results highlight EHR review as a potentially useful method for quantifying variance in symptoms adapted from a questionnaire or survey. Further, using this method indicates that autonomic features are more prevalent in children with genetic disorders conferring risk for ASD and other neurodevelopmental diagnoses.

Assessing Autistic Traits in Toddlers Using a Data-Driven Approach with DSM-5 Mapping

Autistic spectrum disorder (ASD) is a neurodevelopmental condition that characterises a range of people, from individuals who are not able to speak to others who have good verbal communications. The disorder affects the way people see, think, and behave, including their communications and social interactions. Identifying autistic traits, preferably in the early stages, is fundamental for clinicians in expediting referrals, and hence enabling patients to access to required healthcare services. This article investigates various ASD behavioral features in toddlers and proposes a data process using machine-learning techniques. The aims of this study were to identify early behavioral features that can help detect ASD in toddlers and to map these features to the neurodevelopment behavioral areas of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). To achieve these aims, the proposed data process assesses several behavioral features using feature selection techniques, then constructs a classification model based on the chosen features. The empirical results show that during the screening process of toddlers, cognitive features related to communications, social interactions, and repetitive behaviors were most relevant to ASD. For the machine-learning algorithms, the predictive accuracy of Bayesian network (Bayes Net) and logistic regression (LR) models derived from ASD behavioral data subsets were consistent pinpointing to the suitability of ML techniques in predicting ASD.

The usefulness of machine-learning-based evaluation of clinical and pretreatment 18F-FDG-PET/CT radiomic features for predicting prognosis in patients with laryngeal cancer

OBJECTIVE

To examine whether machine learning (ML) analyses involving clinical and 18F-FDG-PET-based radiomic features are helpful in predicting prognosis in patients with laryngeal cancer.

METHODS

This retrospective study included 49 patients with laryngeal cancer who underwent18F-FDG-PET/CT before treatment, and these patients were divided into the training (n = 34) and testing (n = 15) cohorts.Seven clinical (age, sex, tumor size, T stage, N stage, Union for International Cancer Control stage, and treatment) and 40 18F-FDG-PET-based radiomic features were used to predict disease progression and survival. Six ML algorithms (random forest, neural network, k-nearest neighbors, naïve Bayes, logistic regression, and support vector machine) were used for predicting disease progression. Two ML algorithms (cox proportional hazard and random survival forest [RSF] model) considering for time-to-event outcomes were used to assess progression-free survival (PFS), and prediction performance was assessed by the concordance index (C-index).

RESULTS

Tumor size, T stage, N stage, GLZLM_ZLNU, and GLCM_Entropy were the five most important features for predicting disease progression.In both cohorts, the naïve Bayes model constructed by these five features was the best performing classifier (training: AUC = 0.805; testing: AUC = 0.842). The RSF model using the five features (tumor size, GLZLM_ZLNU, GLCM_Entropy, GLRLM_LRHGE and GLRLM_SRHGE) exhibited the highest performance in predicting PFS (training: C-index = 0.840; testing: C-index = 0.808).

CONCLUSION

ML analyses involving clinical and 18F-FDG-PET-based radiomic features may help predict disease progression and survival in patients with laryngeal cancer.

ADVANCES IN KNOWLEDGE

ML approach using clinical and 18F-FDG-PET-based radiomic features has the potential to predict prognosis of laryngeal cancer.

The Usefulness of Machine Learning-Based Evaluation of Clinical and Pretreatment [18F]-FDG-PET/CT Radiomic Features for Predicting Prognosis in Hypopharyngeal Cancer

PURPOSE

To examine whether the machine learning (ML) analyses using clinical and pretreatment 2-deoxy-2-[¹⁸F]fluoro-D-glucose positron emission tomography ([¹⁸F]-FDG-PET)-based radiomic features were useful for predicting prognosis in patients with hypopharyngeal cancer.

PROCEDURES

This retrospective study included 100 patients with hypopharyngeal cancer who underwent [¹⁸F]-FDG-PET/X-ray computed tomography (CT) before treatment, and these patients were allocated to the training (n=80) and validation (n=20) cohorts. Eight clinical (age, sex, histology, T stage, N stage, M stage, UICC stage, and treatment) and 40 [¹⁸F]-FDG-PET-based radiomic features were used to predict disease progression. A feature reduction procedure based on the decrease of the Gini impurity was applied. Six ML algorithms (random forest, neural network, k-nearest neighbors, naïve Bayes, logistic regression, and support vector machine) were compared using the area under the receiver operating characteristic curve (AUC). Progression-free survival (PFS) was assessed using Cox regression analysis.

RESULTS

The five most important features for predicting disease progression were UICC stage, N stage, gray level co-occurrence matrix entropy (GLCM_Entropy), gray level run length matrix run length non-uniformity (GLRLM_RLNU), and T stage. Patients who experienced disease progression displayed significantly higher UICC stage, N stage, GLCM_Entropy, GLRLM_RLNU, and T stage than those without progression (each, p<0.001). In both cohorts, the logistic regression model constructed by these 5 features was the best performing classifier (training: AUC=0.860, accuracy=0.800; validation: AUC=0.803, accuracy=0.700). In the logistic regression model, 5-year PFS was significantly higher in patients with predicted non-progression than those with predicted progression (75.8% vs. 8.3%, p<0.001), and this model was only the independent factor for PFS in multivariate analysis (hazard ratio = 3.22; 95% confidence interval = 1.03-10.11; p=0.045).

CONCLUSIONS

The logistic regression model constructed by UICC, T and N stages and pretreatment [¹⁸F]-FDG-PET-based radiomic features, GLCM_Entropy, and GLRLM_RLNU may be the most important predictor of prognosis in patients with hypopharyngeal cancer.

Gastrointestinal disorders in children with autism: Could artificial intelligence help?

Autism is one of the pervasive neurodevelopmental disorders usually associated with many medical comorbidities. Gastrointestinal (GI) disorders are pervasive in children, with a 46%-84% prevalence rate. Children with Autism have an increased frequency of diarrhea, nausea and/or vomiting, gastroesophageal reflux and/or disease, abdominal pain, chronic flatulence due to various factors as food allergies, gastrointestinal dysmotility, irritable bowel syndrome (IBS), and inflammatory bowel diseases (IBD). These GI disorders have a significant negative impact on both the child and his/her family. Artificial intelligence (AI) could help diagnose and manage Autism by improving children's communication, social, and emotional skills for a long time. AI is an effective method to enhance early detection of GI disorders, including GI bleeding, gastroesophageal reflux disease, Coeliac disease, food allergies, IBS, IBD, and rectal polyps. AI can also help personalize the diet for children with Autism by microbiome modification. It can help to provide modified gluten without initiating an immune response. However, AI has many obstacles in treating digestive diseases, especially in children with Autism. We need to do more studies and adopt specific algorithms for children with Autism. In this article, we will highlight the role of AI in helping children with gastrointestinal disorders, with particular emphasis on children with Autism.

Machine learning based evaluation of clinical and pretreatment 18F-FDG-PET/CT radiomic features to predict prognosis of cervical cancer patients

PURPOSE

To examine the usefulness of machine learning to predict prognosis in cervical cancer using clinical and radiomic features of 2-deoxy-2-[¹⁸F]fluoro-D-glucose (¹⁸F-FDG) positron emission tomography/computed tomography (CT) (¹⁸F-FDG-PET/CT).

METHODS

This retrospective study included 50 cervical cancer patients who underwent ¹⁸F-FDG-PET/CT before treatment. Four clinical (age, histology, stage, and treatment) and 41 ¹⁸F-FDG-PET-based radiomic features were ranked and a subset of useful features for association with disease progression was selected based on decrease of the Gini impurity. Six machine learning algorithms (random forest, neural network, k-nearest neighbors, naive Bayes, logistic regression, and support vector machine) were compared using the areas under the receiver operating characteristic curve (AUC). Progression-free survival (PFS) was assessed using Cox regression analysis.

RESULTS

The five top predictors of disease progression were: stage, surface area, metabolic tumor volume, gray-level run length non-uniformity (GLRLM_RLNU), and gray-level non-uniformity for run (GLRLM_GLNU). The naive Bayes model was the best-performing classifier for predicting disease progression (AUC = 0.872, accuracy = 0.780, F1 score = 0.781, precision = 0.788, and recall = 0.780). In the naive Bayes model, 5-year PFS was significantly higher in predicted non-progression than predicted progression (80.1% vs. 9.1%, p < 0.001) and was only the independent factor for PFS in multivariate analysis (HR, 6.89; 95% CI, 1.92-24.69; p = 0.003).

CONCLUSION

A machine learning approach based on clinical and pretreatment ¹⁸F-FDG PET-based radiomic features may be useful for predicting tumor progression in cervical cancer patients.

Computational Models for Clinical Applications in Personalized Medicine—Guidelines and Recommendations for Data Integration and Model Validation

The future development of personalized medicine depends on a vast exchange of data from different sources, as well as harmonized integrative analysis of large-scale clinical health and sample data. Computational-modelling approaches play a key role in the analysis of the underlying molecular processes and pathways that characterize human biology, but they also lead to a more profound understanding of the mechanisms and factors that drive diseases; hence, they allow personalized treatment strategies that are guided by central clinical questions. However, despite the growing popularity of computational-modelling approaches in different stakeholder communities, there are still many hurdles to overcome for their clinical routine implementation in the future. Especially the integration of heterogeneous data from multiple sources and types are challenging tasks that require clear guidelines that also have to comply with high ethical and legal standards. Here, we discuss the most relevant computational models for personalized medicine in detail that can be considered as best-practice guidelines for application in clinical care. We define specific challenges and provide applicable guidelines and recommendations for study design, data acquisition, and operation as well as for model validation and clinical translation and other research areas.

Food for Thought: Machine Learning in Autism Spectrum Disorder Screening of Infants

Diagnoses of autism spectrum disorders (ASD) are typically made after toddlerhood by examining behavioural patterns. Earlier identification of ASD enables earlier intervention and better outcomes. Machine learning provides a data-driven approach of diagnosing autism at an earlier age. This review aims to summarize recent studies and technologies utilizing machine learning based strategies to screen infants and children under the age of 18 months for ASD, and identify gaps that can be addressed in the future. We reviewed nine studies based on our search criteria, which includes primary studies and technologies conducted within the last 10 years that examine children with ASD or at high risk of ASD with a mean age of less than 18 months old. The studies must use machine learning analysis of behavioural features of ASD as major methodology. A total of nine studies were reviewed, of which the sensitivity ranges from 60.7% to 95.6%, the specificity ranges from 50% to 100%, and the accuracy ranges from 60.9% to 97.7%. Factors that contribute to the inconsistent findings include the varied presentation of ASD among patients and study design differences. Previous studies have shown moderate accuracy, sensitivity and specificity in the differentiation of ASD and non-ASD individuals under the age of 18 months. The application of machine learning and artificial intelligence in the screening of ASD in infants is still in its infancy, as observed by the granularity of data available for review. As such, much work needs to be done before the aforementioned technologies can be applied into clinical practice to facilitate early screening of ASD.

Prediction of Autism Risk From Family Medical History Data Using Machine Learning: A National Cohort Study From Denmark

Background

A family history of specific disorders (e.g., autism, depression, epilepsy) has been linked to risk for autism spectrum disorder (ASD). This study examines whether family history data could be used for ASD risk prediction.

Methods

We followed all Danish live births, from 1980 to 2012, of Denmark-born parents for an ASD diagnosis through April 10, 2017 (N = 1,697,231 births; 26,840 ASD cases). Linking each birth to three-generation family members, we identified 438 morbidity indicators, comprising 73 disorders reported prospectively for each family member. We tested various models using a machine learning approach. From the best-performing model, we calculated a family history risk score and estimated odds ratios and 95% confidence intervals for the risk of ASD.

Results

The best-performing model comprised 41 indicators: eight mental conditions (e.g., ASD, attention-deficit/hyperactivity disorder, neurotic/stress disorders) and nine nonmental conditions (e.g., obesity, hypertension, asthma) across six family member types; model performance was similar in training and test subsamples. The highest risk score group had 17.0% ASD prevalence and a 15.3-fold (95% confidence interval, 14.0-17.1) increased ASD risk compared with the lowest score group, which had 0.6% ASD prevalence. In contrast, individuals with a full sibling with ASD had 9.5% ASD prevalence and a 6.1-fold (95% confidence interval, 5.9-6.4) higher risk than individuals without an affected sibling.

Conclusions

Family history of multiple mental and nonmental conditions can identify more individuals at highest risk for ASD than only considering the immediate family history of ASD. A comprehensive family history may be critical for a clinically relevant ASD risk prediction framework in the future.

Application of a machine learning approach to characterization of liver function using 99mTc-GSA SPECT/CT

PURPOSE

To assess the utility of a machine-learning approach for predicting liver function based on technetium-99 m-galactosyl serum albumin (^99mTc-GSA) single photon emission computed tomography (SPECT)/CT.

METHODS

One hundred twenty-eight patients underwent a ^99mTc-GSA SPECT/CT-based liver function evaluation. All were classified into the low liver-damage or high liver-damage group. Four clinical (age, sex, background liver disease and histological type) and 8 quantitative ^99mTc-GSA SPECT/CT features (receptor index [LHL15], clearance index [HH15], liver-SUVmax, liver-SUVmean, heart-SUVmax, metabolic volume of liver [MVL], total lesion GSA [TL-GSA, liver-SUVmean × MVL] and SUVmax ratio [liver-SUVmax/heart-SUVmax]) were obtained. To predict high liver damage, a machine learning classification with features selection based on Gini impurity and principal component analysis (PCA) were performed using a support vector machine and a random forest (RF) with a five-fold cross-validation scheme. To overcome imbalanced data, stratified sampling was used. The ability to predict high liver damage was evaluated using a receiver operating characteristic (ROC) curve analysis.

RESULTS

Four indices (LHL15, HH15, heart SUVmax and SUVmax ratio) yielded high areas under the ROC curves (AUCs) for predicting high liver damage (range: 0.89-0.93). In a machine learning classification, the RF with selected features (heart SUVmax, SUVmax ratio, LHL15, HH15, and background liver disease) and PCA model yielded the best performance for predicting high liver damage (AUC = 0.956, sensitivity = 96.3%, specificity = 90.0%, accuracy = 91.4%).

CONCLUSION

A machine-learning approach based on clinical and quantitative ^99mTc-GSA SPECT/CT parameters might be useful for predicting liver function.

Maternal health around pregnancy and autism risk: a diagnosis-wide, population-based study

BACKGROUND

Many studies have reported an increased risk of autism spectrum disorder (ASD) associated with some maternal diagnoses in pregnancy. However, such associations have not been studied systematically, accounting for comorbidity between maternal disorders. Therefore our aim was to comprehensively test the associations between maternal diagnoses around pregnancy and ASD risk in offspring.

METHODS

This exploratory case-cohort study included children born in Israel from 1997 to 2008, and followed up until 2015. We used information on all ICD-9 codes received by their mothers during pregnancy and the preceding year. ASD risk associated with each of those conditions was calculated using Cox proportional hazards regression, adjusted for the confounders (birth year, maternal age, socioeconomic status and number of ICD-9 diagnoses during the exposure period).

RESULTS

The analytic sample consisted of 80 187 individuals (1132 cases, 79 055 controls), with 822 unique ICD-9 codes recorded in their mothers. After extensive quality control, 22 maternal diagnoses were nominally significantly associated with offspring ASD, with 16 of those surviving subsequent filtering steps (permutation testing, multiple testing correction, multiple regression). Among those, we recorded an increased risk of ASD associated with metabolic [e.g. hypertension; HR = 2.74 (1.92-3.90), p = 2.43 × 10-8], genitourinary [e.g. non-inflammatory disorders of cervix; HR = 1.88 (1.38-2.57), p = 7.06 × 10-5] and psychiatric [depressive disorder; HR = 2.11 (1.32-3.35), p = 1.70 × 10-3] diagnoses. Meanwhile, mothers of children with ASD were less likely to attend prenatal care appointment [HR = 0.62 (0.54-0.71), p = 1.80 × 10-11].

CONCLUSIONS

Sixteen maternal diagnoses were associated with ASD in the offspring, after rigorous filtering of potential false-positive associations. Replication in other cohorts and further research to understand the mechanisms underlying the observed associations with ASD are warranted.

Application of a Machine Learning Approach for the Analysis of Clinical and Radiomic Features of Pretreatment [18F]-FDG PET/CT to Predict Prognosis of Patients with Endometrial Cancer

PURPOSE

To examine the prognostic significance of pretreatment 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]-FDG) positron emission tomography (PET)-based radiomic features using a machine learning approach in patients with endometrial cancers.

PROCEDURES

Included in this retrospective study were 53 patients with endometrial cancers who underwent [¹⁸F]-FDG PET/X-ray computed tomography (CT) before treatment. Since two different PET scanners were used, post-reconstruction harmonization was performed for all PET parameters using the ComBat harmonization method. Four clinical (age, histological type, stage, and treatment method) and 40 [¹⁸F]-FDG PET-based radiomic features were ranked, and a subset of useful features was selected based on the decrease in the Gini impurity in terms of associations with disease progression. The machine learning algorithms (random forest, neural network, k-nearest neighbors (kNN), naive Bayes, logistic regression, and support vector machine) were compared using the areas under the receiver operating characteristic curve (AUC) and validated by the random sampling method. Progression-free survival (PFS) and overall survival (OS) were assessed by the Cox regression analysis.

RESULTS

The five best predictors of disease progression were coarseness, gray-level run length nonuniformity, stage, treatment method, and gray-level zone length nonuniformity. The kNN model obtained the best performance classifier for predicting the disease progression (AUC =0.890, accuracy =0.849, F1 score =0.848, precision =0.857, and recall =0.849). Coarseness which was the first ranked radiomic feature was selected for survival analyses, and only coarseness remained as a significant and independent factor for both PFS (hazard ratios (HR), 0.65; 95 % confidence interval [CI], 0.49-0.86; p=0.003) and OS (HR, 0.52; 95 % CI, 0.36-0.76; p<0.001) at multivariate Cox regression analysis.

CONCLUSIONS

[¹⁸F]-FDG PET-based radiomic analysis using a machine learning approach may be useful for predicting tumor progression and prognosis in patients with endometrial cancers.

Can machine learning be useful as a screening tool for depression in primary care?

Depression is a widespread disease with a high economic burden and a complex pathophysiology disease that is still not wholly clarified, not to mention it usually is associated as a risk factor for absenteeism at work and suicide. Just 50% of patients with depression are diagnosed in primary care, and only 15% receive treatment. Stigmatization, the coexistence of somatic symptoms, and the need to remember signs in the past two weeks can contribute to explaining this situation. In this context, tools that can serve as diagnostic screening are of great value, as they can reduce the number of undiagnosed patients. Besides, Artificial Intelligence (AI) has enabled several fruitful applications in medicine, particularly in psychiatry. This study aims to evaluate the performance of Machine Learning (ML) algorithms in the detection of depressive patients from the clinical, laboratory, and sociodemographic data obtained from the Brazilian National Network for Research on Cardiovascular Diseases from June 2016 to July 2018. The results obtained are promising. In one of them, Random Forests, the accuracy, sensibility, and area under the receiver operating characteristic curve were, respectively, 0.89, 0.90, and 0.87.