Predictive modeling for classification of positive valence system symptom severity from initial psychiatric evaluation records.
Ontology highlight
ABSTRACT:
SUBMITTER: Posada JD
PROVIDER: S-EPMC5705330 | biostudies-literature | 2017 Nov
REPOSITORIES: biostudies-literature
Publications
Predictive modeling for classification of positive valence system symptom severity from initial psychiatric evaluation records.
Posada Jose D JD Barda Amie J AJ Shi Lingyun L Xue Diyang D Ruiz Victor V Kuan Pei-Han PH Ryan Neal D ND Tsui Fuchiang Rich FR
Journal of biomedical informatics 20170529
In response to the challenges set forth by the CEGS N-GRID 2016 Shared Task in Clinical Natural Language Processing, we describe a framework to automatically classify initial psychiatric evaluation records to one of four positive valence system severities: absent, mild, moderate, or severe. We used a dataset provided by the event organizers to develop a framework comprised of natural language processing (NLP) modules and 3 predictive models (two decision tree models and one Bayesian network mode ...[more]
Similar Datasets
Project description:Background and objectiveEfficiently capturing the severity of positive valence symptoms could aid in risk stratification for adverse outcomes among patients with psychiatric disorders and identify optimal treatment strategies for patient subgroups. Motivated by the success of convolutional neural networks (CNNs) in classification tasks, we studied the application of various CNN architectures and their performance in predicting the severity of positive valence symptoms in patients with psychiatric disorders based on initial psychiatric evaluation records.MethodsPsychiatric evaluation records contain unstructured text and semi-structured data such as question-answer pairs. For a given record, we tokenise and normalise the semi-structured content. Pre-processed tokenised words are represented as one-hot encoded word vectors. We then apply different configurations of convolutional and max pooling layers to automatically learn important features from various word representations. We conducted a series of experiments to explore the effect of different CNN architectures on the classification of psychiatric records.ResultsOur best CNN model achieved a mean absolute error (MAE) of 0.539 and a normalized MAE of 0.785 on the test dataset, which is comparable to the other well-known text classification algorithms studied in this work. Our results also suggest that the normalisation step has a great impact on the performance of the developed models.ConclusionsWe demonstrate that normalisation of the semi-structured contents can improve the MAE among all CNN configurations. Without advanced feature engineering, CNN-based approaches can provide a comparable solution for classifying positive valence symptom severity in initial psychiatric evaluation records. Although word embedding is well known for its ability to capture relatively low-dimensional similarity between words, our experimental results show that pre-trained embeddings do not improve the classification performance. This phenomenon may be due to the inability of word embeddings to capture problem specific contextual semantic information implying the quality of the employing embedding is critical for obtaining an accurate CNN model.
| S-EPMC6191093 | biostudies-literature
Project description:This paper presents a novel method for automatically recognizing symptom severity by using natural language processing of psychiatric evaluation records to extract features that are processed by machine learning techniques to assign a severity score to each record evaluated in the 2016 RDoC for Psychiatry Challenge from CEGS/N-GRID. The natural language processing techniques focused on (a) discerning the discourse information expressed in questions and answers; (b) identifying medical concepts that relate to mental disorders; and (c) accounting for the role of negation. The machine learning techniques rely on the assumptions that (1) the severity of a patient's positive valence symptoms exists on a latent continuous spectrum and (2) all the patient's answers and narratives documented in the psychological evaluation records are informed by the patient's latent severity score along this spectrum. These assumptions motivated our two-step machine learning framework for automatically recognizing psychological symptom severity. In the first step, the latent continuous severity score is inferred from each record; in the second step, the severity score is mapped to one of the four discrete severity levels used in the CEGS/N-GRID challenge. We evaluated three methods for inferring the latent severity score associated with each record: (i) pointwise ridge regression; (ii) pairwise comparison-based classification; and (iii) a hybrid approach combining pointwise regression and the pairwise classifier. The second step was implemented using a tree of cascading support vector machine (SVM) classifiers. While the official evaluation results indicate that all three methods are promising, the hybrid approach not only outperformed the pairwise and pointwise methods, but also produced the second highest performance of all submissions to the CEGS/N-GRID challenge with a normalized MAE score of 84.093% (where higher numbers indicate better performance). These evaluation results enabled us to observe that, for this task, considering pairwise information can produce more accurate severity scores than pointwise regression - an approach widely used in other systems for assigning severity scores. Moreover, our analysis indicates that using a cascading SVM tree outperforms traditional SVM classification methods for the purpose of determining discrete severity levels.
| S-EPMC5705296 | biostudies-literature
Project description:The CEGS N-GRID 2016 Shared Task (Filannino et al., 2017) in Clinical Natural Language Processing introduces the assignment of a severity score to a psychiatric symptom, based on a psychiatric intake report. We present a method that employs the inherent interview-like structure of the report to extract relevant information from the report and generate a representation. The representation consists of a restricted set of psychiatric concepts (and the context they occur in), identified using medical concepts defined in UMLS that are directly related to the psychiatric diagnoses present in the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) ontology. Random Forests provides a generalization of the extracted, case-specific features in our representation. The best variant presented here scored an inverse mean absolute error (MAE) of 80.64%. A concise concept-based representation, paired with identification of concept certainty and scope (family, patient), shows a robust performance on the task.
| S-EPMC5705466 | biostudies-literature
Project description:ObjectiveTo optimally leverage the scalability and unique features of the electronic health records (EHR) for research that would ultimately improve patient care, we need to accurately identify patients and extract clinically meaningful measures. Using multiple sclerosis (MS) as a proof of principle, we showcased how to leverage routinely collected EHR data to identify patients with a complex neurological disorder and derive an important surrogate measure of disease severity heretofore only available in research settings.MethodsIn a cross-sectional observational study, 5,495 MS patients were identified from the EHR systems of two major referral hospitals using an algorithm that includes codified and narrative information extracted using natural language processing. In the subset of patients who receive neurological care at a MS Center where disease measures have been collected, we used routinely collected EHR data to extract two aggregate indicators of MS severity of clinical relevance multiple sclerosis severity score (MSSS) and brain parenchymal fraction (BPF, a measure of whole brain volume).ResultsThe EHR algorithm that identifies MS patients has an area under the curve of 0.958, 83% sensitivity, 92% positive predictive value, and 89% negative predictive value when a 95% specificity threshold is used. The correlation between EHR-derived and true MSSS has a mean R(2) = 0.38±0.05, and that between EHR-derived and true BPF has a mean R(2) = 0.22±0.08. To illustrate its clinical relevance, derived MSSS captures the expected difference in disease severity between relapsing-remitting and progressive MS patients after adjusting for sex, age of symptom onset and disease duration (p = 1.56×10(-12)).ConclusionIncorporation of sophisticated codified and narrative EHR data accurately identifies MS patients and provides estimation of a well-accepted indicator of MS severity that is widely used in research settings but not part of the routine medical records. Similar approaches could be applied to other complex neurological disorders.
| S-EPMC3823928 | biostudies-literature
Project description:Autism spectrum disorder (ASD) affects 1 in 88 children and is characterized by a complex phenotype, including social, communicative, and sensorimotor deficits. Autism spectrum disorder has been linked with atypical connectivity across multiple brain systems, yet the nature of these differences in young children with the disorder is not well understood.To examine connectivity of large-scale brain networks and determine whether specific networks can distinguish children with ASD from typically developing (TD) children and predict symptom severity in children with ASD.Case-control study performed at Stanford University School of Medicine of 20 children 7 to 12 years old with ASD and 20 age-, sex-, and IQ-matched TD children.Between-group differences in intrinsic functional connectivity of large-scale brain networks, performance of a classifier built to discriminate children with ASD from TD children based on specific brain networks, and correlations between brain networks and core symptoms of ASD.We observed stronger functional connectivity within several large-scale brain networks in children with ASD compared with TD children. This hyperconnectivity in ASD encompassed salience, default mode, frontotemporal, motor, and visual networks. This hyperconnectivity result was replicated in an independent cohort obtained from publicly available databases. Using maps of each individual's salience network, children with ASD could be discriminated from TD children with a classification accuracy of 78%, with 75% sensitivity and 80% specificity. The salience network showed the highest classification accuracy among all networks examined, and the blood oxygen-level dependent signal in this network predicted restricted and repetitive behavior scores. The classifier discriminated ASD from TD in the independent sample with 83% accuracy, 67% sensitivity, and 100% specificity.Salience network hyperconnectivity may be a distinguishing feature in children with ASD. Quantification of brain network connectivity is a step toward developing biomarkers for objectively identifying children with ASD.
| S-EPMC3951904 | biostudies-literature
Project description:Predictive modeling of human disease based on the microbiota holds great potential yet remains challenging. Here, 50 adults underwent controlled transitions from naturally occurring gingivitis, to healthy gingivae (baseline), and to experimental gingivitis (EG). In diseased plaque microbiota, 27 bacterial genera changed in relative abundance and functional genes including 33 flagellar biosynthesis-related groups were enriched. Plaque microbiota structure exhibited a continuous gradient along the first principal component, reflecting transition from healthy to diseased states, which correlated with Mazza Gingival Index. We identified two host types with distinct gingivitis sensitivity. Our proposed microbial indices of gingivitis classified host types with 74% reliability, and, when tested on another 41-member cohort, distinguished healthy from diseased individuals with 95% accuracy. Furthermore, the state of the microbiota in naturally occurring gingivitis predicted the microbiota state and severity of subsequent EG (but not the state of the microbiota during the healthy baseline period). Because the effect of disease is greater than interpersonal variation in plaque, in contrast to the gut, plaque microbiota may provide advantages in predictive modeling of oral diseases.
| S-EPMC4139724 | biostudies-literature
Project description:OBJECTIVE:Sleep disturbance may be involved in symptom progression across multiple domains of psychopathology and could represent a target for treatment development in youth. Our objective was to identify sleep patterns that longitudinally change in conjunction with psychiatric symptom severity in at-risk youth. METHOD:The study included 484 Pittsburgh Bipolar Offspring Study (BIOS) youth with at least 2 sleep assessments occurring between 10 and 18 years of age: 267 offspring of parents with bipolar I or II disorder and 217 community comparison offspring. Assessments occurred approximately every 2 years (mean number of assessments, 2.8 ± 0.8; mean follow-up duration, 3.8 ± 1.6 years). Offspring had a range of psychiatric diagnoses at baseline. Multivariate lasso regression was implemented to select offspring-reported sleep patterns associated with changes in five psychiatric symptom measures from baseline through last follow-up (mania, depression, mood lability, anxiety, inattention/externalizing). Analyses accounted for parent psychiatric diagnoses and offspring demographics, psychiatric diagnoses, and medications. RESULTS:Follow-up duration, baseline socioeconomic status, parental history of bipolar disorder, offspring attention-deficit/hyperactivity disorder, and disruptive behavior disorder, and five sleep patterns were identified as predictors of change in all five psychiatric symptom measures. Decreasing sleep duration, later sleep timing preference, longer sleep latency, increasing nighttime awakenings, and greater sleepiness over follow-up were associated with increasing severity the five psychiatric symptom outcomes over follow-up. These 10 predictors explained 16% of the variance in longitudinal psychiatric symptom change, 33% of which was accounted for by sleep predictors. CONCLUSION:A constellation of sleep features were associated with psychiatric symptom changes in youth, and may represent viable targets for future interventions.
| S-EPMC6733405 | biostudies-literature
Project description:ObjectivesNegative psychiatric symptoms are often resistant to treatments, regardless of the disorder in which they appear. One model for a cause of negative symptoms is impairment in higher-order cognition. The current study examined how particular bottom-up and top-down mechanisms of selective attention relate to severity of negative symptoms across a transdiagnostic psychiatric sample.MethodsThe sample consisted of 130 participants: 25 schizophrenia-spectrum disorders, 26 bipolar disorders, 18 unipolar depression, and 61 nonpsychiatric controls. The relationships between attentional event-related potentials following rare visual targets (i.e., N1, N2b, P2a, and P3b) and severity of the negative symptom domains of anhedonia, avolition, and blunted affect were evaluated using frequentist and Bayesian analyses.ResultsP3b and N2b mean amplitudes were inversely related to the Positive and Negative Syndrome Scale-Negative Symptom Factor severity score across the entire sample. Subsequent regression analyses showed a significant negative transdiagnostic relationship between P3b amplitude and blunted affect severity.ConclusionsResults indicate that negative symptoms, and particularly blunted affect, may have a stronger association with deficits in top-down mechanisms of selective attention.SignificanceThis suggests that people with greater severity of blunted affect, independent of diagnosis, do not allocate sufficient cognitive resources when engaging in activities requiring selective attention.
| S-EPMC8223322 | biostudies-literature
Project description:BackgroundGrowing evidence documents the potential of machine learning for developing brain based diagnostic methods for major depressive disorder (MDD). As symptom severity may influence brain activity, we investigated whether the severity of MDD affected the accuracies of machine learned MDD-vs-Control diagnostic classifiers.MethodsForty-five medication-free patients with DSM-IV defined MDD and 19 healthy controls participated in the study. Based on depression severity as determined by the Hamilton Rating Scale for Depression (HRSD), MDD patients were sorted into three groups: mild to moderate depression (HRSD 14-19), severe depression (HRSD 20-23), and very severe depression (HRSD ? 24). We collected functional magnetic resonance imaging (fMRI) data during both resting-state and an emotional-face matching task. Patients in each of the three severity groups were compared against controls in separate analyses, using either the resting-state or task-based fMRI data. We use each of these six datasets with linear support vector machine (SVM) binary classifiers for identifying individuals as patients or controls.ResultsThe resting-state fMRI data showed statistically significant classification accuracy only for the very severe depression group (accuracy 66%, p = 0.012 corrected), while mild to moderate (accuracy 58%, p = 1.0 corrected) and severe depression (accuracy 52%, p = 1.0 corrected) were only at chance. With task-based fMRI data, the automated classifier performed at chance in all three severity groups.ConclusionsBinary linear SVM classifiers achieved significant classification of very severe depression with resting-state fMRI, but the contribution of brain measurements may have limited potential in differentiating patients with less severe depression from healthy controls.
| S-EPMC4983635 | biostudies-literature
Project description:BackgroundThe Borderline Symptom List (BSL-23) is a well-established self-rating instrument to assess the severity of borderline typical psychopathology. However, a classification of severity levels for the BSL-23 is missing.MethodsData from 1.090 adults were used to develop a severity classification for the Borderline Symptom List (BSL-23). The severity grading was based on the distribution of the BSL-23 in 241 individuals with a diagnosis of BPD. Data from three independent samples were used to validate the previously defined severity grades.These validation samples included a group of treatment seeking patients with a diagnosis of BPD (n = 317), a sample of individuals with mental illnesses other than BPD (n = 176), and a healthy control sample (n = 356). The severity grades were validated from comparisons with established assessment instruments such as the International Personality Disorders Examination, the Structured Clinical Interview for DSM-IV, the global severity index of the Symptom Checklist (GSI, SCL-90), the Global Assessment of Functioning (GAF), and the Beck Depression Inventory (BDI-II).ResultsSix grades of symptom severity were defined for the BSL-23 mean score: none or low: 0-0.3; mild: 0.3-0.7; moderate: 0.7-1.7; high: 1.7-2.7; very high: 2.7-3.5; and extremely high: 3.5-4. These grades received consistent empirical support from the independent instruments and samples. For instance, individuals with a severity grade of none or low were virtually free from diagnostic BPD-criteria, had a GSI below the normative population, and a high level of global functioning corresponding to few or no symptoms. Severity grades indicating high to extremely high levels of BPD symptoms were observed at a much higher rate in treatment-seeking patients (70.0%) than in clinical controls (17.6%) and healthy controls (0.0%). The BSL-23 score that best separated treatment-seeking BPD patients and clinical controls was 1.50, whereas the clearest discrimination of BPD patients and healthy controls was found at a score of 0.64.ConclusionsThe grades of BPD-specific symptom severity derived from the distribution of the BSL-23 scores received consistent empirical validation from established assessments for psychopathology. Future studies should expand this validation by including additional instruments e.g., to assess self-esteem, loneliness, connectedness, and quality of life.
| S-EPMC7262769 | biostudies-literature