Text mining applications in psychiatry: a systematic literature review

Predicting treatment resistance in schizophrenia patients: Machine learning highlights the role of early pathophysiologic features

Detecting patients with a high-risk profile for treatment-resistant schizophrenia (TRS) can be beneficial for implementing individually adapted therapeutic strategies and better understanding the TRS etiology. The aim of this study was to explore, with machine learning methods, the impact of demographic and clinical patient characteristics on TRS prediction, for already established risk factors and unexplored ones. This was a retrospective study of 500 patients admitted during 2020 to the University Hospital Group for Paris Psychiatry. We hypothesized potential TRS risk factors. The selected features were coded into structured variables in a new dataset, by processing patients discharge summaries and medical narratives with natural-language processing methods. We compared three machine learning models (XGBoost, logistic elastic net regression, logistic regression without regularization) for predicting TRS outcome. We analysed feature impact on the models, suggesting the following factors as markers of a high-risk TRS profile: early age at first contact with psychiatry, antipsychotic treatment interruptions due to non-adherence, absence of positive symptoms at baseline, educational problems and adolescence mental disorders in the personal psychiatric history. Specifically, we found a significant association with TRS outcome for age at first contact with psychiatry and medication non-adherence. Our findings on TRS risk factors are consistent with the review of the literature and suggest potential in using early pathophysiologic features for TRS prediction. Results were encouraging with the use of natural-langage processing techniques to leverage raw data provided by discharge summaries, combined with machine leaning models. These findings are a promising step for helping clinicians adapt their guidelines to early detection of TRS.

Predicting which patients with cancer will see a psychiatrist or counsellor from their initial oncology consultation document using natural language processing

BACKGROUND

Patients with cancer often have unmet psychosocial needs. Early detection of who requires referral to a counsellor or psychiatrist may improve their care. This work used natural language processing to predict which patients will see a counsellor or psychiatrist from a patient's initial oncology consultation document. We believe this is the first use of artificial intelligence to predict psychiatric outcomes from non-psychiatric medical documents.

METHODS

This retrospective prognostic study used data from 47,625 patients at BC Cancer. We analyzed initial oncology consultation documents using traditional and neural language models to predict whether patients would see a counsellor or psychiatrist in the 12 months following their initial oncology consultation.

RESULTS

Here, we show our best models achieved a balanced accuracy (receiver-operating-characteristic area-under-curve) of 73.1% (0.824) for predicting seeing a psychiatrist, and 71.0% (0.784) for seeing a counsellor. Different words and phrases are important for predicting each outcome.

CONCLUSION

These results suggest natural language processing can be used to predict psychosocial needs of patients with cancer from their initial oncology consultation document. Future research could extend this work to predict the psychosocial needs of medical patients in other settings.

Impact of Collaborative Documentation on Completeness and Length of Clinical Notes in Behavioral Health Settings

OBJECTIVE

This study aimed to examine differences in community mental health visit notes before and after initiation of collaborative documentation, a practice in which clinicians and consumers jointly document clinical encounters.

METHODS

Using a clinical informatics approach, the authors sampled visit notes (N=1,875) from nine providers in one mental health clinic. The authors compared notes from before and after the implementation of collaborative documentation by using fixed-effects regression models, controlling for therapist-level effects.

RESULTS

Significant changes in visit note structure were found after the implementation of collaborative documentation. Most sections (N=6 of 10) contained more information (i.e., higher word and character counts) after collaborative documentation implementation, but sections describing a client's feelings were less likely to have any content (OR=0.01, p<0.001).

CONCLUSIONS

These findings demonstrate that collaborative documentation influences clinical notes, providing much-needed research about a widely adopted practice in community mental health settings.

Large Language Models in Neurology Research and Future Practice

Recent advancements in generative artificial intelligence, particularly using large language models (LLMs), are gaining increased public attention. We provide a perspective on the potential of LLMs to analyze enormous amounts of data from medical records and gain insights on specific topics in neurology. In addition, we explore use cases for LLMs, such as early diagnosis, supporting patient and caregivers, and acting as an assistant for clinicians. We point to the potential ethical and technical challenges raised by LLMs, such as concerns about privacy and data security, potential biases in the data for model training, and the need for careful validation of results. Researchers must consider these challenges and take steps to address them to ensure that their work is conducted in a safe and responsible manner. Despite these challenges, LLMs offer promising opportunities for improving care and treatment of various neurologic disorders.

Natural language processing for identification of refractory status epilepticus in children

OBJECTIVE

Pediatric status epilepticus is one of the most frequent pediatric emergencies, with high mortality and morbidity. Utilizing electronic health records (EHRs) permits analysis of care approaches and disease outcomes at a lower cost than prospective research. However, reviewing EHR manually is time intensive. We aimed to compare refractory status epilepticus (rSE) cases identified by human EHR review with a natural language processing (NLP)-assisted rSE screen followed by a manual review.

METHODS

We used the NLP screening tool Document Review Tool (DrT) to generate regular expressions, trained a bag-of-words NLP classifier on EHRs from 2017 to 2019, and then tested our algorithm on data from February to December 2012. We compared results from manual review to NLP-assisted search followed by manual review.

RESULTS

Our algorithm identified 1528 notes in the test set. After removing notes pertaining to the same event by DrT, the user reviewed a total number of 400 notes to find patients with rSE. Within these 400 notes, we identified 31 rSE cases, including 12 new cases not found in manual review, and 19 of the 20 previously identified cases. The NLP-assisted model found 31 of 32 cases, with a sensitivity of 96.88% (95% CI = 82%-99.84%), whereas manual review identified 20 of 32 cases, with a sensitivity of 62.5% (95% CI = 43.75%-78.34%).

SIGNIFICANCE

DrT provided a highly sensitive model compared to human review and an increase in patient identification through EHRs. The use of DrT is a suitable application of NLP for identifying patients with a history of recent rSE, which ultimately contributes to the implementation of monitoring techniques and treatments in near real time.

Finding the Best Match - a Case Study on the (Text-)Feature and Model Choice in Digital Mental Health Interventions

With the need for psychological help long exceeding the supply, finding ways of scaling, and better allocating mental health support is a necessity. This paper contributes by investigating how to best predict intervention dropout and failure to allow for a need-based adaptation of treatment. We systematically compare the predictive power of different text representation methods (metadata, TF-IDF, sentiment and topic analysis, and word embeddings) in combination with supplementary numerical inputs (socio-demographic, evaluation, and closed-question data). Additionally, we address the research gap of which ML model types - ranging from linear to sophisticated deep learning models - are best suited for different features and outcome variables. To this end, we analyze nearly 16.000 open-text answers from 849 German-speaking users in a Digital Mental Health Intervention (DMHI) for stress. Our research proves that - contrary to previous findings - there is great promise in using neural network approaches on DMHI text data. We propose a task-specific LSTM-based model architecture to tackle the challenge of long input sequences and thereby demonstrate the potential of word embeddings (AUC scores of up to 0.7) for predictions in DMHIs. Despite the relatively small data set, sequential deep learning models, on average, outperform simpler features such as metadata and bag-of-words approaches when predicting dropout. The conclusion is that user-generated text of the first two sessions carries predictive power regarding patients' dropout and intervention failure risk. Furthermore, the match between the sophistication of features and models needs to be closely considered to optimize results, and additional non-text features increase prediction results.

Supplementary Information

The online version contains supplementary material available at 10.1007/s41666-023-00148-z.

Hybrid deep learning models with multi-classification investor sentiment to forecast the prices of China's leading stocks

The prediction of stock prices has long been a captivating subject in academic research. This study aims to forecast the prices of prominent stocks in five key industries of the Chinese A-share market by leveraging the synergistic power of deep learning techniques and investor sentiment analysis. To achieve this, a sentiment multi-classification dataset is for the first time constructed for China's stock market, based on four types of sentiments in modern psychology. The significant heterogeneity of sentiment changes in the sectors' leading stock markets is trained and mined using the Bi-LSTM-ATT model. The impact of multi-classification investor sentiment on stock price prediction was analyzed using the CNN-Bi-LSTM-ATT model. It finds that integrating sentiment indicators into the prediction of industry leading stock prices can enhance the accuracy of the model. Drawing upon four fundamental sentiment types derived from modern psychology, our dataset provides a comprehensive framework for analyzing investor sentiment and its impact on forecasting the stock prices of China's A-share market.

Using patient comments from a standardised experience survey to investigate their perceptions and prioritise improvement actions: a thematic and syntactic analysis

BACKGROUND

Although patient experience surveys flourish in many countries with the aim to improve quality of care, questions remain concerning their ability to become effective drivers of change within institutions. The patient comments from the French national patient experience hospital survey were analysed using an innovative structured approach to characterise patient experience and identify field actions for the institutions.

METHODS

The comments were taken from the two open-ended questions comprised in the patient experience survey of the Hospices Civils de Lyon between 2018 and 2019. The comments analysis methodology consisted in three steps: thematic analysis; syntactic analysis; generation of statistics for the creation of a patient journey and prioritisation of sub-themes. The STROBE statement checklist was followed.

RESULTS

Over a year, 79.7% of the 7 362 respondents left at least one comment at the end of the survey and were included in the study, for a total of 5 868 surveys and 10 061 comments. These led to the identification of 28 general themes and 184 specific sub-themes. From the patient journey created, 23 sub-themes were prioritised and gathered into four key categories: relationship between patient and staff; environment; surgery and pain management; information and care coordination. For each of them, the actions and expectations formulated by the respondents were described.

CONCLUSIONS

The analysis of patient comments obtained from a standardised survey allowed to characterise the patient journey using data that describes patient experience, enabling a prioritisation of actions aiming to improve practice and quality of care at the institution, department, and staff level.

A Review of and Roadmap for Data Science and Machine Learning for the Neuropsychiatric Phenotype of Autism

Autism spectrum disorder (autism) is a neurodevelopmental delay that affects at least 1 in 44 children. Like many neurological disorder phenotypes, the diagnostic features are observable, can be tracked over time, and can be managed or even eliminated through proper therapy and treatments. However, there are major bottlenecks in the diagnostic, therapeutic, and longitudinal tracking pipelines for autism and related neurodevelopmental delays, creating an opportunity for novel data science solutions to augment and transform existing workflows and provide increased access to services for affected families. Several efforts previously conducted by a multitude of research labs have spawned great progress toward improved digital diagnostics and digital therapies for children with autism. We review the literature on digital health methods for autism behavior quantification and beneficial therapies using data science. We describe both case-control studies and classification systems for digital phenotyping. We then discuss digital diagnostics and therapeutics that integrate machine learning models of autism-related behaviors, including the factors that must be addressed for translational use. Finally, we describe ongoing challenges and potential opportunities for the field of autism data science. Given the heterogeneous nature of autism and the complexities of the relevant behaviors, this review contains insights that are relevant to neurological behavior analysis and digital psychiatry more broadly.

[E-Satis : A new method for analysis of Patient-Reported Outcome Measures (PROMs)]

OBJECTIVE

Almost 80% of the patients responding to the nationwide French patient experience and satisfaction survey (e-Satis) provided free text comments. The objective of this article is to describe an innovative methodology for analysis of this qualitative data.

METHODOLOGY

This methodological approach is based on analysis of qualitative data from the comments (verbatims) of respondents to the e-Satis survey. Analysis of the verbatims consists in three main steps: (i) analysis of the meaning of the words, with constitution of a thematic dictionary through exploratory research without preconceived notions; (ii) analysis of the syntax, i.e., the way in which the ideas are articulated, which will enable calculation of a linguistic indicator of speakers' involvement in their speech; (iii) production of statistics and characterisation of the themes, which will include three indicators: occurrence of the themes, the average satisfaction shown in the respondents' discourse, and the positive and negative involvement with which they express themselves. Given these results, a priority matrix of four categories of action is established: strong points, priority areas, good practices, and weak signals.

RESULTS

This methodological approach was applied to 5868 e-Satis questionnaires out of a total of 10,061 verbatims by respondents hospitalised at the Hospices Civils de Lyon between 2018 and 2019. The analysis identified 28 major themes with 184 sub-themes. An extract is presented in this article for illustration purposes.

DISCUSSION

A methodological approach based on analysis of qualitative data will enable transformation of unstructured data (verbatims) into measurable and comparable data. This methodology is structured to overcome the limitations of closed questions; open questions allow respondents to describe their experiences and perceptions in their own words. Moreover, it is a first step toward comparability of results over time with those of other establishments. This approach is unique in France on account of (a) its exploratory thematic research without preconceived notions and (b) its syntactic analysis of verbatims.

CONCLUSIONS

This verbatim analysis methodology should enable precise and operational characterization of Patient Experience and induce prioritized improvement actions in healthcare institutions.

Comparison of relative areas of interest between major depression disorder and postpartum depression

INTRODUCTION

Postpartum depression (PPD) is defined as a major depressive disorder (MDD) beginning after childbirth. Wide debates aim to better understand PPD's specificities compared with MDD. One of the keys in differentiating PPD from MDD is to systematically study scientific "Areas Of Interest" (AOIs) of these disorders.

METHODS

In November 2021, we performed an extraction and textual computational analysis of associated terms for PPD and MDD, using the biomedical database PubMed. We performed an undirected lexical network analysis to map the 150 first terms in space. Then, we used an unsupervised machine learning technique to detect word patterns and automatically cluster AOIs with a topic-modeling analysis.

RESULTS

We identified 30,000 articles of the 554,724 articles for MDD and 15,642 articles for PPD. Four AOIs were detected in the MDD network: mood disorders and their treatments, risk factors, consequences and quality of life, and mental health and comorbidities. Five AOIs were detected in the PPD network: mood disorders and treatments, risk factors, consequences and child health, patient's background, and the challenges of screening.

DISCUSSION AND CONCLUSION

Limitations are both methodological, in particular due to the qualitative interpretation of AOIs, and are also related to the difficult transferability of these research results to the clinical practice. The partial overlap between AOIs for MDD and for PPD suggest that the latter is a particular form of the former.

Transforming epilepsy research: A systematic review on natural language processing applications

Despite improved ancillary investigations in epilepsy care, patients' narratives remain indispensable for diagnosing and treatment monitoring. This wealth of information is typically stored in electronic health records and accumulated in medical journals in an unstructured manner, thereby restricting complete utilization in clinical decision-making. To this end, clinical researchers increasing apply natural language processing (NLP)-a branch of artificial intelligence-as it removes ambiguity, derives context, and imbues standardized meaning from free-narrative clinical texts. This systematic review presents an overview of the current NLP applications in epilepsy and discusses the opportunities and drawbacks of NLP alongside its future implications. We searched the PubMed and Embase databases with a "natural language processing" and "epilepsy" query (March 4, 2022) and included original research articles describing the application of NLP techniques for textual analysis in epilepsy. Twenty-six studies were included. Fifty-eight percent of these studies used NLP to classify clinical records into predefined categories, improving patient identification and treatment decisions. Other applications of NLP had structured clinical information retrieval from electronic health records, scientific papers, and online posts of patients. Challenges and opportunities of NLP applications for enhancing epilepsy care and research are discussed. The field could further benefit from NLP by replicating successes in other health care domains, such as NLP-aided quality evaluation for clinical decision-making, outcome prediction, and clinical record summarization.

Recent advances in psychoradiology

Psychiatry, as a field, lacks objective markers for diagnosis, progression, treatment planning, and prognosis, in part due to difficulties studying the brain in vivo, and diagnoses are based on self-reported symptoms and observation of patient behavior and cognition. Rapid advances in brain imaging techniques allow clinical investigators to noninvasively quantify brain features at the structural, functional, and molecular levels. Psychoradiology is an emerging discipline at the intersection of psychiatry and radiology. Psychoradiology applies medical imaging technologies to psychiatry and promises not only to improve insight into structural and functional brain abnormalities in patients with psychiatric disorders but also to have potential clinical utility. We searched for representative studies related to recent advances in psychoradiology through May 1, 2022, and conducted a selective review of 165 references, including 75 research articles. We summarize the novel dynamic imaging processing methods to model brain networks and present imaging genetics studies that reveal the relationship between various neuroimaging endophenotypes and genetic markers in psychiatric disorders. Furthermore, we survey recent advances in psychoradiology, with a focus on future psychiatric diagnostic approaches with dimensional analysis and a shift from group-level to individualized analysis. Finally, we examine the application of machine learning in psychoradiology studies and the potential of a novel option for brain stimulation treatment based on psychoradiological findings in precision medicine. Here, we provide a summary of recent advances in psychoradiology research, and we hope this review will help guide the practice of psychoradiology in the scientific and clinical fields.

Machine learning in the prediction of postpartum depression: A review

BACKGROUND

Current screening options in the setting of postpartum depression (PPD) are firmly rooted in self-report symptom-based tools. The implementation of the modern machine learning (ML) approaches might, in this context, represent a way to refine patient screening by precisely identifying possible PPD predictors and, subsequently, a population at risk of developing the disease, in an effort to lower its morbidity, mortality and its economic burden.

METHODS

We performed a bibliographic search on PubMed and Embase looking for studies aimed at the identification of PPD predictors using ML techniques.

RESULTS

Among the 482 articles retrieved, 11 met the inclusion criteria. The most used algorithm was the support vector machine. Notably, all studies reached an area under the curve above 0.7, ultimately suggesting that the prediction of PPD could be feasible. Variables obtained from sociodemographic and clinical aspects (psychiatric and gynecological factors) seem to be the most reliable. Only three studies employed biological variables, in the form of blood, genetic and epigenetic predictors, while no study employed imaging techniques.

LIMITATIONS

The literature on PPD prediction via ML techniques is currently scarce, with most studies employing different variables selection and ML algorithms, ultimately reducing the generalizability of the results.

CONCLUSIONS

The identification of a population at risk of developing PPD might be feasible with current technology and clinical knowledge. Further studies are necessary to clarify how such an approach could be implemented into clinical practice.

Spontaneously generated online patient experience data - how and why is it being used in health research: an umbrella scoping review

PURPOSE

Social media has led to fundamental changes in the way that people look for and share health related information. There is increasing interest in using this spontaneously generated patient experience data as a data source for health research. The aim was to summarise the state of the art regarding how and why SGOPE data has been used in health research. We determined the sites and platforms used as data sources, the purposes of the studies, the tools and methods being used, and any identified research gaps.

METHODS

A scoping umbrella review was conducted looking at review papers from 2015 to Jan 2021 that studied the use of SGOPE data for health research. Using keyword searches we identified 1759 papers from which we included 58 relevant studies in our review.

RESULTS

Data was used from many individual general or health specific platforms, although Twitter was the most widely used data source. The most frequent purposes were surveillance based, tracking infectious disease, adverse event identification and mental health triaging. Despite the developments in machine learning the reviews included lots of small qualitative studies. Most NLP used supervised methods for sentiment analysis and classification. Very early days, methods need development. Methods not being explained. Disciplinary differences - accuracy tweaks vs application. There is little evidence of any work that either compares the results in both methods on the same data set or brings the ideas together.

CONCLUSION

Tools, methods, and techniques are still at an early stage of development, but strong consensus exists that this data source will become very important to patient centred health research.

Perceived barriers to psychiatric help-seeking in South Korea by age groups: text mining analyses of social media big data

BACKGROUND

The psychiatric treatment gap is substantial in Korea, implying barriers in seeking help.

OBJECTIVES

This study aims to explore barriers of seeing psychiatrists, expressed on the internet by age groups.

METHODS

A corpus of data was garnered extensively from internet communities, blogs and social network services from 1 January 2016 to 31 July 2019. Among the texts collected, texts containing words linked to psychiatry were selected. Then the corpus was dismantled into words by using natural language processing. Words linked to barriers to seeking help were identified and classified. Then the words from web communities that we were able to identify the age groups were additionally organized by age groups.

RESULTS

97,730,360 articles were identified and 6,097,369 were included in the analysis. Words implying the barriers were selected and classified into four groups of structural discrimination, public prejudice, low accessibility, and adverse drug effects. Structural discrimination was the greatest barrier occupying 34%, followed by public prejudice (27.8%), adverse drug effects (18.6%), and cost/low accessibility (16.1%). In the analysis by age groups, structural discrimination caused teenagers (51%), job seekers (64%) and mothers with children (43%) the most concern. In contrast, the public prejudice (49%) was the greatest barriers in the senior group.

CONCLUSIONS

Although structural discrimination may most contribute to barriers to visiting psychiatrists in Korea, variation by generations may exist. Along with the general attempt to tackle the discrimination, customized approach might be needed.

Natural Language Processing and Machine Learning Supporting the Work of a Psychologist and Its Evaluation on the Example of Support for Psychological Diagnosis of Anorexia

Objective: This study sought to address the use of computer-aided diagnosis and therapy for anorexia nervosa. This paper presents the means by which the use of natural language processing methods can augment the work of psychologists. Method: We evaluated this method based on its efficacy when diagnosing anorexia nervosa. Using natural language processing and machine learning, we developed methods for analyzing five basic emotions, analyzing a patient’s body perception, and detecting six potential areas of difficulties for computer support of psychological diagnosis of anorexia. We surveyed 43 psychologists to obtain feedback on these tools. Results: We evaluated efficacy in terms of patient relationship, substantive aspects of the diagnosis, and diagnostic procedures. In terms of patient relationship, we found a noticeable decrease in the patient’s resistance and better support in verifying the substantive scope of the diagnostic thesis. Discussion: The presented methods can be a supporting tool for monitoring the diagnostic process and increasing the degree of self-diagnosis and self-reflection by the patient. This tool can increase the accuracy of the diagnostic process by reducing patient resistance. This will increase access to the patient’s psychopathology.

Improving Web-Based Treatment Intake for Multiple Mental and Substance Use Disorders by Text Mining and Machine Learning: Algorithm Development and Validation

BACKGROUND

Text mining and machine learning are increasingly used in mental health care practice and research, potentially saving time and effort in the diagnosis and monitoring of patients. Previous studies showed that mental disorders can be detected based on text, but they focused on screening for a single predefined disorder instead of multiple disorders simultaneously.

OBJECTIVE

The aim of this study is to develop a Dutch multi-class text-classification model to screen for a range of mental disorders to refer new patients to the most suitable treatment.

METHODS

On the basis of textual responses of patients (N=5863) to a questionnaire currently used for intake and referral, a 7-class classifier was developed to distinguish among anxiety, panic, posttraumatic stress, mood, eating, substance use, and somatic symptom disorders. A linear support vector machine was fitted using nested cross-validation grid search.

RESULTS

The highest classification rate was found for eating disorders (82%). The scores for panic (55%), posttraumatic stress (52%), mood (50%), somatic symptom (50%), anxiety (35%), and substance use disorders (33%) were lower, likely because of overlapping symptoms. The overall classification accuracy (49%) was reasonable for a 7-class classifier.

CONCLUSIONS

A classification model was developed that could screen text for multiple mental health disorders. The screener resulted in an additional outcome score that may serve as input for a formal diagnostic interview and referral. This may lead to a more efficient and standardized intake process.

Language as a Window Into the Altered State of Consciousness Elicited by Psychedelic Drugs

Psychedelics are drugs capable of eliciting profound alterations in the subjective experience of the users, sometimes with long-lasting consequences. Because of this, psychedelic research tends to focus on human subjects, given their capacity to construct detailed narratives about the contents of their consciousness experiences. In spite of its relevance, the interaction between serotonergic psychedelics and language production is comparatively understudied in the recent literature. This review is focused on two aspects of this interaction: how the acute effects of psychedelic drugs impact on speech organization regardless of its semantic content, and how to characterize the subjective effects of psychedelic drugs by analyzing the semantic content of written retrospective reports. We show that the computational characterization of language production is capable of partially predicting the therapeutic outcome of individual experiences, relate the effects elicited by psychedelics with those associated with other altered states of consciousness, draw comparisons between the psychedelic state and the symptomatology of certain psychiatric disorders, and investigate the neurochemical profile and mechanism of action of different psychedelic drugs. We conclude that researchers studying psychedelics can considerably expand the range of their potential scientific conclusions by analyzing brief interviews obtained before, during and after the acute effects. Finally, we list a series of questions and open problems that should be addressed to further consolidate this approach.

iHealth: The ethics of artificial intelligence and big data in mental healthcare

The concept of intelligent health (iHealth) in mental healthcare integrates artificial intelligence (AI) and Big Data analytics. This article is an attempt to outline ethical aspects linked to iHealth by focussing on three crucial elements that have been defined in the literature: self-monitoring, ecological momentary assessment (EMA), and data mining. The material for the analysis was obtained by a database search. Studies and reviews providing outcome data for each of the three elements were analyzed. An ethical framing of the results was conducted that shows the chances and challenges of iHealth. The synergy between self-monitoring, EMA, and data mining might enable the prevention of mental illness, the prediction of its onset, the personalization of treatment, and the participation of patients in the treatment process. Challenges arise when it comes to the autonomy of users, privacy and data security of users, and potential bias.

Experiences of Siblings of Children With Neurodevelopmental Disorders: Comparing Qualitative Analysis and Machine Learning to Study Narratives

INTRODUCTION

Relatively few studies have focused on the wellbeing, experiences and needs of the siblings of children with a psychiatric diagnosis. However, the studies that have been conducted suggest that the impact of such circumstances on these siblings is significant. Studying narratives of diagnosed children or relatives has proven to be a successful approach to gain insights that could help improve care. Only a few attempts have been made to study narratives in psychiatry utilizing a machine learning approach.

METHOD

In this current study, 13 narratives of the experiences of siblings of children with a neurodevelopmental disorders were collected through largely unstructured interviews. The interviews were analyzed using the traditional qualitative, hermeneutic phenomenology method as well as latent Dirichlet allocation (LDA), an unsupervised machine learning method clustering words from documents into topics. One aim of this study was to evaluate the experiences of the siblings in order to find leads to improve care and support for these siblings. Furthermore, the outcomes of both analyses were compared to evaluate the role of machine learning in analyzing narratives.

RESULTS

Qualitative analysis of the interviews led to the formulation of nine main themes: confrontation with conflicts, coping strategies siblings, need for rest and time for myself, need for support and attention from personal circle, wish for normality, influence on personal choices and possibilities for development, doing things together, recommendations and advices, ambivalence and loyalty. Using unsupervised machine learning (LDA) 24 topics were formed that mostly overlapped with the qualitative themes found. Both the qualitative analysis and the LDA analysis detected themes that were unique to the respective analysis.

CONCLUSION

The present study found that studying narratives of siblings of children with a neurodevelopmental disorder contributes to a better understanding of the subjects' experiences. Siblings cope with ambivalent feelings toward their brother or sister and this emotional conflict often leads to adapted behavior. Several coping strategies are developed to deal with the behavior of their brother or sister like seeking support or ignoring. Devoted support, time and attention from close relatives, especially parents, is needed. The LDA analysis didn't appear useful to distract meaning and context from the narratives, but it was proposed that machine learning could be a valuable and quick addition to the traditional qualitative methods by finding overlooked topics and giving a rudimental overview of topics in narratives.

Natural language processing in clinical neuroscience and psychiatry: A review

Natural language processing (NLP) is rapidly becoming an important topic in the medical community. The ability to automatically analyze any type of medical document could be the key factor to fully exploit the data it contains. Cutting-edge artificial intelligence (AI) architectures, particularly machine learning and deep learning, have begun to be applied to this topic and have yielded promising results. We conducted a literature search for 1,024 papers that used NLP technology in neuroscience and psychiatry from 2010 to early 2022. After a selection process, 115 papers were evaluated. Each publication was classified into one of three categories: information extraction, classification, and data inference. Automated understanding of clinical reports in electronic health records has the potential to improve healthcare delivery. Overall, the performance of NLP applications is high, with an average F1-score and AUC above 85%. We also derived a composite measure in the form of Z-scores to better compare the performance of NLP models and their different classes as a whole. No statistical differences were found in the unbiased comparison. Strong asymmetry between English and non-English models, difficulty in obtaining high-quality annotated data, and train biases causing low generalizability are the main limitations. This review suggests that NLP could be an effective tool to help clinicians gain insights from medical reports, clinical research forms, and more, making NLP an effective tool to improve the quality of healthcare services.

Identifying subtypes of depression in clinician-annotated text: a retrospective cohort study

Current criteria for depression are imprecise and do not accurately characterize its distinct clinical presentations. As a result, its diagnosis lacks clinical utility in both treatment and research settings. Data-driven efforts to refine criteria have typically focused on a limited set of symptoms that do not reflect the disorder's heterogeneity. By contrast, clinicians often write about patients in depth, creating descriptions that may better characterize depression. However, clinical text is not commonly used to this end. Here we show that clinically relevant depressive subtypes can be derived from unstructured electronic health records. Five subtypes were identified amongst 18,314 patients with depression treated at a large mental healthcare provider by using unsupervised machine learning: severe-typical, psychotic, mild-typical, agitated, and anergic-apathetic. Subtypes were used to place patients in groups for validation; groups were found to be associated with future outcomes and characteristics that were consistent with the subtypes. These associations suggest that these categorizations are actionable due to their validity with respect to disease prognosis. Moreover, they were derived with automated techniques that might theoretically be widely implemented, allowing for future analyses in more varied populations and settings. Additional research, especially with respect to treatment response, may prove useful in further evaluation.

Identifying themes for assessing cancer-related cognitive impairment identified by topic modeling and qualitative content analysis of public online comments (Preprint)

Background

Cancer-related cognitive impairment (CRCI) is a common and significant adverse effect of cancer and its therapies. However, its definition and assessment remain difficult due to limitations of currently available measurement tools.

Objective

This study aims to evaluate qualitative themes related to the cognitive effects of cancer to help guide development of assessments that are more specific than what is currently available.

Methods

We applied topic modeling and inductive qualitative content analysis to 145 public online comments related to cognitive effects of cancer.

Results

Topic modeling revealed 2 latent topics that we interpreted as representing internal and external factors related to cognitive effects. These findings lead us to hypothesize regarding the potential contribution of locus of control to CRCI. Content analysis suggested several major themes including symptoms, emotional/psychological impacts, coping, “chemobrain” is real, change over time, and function. There was some conceptual overlap between the 2 methods regarding internal and external factors related to patient experiences of cognitive effects.

Conclusions

Our findings indicate that coping mechanisms and locus of control may be important themes to include in assessments of CRCI. Future directions in this field include prospective acquisition of free-text responses to guide development of assessments that are more sensitive and specific to cognitive function in patients with cancer.

Spontaneously Generated Online Patient Experience of Modafinil: A Qualitative and NLP Analysis

Objective: To compare the findings from a qualitative and a natural language processing (NLP) based analysis of online patient experience posts on patient experience of the effectiveness and impact of the drug Modafinil.

Methods: Posts (n = 260) from 5 online social media platforms where posts were publicly available formed the dataset/corpus. Three platforms asked posters to give a numerical rating of Modafinil. Thematic analysis: data was coded and themes generated. Data were categorized into PreModafinil, Acquisition, Dosage, and PostModafinil and compared to identify each poster's own view of whether taking Modafinil was linked to an identifiable outcome. We classified this as positive, mixed, negative, or neutral and compared this with numerical ratings. NLP: Corpus text was speech tagged and keywords and key terms extracted. We identified the following entities: drug names, condition names, symptoms, actions, and side-effects. We searched for simple relationships, collocations, and co-occurrences of entities. To identify causal text, we split the corpus into PreModafinil and PostModafinil and used n-gram analysis. To evaluate sentiment, we calculated the polarity of each post between −1 (negative) and +1 (positive). NLP results were mapped to qualitative results.

Results: Posters had used Modafinil for 33 different primary conditions. Eight themes were identified: the reason for taking (condition or symptom), impact of symptoms, acquisition, dosage, side effects, other interventions tried or compared to, effectiveness of Modafinil, and quality of life outcomes. Posters reported perceived effectiveness as follows: 68% positive, 12% mixed, 18% negative. Our classification was consistent with poster ratings. Of the most frequent 100 keywords/keyterms identified by term extraction 88/100 keywords and 84/100 keyterms mapped directly to the eight themes. Seven keyterms indicated negation and temporal states. Sentiment was as follows 72% positive sentiment 4% neutral 24% negative. Matching of sentiment between the qualitative and NLP methods was accurate in 64.2% of posts. If we allow for one category difference matching was accurate in 85% of posts.

Conclusions: User generated patient experience is a rich resource for evaluating real world effectiveness, understanding patient perspectives, and identifying research gaps. Both methods successfully identified the entities and topics contained in the posts. In contrast to current evidence, posters with a wide range of other conditions found Modafinil effective. Perceived causality and effectiveness were identified by both methods demonstrating the potential to augment existing knowledge.

Use of Automated Thematic Annotations for Small Data Sets in a Psychotherapeutic Context: Systematic Review of Machine Learning Algorithms

BACKGROUND

A growing body of literature has detailed the use of qualitative analyses to measure the therapeutic processes and intrinsic effectiveness of psychotherapies, which yield small databases. Nonetheless, these approaches have several limitations and machine learning algorithms are needed.

OBJECTIVE

The objective of this study is to conduct a systematic review of the use of machine learning for automated text classification for small data sets in the fields of psychiatry, psychology, and social sciences. This review will identify available algorithms and assess if automated classification of textual entities is comparable to the classification done by human evaluators.

METHODS

A systematic search was performed in the electronic databases of Medline, Web of Science, PsycNet (PsycINFO), and Google Scholar from their inception dates to 2021. The fields of psychiatry, psychology, and social sciences were selected as they include a vast array of textual entities in the domain of mental health that can be reviewed. Additional records identified through cross-referencing were used to find other studies.

RESULTS

This literature search identified 5442 articles that were eligible for our study after the removal of duplicates. Following abstract screening, 114 full articles were assessed in their entirety, of which 107 were excluded. The remaining 7 studies were analyzed. Classification algorithms such as naive Bayes, decision tree, and support vector machine classifiers were identified. Support vector machine is the most used algorithm and best performing as per the identified articles. Prediction classification scores for the identified algorithms ranged from 53%-91% for the classification of textual entities in 4-7 categories. In addition, 3 of the 7 studies reported an interjudge agreement statistic; these were consistent with agreement statistics for text classification done by human evaluators.

CONCLUSIONS

A systematic review of available machine learning algorithms for automated text classification for small data sets in several fields (psychiatry, psychology, and social sciences) was conducted. We compared automated classification with classification done by human evaluators. Our results show that it is possible to automatically classify textual entities of a transcript based solely on small databases. Future studies are nevertheless needed to assess whether such algorithms can be implemented in the context of psychotherapies.

Forecasting the Suitability of Online Mental Health Information for Effective Self-Care Developing Machine Learning Classifiers Using Natural Language Features

Background: Online mental health information represents important resources for people living with mental health issues. Suitability of mental health information for effective self-care remains understudied, despite the increasing needs for more actionable mental health resources, especially among young people. Objective: We aimed to develop Bayesian machine learning classifiers as data-based decision aids for the assessment of the actionability of credible mental health information for people with mental health issues and diseases. Methods: We collected and classified creditable online health information on mental health issues into generic mental health (GEN) information and patient-specific (PAS) mental health information. GEN and PAS were both patient-oriented health resources developed by health authorities of mental health and public health promotion. GENs were non-classified online health information without indication of targeted readerships; PASs were developed purposefully for specific populations (young, elderly people, pregnant women, and men) as indicated by their website labels. To ensure the generalisability of our model, we chose to develop a sparse Bayesian machine learning classifier using Relevance Vector Machine (RVM). Results: Using optimisation and normalisation techniques, we developed a best-performing classifier through joint optimisation of natural language features and min-max normalisation of feature frequencies. The AUC (0.957), sensitivity (0.900), and specificity (0.953) of the best model were statistically higher (p < 0.05) than other models using parallel optimisation of structural and semantic features with or without feature normalisation. We subsequently evaluated the diagnostic utility of our model in the clinic by comparing its positive (LR+) and negative likelihood ratios (LR−) and 95% confidence intervals (95% C.I.) as we adjusted the probability thresholds with the range of 0.1 and 0.9. We found that the best pair of LR+ (18.031, 95% C.I.: 10.992, 29.577) and LR− (0.100, 95% C.I.: 0.068, 0.148) was found when the probability threshold was set to 0.45 associated with a sensitivity of 0.905 (95%: 0.867, 0.942) and specificity of 0.950 (95% C.I.: 0.925, 0.975). These statistical properties of our model suggested its applicability in the clinic. Conclusion: Our study found that PAS had significant advantage over GEN mental health information regarding information actionability, engagement, and suitability for specific populations with distinct mental health issues. GEN is more suitable for general mental health information acquisition, whereas PAS can effectively engage patients and provide more effective and needed self-care support. The Bayesian machine learning classifier developed provided automatic tools to support decision making in the clinic to identify more actionable resources, effective to support self-care among different populations.

DES-Tcell is a knowledgebase for exploring immunology-related literature

T-cells are a subtype of white blood cells circulating throughout the body, searching for infected and abnormal cells. They have multifaceted functions that include scanning for and directly killing cells infected with intracellular pathogens, eradicating abnormal cells, orchestrating immune response by activating and helping other immune cells, memorizing encountered pathogens, and providing long-lasting protection upon recurrent infections. However, T-cells are also involved in immune responses that result in organ transplant rejection, autoimmune diseases, and some allergic diseases. To support T-cell research, we developed the DES-Tcell knowledgebase (KB). This KB incorporates text- and data-mined information that can expedite retrieval and exploration of T-cell relevant information from the large volume of published T-cell-related research. This KB enables exploration of data through concepts from 15 topic-specific dictionaries, including immunology-related genes, mutations, pathogens, and pathways. We developed three case studies using DES-Tcell, one of which validates effective retrieval of known associations by DES-Tcell. The second and third case studies focuses on concepts that are common to Grave’s disease (GD) and Hashimoto’s thyroiditis (HT). Several reports have shown that up to 20% of GD patients treated with antithyroid medication develop HT, thus suggesting a possible conversion or shift from GD to HT disease. DES-Tcell found miR-4442 links to both GD and HT, and that miR-4442 possibly targets the autoimmune disease risk factor CD6, which provides potential new knowledge derived through the use of DES-Tcell. According to our understanding, DES-Tcell is the first KB dedicated to exploring T-cell-relevant information via literature-mining, data-mining, and topic-specific dictionaries.

Domestic Violence in Residential Care Facilities in New South Wales, Australia: A Text Mining Study

BACKGROUND AND OBJECTIVES

The police are often the first to attend domestic violence events in New South Wales (NSW), Australia, recording related details as structured information (e.g., date of the event, type of incident, premises type) and as text narratives which contain important information (e.g., mental health status, abuse types) for victims and perpetrators. This study examined the characteristics of victims and persons of interest (POIs) suspected and/or charged with perpetrating a domestic violence related crime in residential care facilities.

RESEARCH DESIGN AND METHODS

The study employed a text mining method that extracted key information from 700 police recorded domestic violence events in NSW residential care facilities.

RESULTS

Victims were mostly female (65.4%) and older adults (median age 80.3). POIs were predominantly male (67.0%) and were younger than the victims (median age 57.0). While low rates of mental illnesses were recorded (29.1% in victims; 17.4% in POIs), 'dementia' was the most common condition among POIs (55.7%) and victims (73.0%). 'Physical abuse' was the most common abuse type (80.2%) with 'bruising' the most common injury (36.8%). The most common relationship between perpetrator and victim was 'carer' (76.6%).

DISCUSSION AND IMPLICATIONS

These findings highlight the opportunity provided by police text-based data to provide insights into elder abuse within residential care facilities.

Using Natural Language Processing to Measure and Improve Quality of Diabetes Care: A Systematic Review

BACKGROUND

Real-world evidence research plays an increasingly important role in diabetes care. However, a large fraction of real-world data are "locked" in narrative format. Natural language processing (NLP) technology offers a solution for analysis of narrative electronic data.

METHODS

We conducted a systematic review of studies of NLP technology focused on diabetes. Articles published prior to June 2020 were included.

RESULTS

We included 38 studies in the analysis. The majority (24; 63.2%) described only development of NLP tools; the remainder used NLP tools to conduct clinical research. A large fraction (17; 44.7%) of studies focused on identification of patients with diabetes; the rest covered a broad range of subjects that included hypoglycemia, lifestyle counseling, diabetic kidney disease, insulin therapy and others. The mean F₁ score for all studies where it was available was 0.882. It tended to be lower (0.817) in studies of more linguistically complex concepts. Seven studies reported findings with potential implications for improving delivery of diabetes care.

CONCLUSION

Research in NLP technology to study diabetes is growing quickly, although challenges (e.g. in analysis of more linguistically complex concepts) remain. Its potential to deliver evidence on treatment and improving quality of diabetes care is demonstrated by a number of studies. Further growth in this area would be aided by deeper collaboration between developers and end-users of natural language processing tools as well as by broader sharing of the tools themselves and related resources.

Data Mining of Free-Text Responses: An Innovative Approach to Analyzing Patient Perspectives on Treatment for Chronic Rhinosinusitis with Nasal Polyps in a Phase IIa Proof-of-Concept Study for Dupilumab

PURPOSE

Patient perspective is an important and increasingly sought-after complement to clinical assessment. The aim of this study was to transcribe individual patients' experience of treatment in a dupilumab clinical trial through free-text responses with analysis using natural language processing (NLP) to obtain the unique perspective of patients on disease impact and unmet needs with existing treatment to inform future trial design.

PATIENTS AND METHODS

Patients with chronic rhinosinusitis with nasal polyps (CRSwNP) who were enrolled in a Phase IIa randomized controlled trial comparing dupilumab with placebo (NCT01920893) were invited to complete a self-assessment of treatment (SAT) tool at the end of treatment, asking, "What is your opinion on the treatment you had during the trial? What did you like or dislike about the treatment?" Free-text responses were analyzed for the overall cohort and according to treatment assignment using natural language processing including sentiment scoring. In a mixed-methods approach, quantitative patient-reported outcome (PRO) results were utilized to complement the qualitative analysis of free-text responses.

RESULTS

Of 60 patients enrolled in the study, 43 (71.6%) completed the SAT and responses from 37 patients were analyzed (placebo, n = 16; dupilumab, n = 21). Word analyses showed that the most common words were "smell," "improve," "staff," "great," "time," and "good." Across the whole cohort, "smell" was the most common symptom-related word. The words "smell" and "experience" were more likely to occur in patients treated with dupilumab. Patients treated with dupilumab also had more positive sentiment in their SAT responses than those who received placebo. The results from this qualitative analysis were reflected in quantitative PRO results.

CONCLUSION

"Smell" was important to patients with CRSwNP, highlighting its importance as a patient-centric efficacy outcome measure in the context of clinical trials in CRSwNP.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT01920893. Registered 12 August 2013, https://www.clinicaltrials.gov/ct2/show/NCT01920893.

Surveillance of Domestic Violence Using Text Mining Outputs From Australian Police Records

In Australia, domestic violence reports are mostly based on data from the police, courts, hospitals, and ad hoc surveys. However, gaps exist in reporting information such as victim injuries, mental health status and abuse types. The police record details of domestic violence events as structured information (e.g., gender, postcode, ethnicity), but also in text narratives describing other details such as injuries, substance use, and mental health status. However, the voluminous nature of the narratives has prevented their use for surveillance purposes. We used a validated text mining methodology on 492,393 police-attended domestic violence event narratives from 2005 to 2016 to extract mental health mentions on persons of interest (POIs) (individuals suspected/charged with a domestic violence offense) and victims, abuse types, and victim injuries. A significant increase was observed in events that recorded an injury type (28.3% in 2005 to 35.6% in 2016). The pattern of injury and abuse types differed between male and female victims with male victims more likely to be punched and to experience cuts and bleeding and female victims more likely to be grabbed and pushed and have bruises. The four most common mental illnesses (alcohol abuse, bipolar disorder, depression schizophrenia) were the same in male and female POIs. An increase from 5.0% in 2005 to 24.3% in 2016 was observed in the proportion of events with a reported mental illness with an increase between 2005 and 2016 in depression among female victims. These findings demonstrate that extracting information from police narratives can provide novel insights into domestic violence patterns including confounding factors (e.g., mental illness) and thus enable policy responses to address this significant public health problem.

Prevalence of Mental Illnesses in Domestic Violence Police Records: Text Mining Study

Background

The New South Wales Police Force (NSWPF) records details of significant numbers of domestic violence (DV) events they attend each year as both structured quantitative data and unstructured free text. Accessing information contained in the free text such as the victim’s and persons of interest (POI's) mental health status could be useful in the better management of DV events attended by the police and thus improve health, justice, and social outcomes.

Objective

The aim of this study is to present the prevalence of extracted mental illness mentions for POIs and victims in police-recorded DV events.

Methods

We applied a knowledge-driven text mining method to recognize mental illness mentions for victims and POIs from police-recorded DV events.

Results

In 416,441 police-recorded DV events with single POIs and single victims, we identified 64,587 events (15.51%) with at least one mental illness mention versus 4295 (1.03%) recorded in the structured fixed fields. Two-thirds (67,582/85,880, 78.69%) of mental illnesses were associated with POIs versus 21.30% (18,298/85,880) with victims; depression was the most common condition in both victims (2822/12,589, 22.42%) and POIs (7496/39,269, 19.01%). Mental illnesses were most common among POIs aged 0-14 years (623/1612, 38.65%) and in victims aged over 65 years (1227/22,873, 5.36%).

Conclusions

A wealth of mental illness information exists within police-recorded DV events that can be extracted using text mining. The results showed mood-related illnesses were the most common in both victims and POIs. Further investigation is required to determine the reliability of the mental illness mentions against sources of diagnostic information.

Pandemics, epidemics, viruses, plagues, and disease: Comparative frequency analysis of a cultural pathology reflected in science fiction magazines from 1926 to 2015

Science fiction includes many dystopian narratives, often featuring epidemics, pandemics, plagues, viruses, and disease. As science fiction has grown in popularity and prevalence it appeals to an increasingly broad demographic, is employed in research communication and education, and as a genre it is frequently argued that it reflects contemporary cultural interests and concerns. To identify the relevance of science fiction as an indicator of popular trends relating to the pathologies of disease, a word frequency comparison of selected key words found in the Google Books 2012 English Corpus has been made to a representative corpus of science fiction magazines dating between 1926 and 2015. Selected issues were reviewed to identify concepts, situations, and outcomes that could readily be measured against real-world examples from current and recent pandemics. The findings indicate that science fiction does appear to mirror and magnify contemporary literary trends, and provides potentially revealing correlations to real-world historical events. In this regard, science fiction might be regarded as a form of ‘cultural pathology’ of popular interests related to the spread and impact of disease that may be valuable in gauging the degree to which society is engaged with these topics at any specific time.

•

Science fiction topics tend to reflect real-world historical events.

•

Comparison of English corpus Google Books word frequencies to science fiction.

•

Science fiction investigates social, cultural and psychological concerns.

•

Science fiction content indicates a ‘cultural pathology’ of popular interests.

Effectiveness of Social Cognitive Theory-Based Interventions for Glycemic Control in Adults With Type 2 Diabetes Mellitus: Protocol for a Systematic Review and Meta-Analysis

BACKGROUND

For those living with type 2 diabetes mellitus (T2DM), failing to engage in self-management behaviors leads to poor glycemic control. Social cognitive theory (SCT) has been shown to improve health behaviors by altering cognitive processes and increasing an individual's belief in their ability to accomplish a task.

OBJECTIVE

We aim to present a protocol for a systematic review and meta-analysis to systematically identify, evaluate, and analyze the effect of SCT-based interventions to improve glycemic control in adults with T2DM.

METHODS

This protocol follows the 2009 Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. Data sources will include PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL), PsychINFO, Cochrane Library, and Web of Science, and data will be reviewed with the use of customized text mining software. Studies examining SCT-based behavioral interventions for adults diagnosed with T2DM in randomized controlled trials located in the outpatient setting will be included. Intervention effectiveness will be compared with routine care. Screening and data collection will be performed in multiple stages with three reviewers as follows: (1) an independent review of titles/abstracts, (2) a full review, and (3) data collection with alternating teams of two reviewers for disputes to be resolved by a third reviewer. Study quality and risk of bias will be assessed by three reviewers using the Cochrane risk of bias tool. Standardized mean differences will be used to describe the intervention effect sizes with regard to self-efficacy and diabetes knowledge. The raw mean difference of HbA1c will be provided in a random effects model and presented in a forest plot. The expected limitations of this study are incomplete data, the need to contact authors, and analysis of various types of glycemic control measures accurately within the same data set.

RESULTS

This protocol was granted institutional review board exemption on October 7, 2019. PROSPERO registration (ID: CRD42020147105) was received on April 28, 2020. The review began on April 29, 2020. The results of the review will be disseminated through conference presentations, peer-reviewed journals, and meetings.

CONCLUSIONS

This systematic review will appraise the effectiveness of SCT-based interventions for adults diagnosed with T2DM and provide the most effective interventions for improving health behaviors in these patients.

TRIAL REGISTRATION

PROSPERO CRD42020147105; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=147105.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

PRR1-10.2196/17148.

Qualitative Assessment of Adult Patients' Perception of Atopic Dermatitis Using Natural Language Processing Analysis in a Cross-Sectional Study

INTRODUCTION

Atopic dermatitis (AD) is an incurable, inflammatory skin disease characterized by skin barrier disruption and immune dysregulation. Although AD is considered a childhood disease, adult onset is possible, presenting with daily sleep disturbance and functional impairment associated with itch, neuropsychiatric issues (anxiety and depression), and reduced health-related quality of life. Although such aspects of adult AD disease burden have been measured through standardized assessments and based on population-level data, the understanding of the disease experienced at the patient level remains poor. This text-mining study assessed the impact of AD on the lives of adult patients as described from an experiential perspective.

METHODS

Natural language processing (NLP) was applied to qualitative patient response data from two large-scale international cross-sectional surveys conducted in the USA and countries outside of the USA (non-USA; Canada, France, Germany, Italy, Spain, and the UK). Descriptive analysis was conducted on patient responses to an open-ended question on how they felt about their AD and how the disease affected their life. Character length, word count, and stop word (common words) count were evaluated; centrality analysis identified concepts that were most strongly interlinked.

RESULTS

Patients with AD in all countries were most frequently impacted by itch, pain, and embarrassment across all levels of disease severity. Patients with moderate-to-severe AD were more likely than patients with mild AD to describe sleep disturbances, fatigue, and feelings of depression, anxiety, and a lack of hope that were directly associated with AD. Centrality analysis revealed sleep disturbance was strongly linked with itch. Collectively, these concepts revealed that patients with AD are impacted by both physical and emotional burdens that are intricately connected.

CONCLUSIONS

Qualitative data from NLP, being more patient-centric than data from clinical standardized measures, provide a more comprehensive view of the burden of AD to inform disease management.

DES-ROD: Exploring Literature to Develop New Links between RNA Oxidation and Human Diseases

Normal cellular physiology and biochemical processes require undamaged RNA molecules. However, RNAs are frequently subjected to oxidative damage. Overproduction of reactive oxygen species (ROS) leads to RNA oxidation and disturbs redox (oxidation-reduction reaction) homeostasis. When oxidation damage affects RNA carrying protein-coding information, this may result in the synthesis of aberrant proteins as well as a lower efficiency of translation. Both of these, as well as imbalanced redox homeostasis, may lead to numerous human diseases. The number of studies on the effects of RNA oxidative damage in mammals is increasing by year due to the understanding that this oxidation fundamentally leads to numerous human diseases. To enable researchers in this field to explore information relevant to RNA oxidation and effects on human diseases, we developed DES-ROD, an online knowledgebase that contains processed information from 298,603 relevant documents that consist of PubMed abstracts and PubMed Central full-text articles. The system utilizes concepts/terms from 38 curated thematic dictionaries mapped to the analyzed documents. Researchers can explore enriched concepts, as well as enriched pairs of putatively associated concepts. In this way, one can explore mutual relationships between any combinations of two concepts from used dictionaries. Dictionaries cover a wide range of biomedical topics, such as human genes and proteins, pathways, Gene Ontology categories, mutations, noncoding RNAs, enzymes, toxins, metabolites, and diseases. This makes insights into different facets of the effects of RNA oxidation and the control of this process possible. The usefulness of the DES-ROD system is demonstrated by case studies on some known information, as well as potentially novel information involving RNA oxidation and diseases. DES-ROD is the first knowledgebase based on text and data mining that focused on the exploration of RNA oxidation and human diseases.

Automated assessment of psychiatric disorders using speech: A systematic review

OBJECTIVE

There are many barriers to accessing mental health assessments including cost and stigma. Even when individuals receive professional care, assessments are intermittent and may be limited partly due to the episodic nature of psychiatric symptoms. Therefore, machine-learning technology using speech samples obtained in the clinic or remotely could one day be a biomarker to improve diagnosis and treatment. To date, reviews have only focused on using acoustic features from speech to detect depression and schizophrenia. Here, we present the first systematic review of studies using speech for automated assessments across a broader range of psychiatric disorders.

METHODS

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. We included studies from the last 10 years using speech to identify the presence or severity of disorders within the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). For each study, we describe sample size, clinical evaluation method, speech-eliciting tasks, machine learning methodology, performance, and other relevant findings.

RESULTS

1395 studies were screened of which 127 studies met the inclusion criteria. The majority of studies were on depression, schizophrenia, and bipolar disorder, and the remaining on post-traumatic stress disorder, anxiety disorders, and eating disorders. 63% of studies built machine learning predictive models, and the remaining 37% performed null-hypothesis testing only. We provide an online database with our search results and synthesize how acoustic features appear in each disorder.

CONCLUSION

Speech processing technology could aid mental health assessments, but there are many obstacles to overcome, especially the need for comprehensive transdiagnostic and longitudinal studies. Given the diverse types of data sets, feature extraction, computational methodologies, and evaluation criteria, we provide guidelines for both acquiring data and building machine learning models with a focus on testing hypotheses, open science, reproducibility, and generalizability.

LEVEL OF EVIDENCE

3a.

Stigma, biomarkers, and algorithmic bias: recommendations for precision behavioral health with artificial intelligence

Effective implementation of artificial intelligence in behavioral healthcare delivery depends on overcoming challenges that are pronounced in this domain. Self and social stigma contribute to under-reported symptoms, and under-coding worsens ascertainment. Health disparities contribute to algorithmic bias. Lack of reliable biological and clinical markers hinders model development, and model explainability challenges impede trust among users. In this perspective, we describe these challenges and discuss design and implementation recommendations to overcome them in intelligent systems for behavioral and mental health.

Using text mining to extract depressive symptoms and to validate the diagnosis of major depressive disorder from electronic health records

BACKGROUND

Many studies have used Taiwan's National Health Insurance Research database (NHIRD) to conduct psychiatric research. However, the accuracy of the diagnostic codes for psychiatric disorders in NHIRD is not validated, and the symptom profiles are not available either. This study aimed to evaluate the accuracy of diagnostic codes and use text mining to extract symptom profile and functional impairment from electronic health records (EHRs) to overcome the above research limitations.

METHODS

A total of 500 discharge notes were randomly selected from a medical center's database. Three annotators reviewed the notes to establish gold standards. The accuracy of diagnostic codes for major psychiatric illness was evaluated. Text mining approaches were applied to extract depressive symptoms and function profiles and to identify patients with major depressive disorder.

RESULTS

The accuracy of the diagnostic code for major depressive disorder, schizophrenia, and dementia was acceptable but that of bipolar disorder and minor depression was less satisfactory. The performance of text mining approach to recognize depressive symptoms is satisfactory; however, the recall for functional impairment is lower resulting in lower F-scores of 0.774-0.753. Using the text mining approach to identify major depressive disorder, the recall was 0.85 but precision was only 0.69.

CONCLUSIONS

The accuracy of the diagnostic code for major depressive disorder in discharge notes was generally acceptable. This finding supports the utilization of psychiatric diagnoses in claims databases. The application of text mining to EHRs might help in overcoming current limitations in research using claims databases.

Towards text mining therapeutic change: A systematic review of text-based methods for Therapeutic Change Process Research

Therapeutic Change Process Research (TCPR) connects within-therapeutic change processes to outcomes. The labour intensity of qualitative methods limit their use to small scale studies. Automated text-analyses (e.g. text mining) provide means for analysing large scale text patterns. We aimed to provide an overview of the frequently used qualitative text-based TCPR methods and assess the extent to which these methods are reliable and valid, and have potential for automation. We systematically reviewed PsycINFO, Scopus, and Web of Science to identify articles concerning change processes and text or language. We evaluated the reliability and validity based on replicability, the availability of code books, training data and inter-rater reliability, and evaluated the potential for automation based on the example- and rule-based approach. From 318 articles we identified four often used methods: Innovative Moments Coding Scheme, the Narrative Process Coding Scheme, Assimilation of Problematic Experiences Scale, and Conversation Analysis. The reliability and validity of the first three is sufficient to hold promise for automation. While some text features (content, grammar) lend themselves for automation through a rule-based approach, it should be possible to automate higher order constructs (e.g. schemas) when sufficient annotated data for an example-based approach are available.

KLOSURE: Closing in on open-ended patient questionnaires with text mining

Background

Knee injury and Osteoarthritis Outcome Score (KOOS) is an instrument used to quantify patients’ perceptions about their knee condition and associated problems. It is administered as a 42-item closed-ended questionnaire in which patients are asked to self-assess five outcomes: pain, other symptoms, activities of daily living, sport and recreation activities, and quality of life. We developed KLOG as a 10-item open-ended version of the KOOS questionnaire in an attempt to obtain deeper insight into patients’ opinions including their unmet needs. However, the open–ended nature of the questionnaire incurs analytical overhead associated with the interpretation of responses. The goal of this study was to automate such analysis. We implemented KLOSURE as a system for mining free–text responses to the KLOG questionnaire. It consists of two subsystems, one concerned with feature extraction and the other one concerned with classification of feature vectors. Feature extraction is performed by a set of four modules whose main functionalities are linguistic pre-processing, sentiment analysis, named entity recognition and lexicon lookup respectively. Outputs produced by each module are combined into feature vectors. The structure of feature vectors will vary across the KLOG questions. Finally, Weka, a machine learning workbench, was used for classification of feature vectors.

Results

The precision of the system varied between 62.8 and 95.3%, whereas the recall varied from 58.3 to 87.6% across the 10 questions. The overall performance in terms of F–measure varied between 59.0 and 91.3% with an average of 74.4% and a standard deviation of 8.8.

Conclusions

We demonstrated the feasibility of mining open-ended patient questionnaires. By automatically mapping free text answers onto a Likert scale, we can effectively measure the progress of rehabilitation over time. In comparison to traditional closed-ended questionnaires, our approach offers much richer information that can be utilised to support clinical decision making. In conclusion, we demonstrated how text mining can be used to combine the benefits of qualitative and quantitative analysis of patient experiences.

Literature-Based Enrichment Insights into Redox Control of Vascular Biology

In cellular physiology and signaling, reactive oxygen species (ROS) play one of the most critical roles. ROS overproduction leads to cellular oxidative stress. This may lead to an irrecoverable imbalance of redox (oxidation-reduction reaction) function that deregulates redox homeostasis, which itself could lead to several diseases including neurodegenerative disease, cardiovascular disease, and cancers. In this study, we focus on the redox effects related to vascular systems in mammals. To support research in this domain, we developed an online knowledge base, DES-RedoxVasc, which enables exploration of information contained in the biomedical scientific literature. The DES-RedoxVasc system analyzed 233399 documents consisting of PubMed abstracts and PubMed Central full-text articles related to different aspects of redox biology in vascular systems. It allows researchers to explore enriched concepts from 28 curated thematic dictionaries, as well as literature-derived potential associations of pairs of such enriched concepts, where associations themselves are statistically enriched. For example, the system allows exploration of associations of pathways, diseases, mutations, genes/proteins, miRNAs, long ncRNAs, toxins, drugs, biological processes, molecular functions, etc. that allow for insights about different aspects of redox effects and control of processes related to the vascular system. Moreover, we deliver case studies about some existing or possibly novel knowledge regarding redox of vascular biology demonstrating the usefulness of DES-RedoxVasc. DES-RedoxVasc is the first compiled knowledge base using text mining for the exploration of this topic.

Discovering major opioid-related research themes over time: A text mining technique

The recent health crisis concerning opioid overdose has prompted watershed levels of publications. This study explores how opioid-related research themes have changed since 2000, using a text mining technique. The textual data were obtained from PubMed, and the research periods were divided into three periods. While a few topics appear throughout each period, many new health problems emerged as opioid abuse problems magnified. Topics such as HIV, methadone maintenance treatment, and world health organization appear consistently but diminish over time, while topics like injecting drugs, neonatal abstinence syndrome, and public health concerns are rapidly increasing. Recent widespread opioid abuse problems led to new research topics, including prescription drug monitoring programs, veteran's health issue, posttraumatic stress disorder, HCV, opioid-related deaths, and emergency department visits. Examination of scholarly publications reveals that the expanded use of opioids worsened opioid abuse problems and accelerated the emergence of new health problems.

Natural Language Processing of Clinical Notes on Chronic Diseases: Systematic Review

Background

Novel approaches that complement and go beyond evidence-based medicine are required in the domain of chronic diseases, given the growing incidence of such conditions on the worldwide population. A promising avenue is the secondary use of electronic health records (EHRs), where patient data are analyzed to conduct clinical and translational research. Methods based on machine learning to process EHRs are resulting in improved understanding of patient clinical trajectories and chronic disease risk prediction, creating a unique opportunity to derive previously unknown clinical insights. However, a wealth of clinical histories remains locked behind clinical narratives in free-form text. Consequently, unlocking the full potential of EHR data is contingent on the development of natural language processing (NLP) methods to automatically transform clinical text into structured clinical data that can guide clinical decisions and potentially delay or prevent disease onset.

Objective

The goal of the research was to provide a comprehensive overview of the development and uptake of NLP methods applied to free-text clinical notes related to chronic diseases, including the investigation of challenges faced by NLP methodologies in understanding clinical narratives.

Methods

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed and searches were conducted in 5 databases using “clinical notes,” “natural language processing,” and “chronic disease” and their variations as keywords to maximize coverage of the articles.

Results

Of the 2652 articles considered, 106 met the inclusion criteria. Review of the included papers resulted in identification of 43 chronic diseases, which were then further classified into 10 disease categories using the International Classification of Diseases, 10th Revision. The majority of studies focused on diseases of the circulatory system (n=38) while endocrine and metabolic diseases were fewest (n=14). This was due to the structure of clinical records related to metabolic diseases, which typically contain much more structured data, compared with medical records for diseases of the circulatory system, which focus more on unstructured data and consequently have seen a stronger focus of NLP. The review has shown that there is a significant increase in the use of machine learning methods compared to rule-based approaches; however, deep learning methods remain emergent (n=3). Consequently, the majority of works focus on classification of disease phenotype with only a handful of papers addressing extraction of comorbidities from the free text or integration of clinical notes with structured data. There is a notable use of relatively simple methods, such as shallow classifiers (or combination with rule-based methods), due to the interpretability of predictions, which still represents a significant issue for more complex methods. Finally, scarcity of publicly available data may also have contributed to insufficient development of more advanced methods, such as extraction of word embeddings from clinical notes.

Conclusions

Efforts are still required to improve (1) progression of clinical NLP methods from extraction toward understanding; (2) recognition of relations among entities rather than entities in isolation; (3) temporal extraction to understand past, current, and future clinical events; (4) exploitation of alternative sources of clinical knowledge; and (5) availability of large-scale, de-identified clinical corpora.

Machine Learning, Natural Language Processing, and the Electronic Health Record: Innovations in Mental Health Services Research

An unprecedented amount of clinical information is now available via electronic health records (EHRs). These massive data sets have stimulated opportunities to adapt computational approaches to track and identify target areas for quality improvement in mental health care. In this column, three key areas of EHR data science are described: EHR phenotyping, natural language processing, and predictive modeling. For each of these computational approaches, case examples are provided to illustrate their role in mental health services research. Together, adaptation of these methods underscores the need for standardization and transparency while recognizing the opportunities and challenges ahead.

Automated Analysis of Domestic Violence Police Reports to Explore Abuse Types and Victim Injuries: Text Mining Study

Background

The police attend numerous domestic violence events each year, recording details of these events as both structured (coded) data and unstructured free-text narratives. Abuse types (including physical, psychological, emotional, and financial) conducted by persons of interest (POIs) along with any injuries sustained by victims are typically recorded in long descriptive narratives.

Objective

We aimed to determine if an automated text mining method could identify abuse types and any injuries sustained by domestic violence victims in narratives contained in a large police dataset from the New South Wales Police Force.

Methods

We used a training set of 200 recorded domestic violence events to design a knowledge-driven approach based on syntactical patterns in the text and then applied this approach to a large set of police reports.

Results

Testing our approach on an evaluation set of 100 domestic violence events provided precision values of 90.2% and 85.0% for abuse type and victim injuries, respectively. In a set of 492,393 domestic violence reports, we found 71.32% (351,178) of events with mentions of the abuse type(s) and more than one-third (177,117 events; 35.97%) contained victim injuries. “Emotional/verbal abuse” (33.46%; 117,488) was the most common abuse type, followed by “punching” (86,322 events; 24.58%) and “property damage” (22.27%; 78,203 events). “Bruising” was the most common form of injury sustained (51,455 events; 29.03%), with “cut/abrasion” (28.93%; 51,284 events) and “red marks/signs” (23.71%; 42,038 events) ranking second and third, respectively.

Conclusions

The results suggest that text mining can automatically extract information from police-recorded domestic violence events that can support further public health research into domestic violence, such as examining the relationship of abuse types with victim injuries and of gender and abuse types with risk escalation for victims of domestic violence. Potential also exists for this extracted information to be linked to information on the mental health status.

Automatic Extraction of Mental Health Disorders From Domestic Violence Police Narratives: Text Mining Study

BACKGROUND

Vast numbers of domestic violence (DV) incidents are attended by the New South Wales Police Force each year in New South Wales and recorded as both structured quantitative data and unstructured free text in the WebCOPS (Web-based interface for the Computerised Operational Policing System) database regarding the details of the incident, the victim, and person of interest (POI). Although the structured data are used for reporting purposes, the free text remains untapped for DV reporting and surveillance purposes.

OBJECTIVE

In this paper, we explore whether text mining can automatically identify mental health disorders from this unstructured text.

METHODS

We used a training set of 200 DV recorded events to design a knowledge-driven approach based on lexical patterns in text suggesting mental health disorders for POIs and victims.

RESULTS

The precision returned from an evaluation set of 100 DV events was 97.5% and 87.1% for mental health disorders related to POIs and victims, respectively. After applying our approach to a large-scale corpus of almost a half million DV events, we identified 77,995 events (15.83%) that mentioned mental health disorders, with 76.96% (60,032/77,995) of those linked to POIs versus 16.47% (12,852/77,995) for the victims and 6.55% (5111/77,995) for both. Depression was the most common mental health disorder mentioned in both victims (22.25%, 3269) and POIs (18.70%, 8944), followed by alcohol abuse for POIs (12.19%, 5829) and various anxiety disorders (eg, panic disorder, generalized anxiety disorder) for victims (11.66%, 1714).

CONCLUSIONS

The results suggest that text mining can automatically extract targeted information from police-recorded DV events to support further public health research into the nexus between mental health disorders and DV.