David Chandran

Replicable and Coupled Changes in Innate and Adaptive Immune Gene Expression in Two Case-Control Studies of Blood Microarrays in Major Depressive Disorder

Background Peripheral inflammation is often associated with major depressive disorder (MDD), and immunological biomarkers of depression remain a focus of investigation. Methods We used microarray data on whole blood from two independent case-control studies of MDD: the GlaxoSmithKline–High-Throughput Disease-specific target Identification Program [GSK-HiTDiP] study (113 patients and 57 healthy control subjects) and the Janssen–Brain Resource Company study (94 patients and 100 control subjects). Genome-wide differential gene expression analysis (18,863 probes) resulted in a p value for each gene in each study. A Bayesian method identified the largest p-value threshold (q = .025) associated with twice the number of genes differentially expressed in both studies compared with the number of coincidental case-control differences expected by chance. Results A total of 165 genes were differentially expressed in both studies with concordant direction of fold change. The 90 genes overexpressed (or UP genes) in MDD were significantly enriched for immune response to infection, were concentrated in a module of the gene coexpression network associated with innate immunity, and included clusters of genes with correlated expression in monocytes, monocyte-derived dendritic cells, and neutrophils. In contrast, the 75 genes underexpressed (or DOWN genes) in MDD were associated with the adaptive immune response and included clusters of genes with correlated expression in T cells, natural killer cells, and erythroblasts. Consistently, the MDD patients with overexpression of UP genes also had underexpression of DOWN genes (correlation > .70 in both studies). Conclusions MDD was replicably associated with proinflammatory activation of the peripheral innate immune system, coupled with relative inactivation of the adaptive immune system, indicating the potential of transcriptional biomarkers for immunological stratification of patients with depression.

Repeated exposure to systemic inflammation and risk of new depressive symptoms among older adults

Evidence on systemic inflammation as a risk factor for future depression is inconsistent, possibly due to a lack of regard for persistency of exposure. We examined whether being inflamed on multiple occasions increases risk of new depressive symptoms using prospective data from a population-based sample of adults aged 50 years or older (the English Longitudinal Study of Ageing). Participants with less than four of eight depressive symptoms in 2004/05 and 2008/09 based on the Eight-item Centre for Epidemiologic Studies Depression scale were analysed. The number of occasions with C-reactive protein ⩾3 mg l−1 over the same initial assessments (1 vs 0 occasion, and 2 vs 0 occasions) was examined in relation to change in depressive symptoms between 2008/09 and 2012/13 and odds of developing depressive symptomology (having more than or equal to four of eight symptoms) in 2012/13. In multivariable-adjusted regression models (n=2068), participants who were inflamed on 1 vs 0 occasion showed no increase in depressive symptoms nor raised odds of developing depressive symptomology; those inflamed on 2 vs 0 occasions showed a 0.10 (95% confidence intervals (CIs)=−0.07, 0.28) symptom increase and 1.60 (95% CI=1.00, 2.55) times higher odds. In further analyses, 2 vs 0 occasions of inflammation were associated with increased odds of developing depressive symptoms among women (odds ratio (OR)=2.75, 95% CI=1.53, 4.95), but not among men (OR=0.70, 95% CI=0.29, 1.68); P-for-sex interaction=0.035. In this cohort study of older adults, repeated but not transient exposure to systemic inflammation was associated with increased risk of future depressive symptoms among women; this subgroup finding requires confirmation of validity.

FAST: A fuzzy semantic sentence similarity measure

A problem in the field of semantic sentence similarity is the inability of sentence similarity measures to accurately represent perception based (fuzzy) words that are commonly used in natural language. This paper presents a new sentence similarity measure that attempts to solve this problem. The new measure, Fuzzy Algorithm for Similarity Testing (FAST) is an ontology based similarity measure that uses concepts of fuzzy and computing with words to allow for the accurate representation of fuzzy based words. Through human experimentation fuzzy sets were created for six categories of words based on their levels of association with particular concepts. These fuzzy sets were then defuzzified and the results used to create new ontological relations between the words. Using these relationships allows for the creation of a new ontology based semantic text similarity algorithm that is able to show the effect of fuzzy words on computing sentence similarity as well as the effect that fuzzy words have on non-fuzzy words within a sentence. Experiments on FAST were conducted using a new fuzzy dataset, the creation of which is described in this paper. The results of the evaluation showed that there was an improved level of correlation between FAST and human test results over two existing sentence similarity measures.

On the creation of a fuzzy dataset for the evaluation of fuzzy semantic similarity measures

Short text semantic similarity (STSS) measures are algorithms designed to compare short texts and return a level of similarity between them. However, until recently such measures have ignored perception or fuzzy based words (i.e. very hot, cold less cold) in calculations of both word and sentence similarity. Evaluation of such measures is usually achieved through the use of benchmark data sets comprising of a set of rigorously collected sentence pairs which have been evaluated by human participants. A weakness of these datasets is that the sentences pairs include limited, if any, fuzzy based words that makes them impractical for evaluating fuzzy sentence similarity measures. In this paper, a method is presented for the creation of a new benchmark dataset known as SFWD (Single Fuzzy Word Dataset). After creation the data set is then used in the evaluation of FAST, an ontology based fuzzy algorithm for semantic similarity testing that uses concepts of fuzzy and computing with words to allow for the accurate representation of fuzzy based words. The SFWD is then used to undertake a comparative analysis of other established STSS measures.

An automatic corpus based method for a building Multiple Fuzzy Word Dataset

Fuzzy sentence semantic similarity measures are designed to be applied to real world problems where a computer system is required to assess the similarity between human natural language and words or prototype sentences stored within a knowledge base. Such measures are often developed for a specific corpus/domain where a limited set of words and sentences are evaluated. As new “fuzzy” measures are developed the research challenge is on how to evaluate them. Traditional approaches have involved rigorous and complex human involvement in compiling benchmark datasets and obtaining human similarity measures. Existing datasets often contain limited fuzzy words and do allow the fuzzy measures to be exhaustively tested. This paper presents an automatic method for the generation of a Multiple Fuzzy Word Dataset (MFWD) from a corpus. A Fuzzy Sentence Pairing Algorithm is used to extract and augment high, medium and low similarity sentence pairs with multiple fuzzy words. Human ratings are collected through crowdsourcing and the MFWD is evaluated using both fuzzy and traditional sentence similarity measures. The results indicated that fuzzy measures returned a higher correlation with human ratings compared with traditional measures.

Identifying Suicide Ideation and Suicidal Attempts in a Psychiatric Clinical Research Database using Natural Language Processing

Research into suicide prevention has been hampered by methodological limitations such as low sample size and recall bias. Recently, Natural Language Processing (NLP) strategies have been used with Electronic Health Records to increase information extraction from free text notes as well as structured fields concerning suicidality and this allows access to much larger cohorts than previously possible. This paper presents two novel NLP approaches – a rule-based approach to classify the presence of suicide ideation and a hybrid machine learning and rule-based approach to identify suicide attempts in a psychiatric clinical database. Good performance of the two classifiers in the evaluation study suggest they can be used to accurately detect mentions of suicide ideation and attempt within free-text documents in this psychiatric database. The novelty of the two approaches lies in the malleability of each classifier if a need to refine performance, or meet alternate classification requirements arises. The algorithms can also be adapted to fit infrastructures of other clinical datasets given sufficient clinical recording practice knowledge, without dependency on medical codes or additional data extraction of known risk factors to predict suicidal behaviour.

Demographic and clinical factors associated with different antidepressant treatments: a retrospective cohort study design in a UK psychiatric healthcare setting

Objective To investigate the demographic and clinical factors associated with antidepressant use for depressive disorder in a psychiatric healthcare setting using a retrospective cohort study design. Setting Data were extracted from a de-identified data resource sourced from the electronic health records of a London mental health service. Relative risk ratios (RRRs) were obtained from multinomial logistic regression analysis to ascertain the probability of receiving common antidepressant treatments relative to sertraline. Participants Patients were included if they received mental healthcare and a diagnosis of depression with antidepressant treatment between March and August 2015 and exposures were measured over the preceding 12 months. Results Older age was associated with increased use of all antidepressants compared with sertraline, except for negative associations with fluoxetine (RRR 0.98; 95% CI 0.96 to 0.98) and a combination of two selective serotonin reuptake inhibitors (SSRIs) (0.98; 95% CI 0.96 to 0.99), and no significant association with escitalopram. Male gender was associated with increased use of mirtazapine compared with sertraline (2.57; 95% CI 1.85 to 3.57). Previous antidepressant, antipsychotic and mood stabiliser use were associated with newer antidepressant use (ie, selective norepinephrine reuptake inhibitors, mirtazapine or a combination of both), while affective symptoms were associated with reduced use of citalopram (0.58; 95% CI 0.27 to 0.83) and fluoxetine (0.42; 95% CI 0.22 to 0.72) and somatic symptoms were associated with increased use of mirtazapine (1.60; 95% CI 1.00 to 2.75) relative to sertraline. In patients older than 25 years, past benzodiazepine use was associated with a combination of SSRIs (2.97; 95% CI 1.32 to 6.68), mirtazapine (1.94; 95% CI 1.20 to 3.16) and venlafaxine (1.87; 95% CI 1.04 to 3.34), while past suicide attempts were associated with increased use of fluoxetine (2.06; 95% CI 1.10 to 3.87) relative to sertraline. Conclusion There were several factors associated with different antidepressant receipt in psychiatric healthcare. In patients aged >25, those on fluoxetine were more likely to have past suicide attempt, while past use of antidepressant and non-antidepressant use was also associated with use of new generation antidepressants, potentially reflecting perceived treatment resistance.

Application of fuzzy semantic similarity measures to event detection within tweets

This paper examines the suitability of applying fuzzy semantic similarity measures (FSSM) to the task of detecting potential future events through the use of a group of prototypical event tweets. FSSM are ideal measures to be used to analyse the semantic textual content of tweets due to the ability to deal equally with not only nouns, verbs, adjectives and adverbs, but also perception based fuzzy words. The proposed methodology first creates a set of prototypical event related tweets and a control group of tweets from a data source, then calculates the semantic similarity against an event dataset compiled from tweets issued during the 2011 London riots. The dataset of tweets contained a proportion of tweets that the Guardian Newspaper publically released that were attributed to 200 influential Twitter users during the actual riot. The effects of changing the semantic similarity threshold are investigated in order to evaluate if Twitter tweets can be used in conjunction with fuzzy short text similarity measures and prototypical event related tweets to determine if an event is more likely to occur. By looking at the increase in frequency of tweets in the dataset, over a certain similarity threshold when matched with prototypical event tweets about riots, the results have shown that a potential future event can be detected.

Identification of the delivery of cognitive behavioural therapy for psychosis (CBTp) using a cross-sectional sample from electronic health records and open-text information in a large UK-based mental health case register

Objective Our primary objective was to identify cognitive behavioural therapy (CBT) delivery for people with psychosis (CBTp) using an automated method in a large electronic health record database. We also examined what proportion of service users with a diagnosis of psychosis were recorded as having received CBTp within their episode of care during defined time periods provided by early intervention or promoting recovery community services for people with psychosis, compared with published audits and whether demographic characteristics differentially predicted the receipt of CBTp. Methods Both free text using natural language processing (NLP) techniques and structured methods of identifying CBTp were combined and evaluated for positive predictive value (PPV) and sensitivity. Using inclusion criteria from two published audits, we identified anonymised cross-sectional samples of 2579 and 2308 service users respectively with a case note diagnosis of schizophrenia or psychosis for further analysis. Results The method achieved PPV of 95% and sensitivity of 96%. Using the National Audit of Schizophrenia 2 criteria, 34.6% service users were identified as ever having received at least one session and 26.4% at least two sessions of CBTp; these are higher percentages than previously reported by manual audit of a sample from the same trust that returned 20.0%. In the fully adjusted analysis, CBTp receipt was significantly (p<0.05) more likely in younger patients, in white and other when compared with black ethnic groups and patients with a diagnosis of other schizophrenia spectrum and schizoaffective disorder when compared with schizophrenia. Conclusions The methods presented here provided a potential method for evaluating delivery of CBTp on a large scale, providing more scope for routine monitoring, cross-site comparisons and the promotion of equitable access.

Fuzzy ontologies in semantic similarity measures

Ontologies are a fundamental part of the development of short text semantic similarity measures. The most known ontology used within the field was developed from the lexical database known as WordNet which is used as a semantic resource for determining word similarity using the semantic distance between words. The original WordNet does not include in its hierarchy fuzzy words - those which are subjective to humans and often context dependent. The recent development of fuzzy semantic similarity measures requires research into the development of different ontological structures which are suitable for the representation of fuzzy categories of words where quantification of words is undertaken by human participations. This paper proposes two different fuzzy ontology structures which are based on a human quantified scale for a collection of fuzzy words across six fuzzy categories. The methodology of ontology creation utilizes human participants to populate fuzzy categories and quantify fuzzy words. Each ontology is evaluated within a known fuzzy semantic similarity measure and experiments are conducted using human participants and two benchmark fuzzy word datasets. Correlations with human similarity ratings show only one ontological structure was naturally representative of human perceptions of fuzzy words.