scholarly journals Artificial intelligence applications in social media for depression screening: A systematic review protocol for content validity processes

PLoS ONE ◽  
2021 ◽  
Vol 16 (11) ◽  
pp. e0259499
Author(s):  
Priscilla N. Owusu ◽  
Ulrich Reininghaus ◽  
Georgia Koppe ◽  
Irene Dankwa-Mullan ◽  
Till Bärnighausen
Keyword(s):  
Artificial Intelligence ◽  
Systematic Review ◽  
Social Media ◽  
Content Validity ◽  
Machine Learning Techniques ◽  
Measurement Instruments ◽  
Normative Judgment ◽  
Current Application ◽  
Randomized Controlled Studies ◽  
Secondary Outcomes

Background The popularization of social media has led to the coalescing of user groups around mental health conditions; in particular, depression. Social media offers a rich environment for contextualizing and predicting users’ self-reported burden of depression. Modern artificial intelligence (AI) methods are commonly employed in analyzing user-generated sentiment on social media. In the forthcoming systematic review, we will examine the content validity of these computer-based health surveillance models with respect to standard diagnostic frameworks. Drawing from a clinical perspective, we will attempt to establish a normative judgment about the strengths of these modern AI applications in the detection of depression. Methods We will perform a systematic review of English and German language publications from 2010 to 2020 in PubMed, APA PsychInfo, Science Direct, EMBASE Psych, Google Scholar, and Web of Science. The inclusion criteria span cohort, case-control, cross-sectional studies, randomized controlled studies, in addition to reports on conference proceedings. The systematic review will exclude some gray source materials, specifically editorials, newspaper articles, and blog posts. Our primary outcome is self-reported depression, as expressed on social media. Secondary outcomes will be the types of AI methods used for social media depression screen, and the clinical validation procedures accompanying these methods. In a second step, we will utilize the evidence-strengthening Population, Intervention, Comparison, Outcomes, Study type (PICOS) tool to refine our inclusion and exclusion criteria. Following the independent assessment of the evidence sources by two authors for the risk of bias, the data extraction process will culminate in a thematic synthesis of reviewed studies. Discussion We present the protocol for a systematic review which will consider all existing literature from peer reviewed publication sources relevant to the primary and secondary outcomes. The completed review will discuss depression as a self-reported health outcome in social media material. We will examine the computational methods, including AI and machine learning techniques which are commonly used for online depression surveillance. Furthermore, we will focus on standard clinical assessments, as indicating content validity, in the design of the algorithms. The methodological quality of the clinical construct of the algorithms will be evaluated with the COnsensus-based Standards for the selection of health status Measurement Instruments (COSMIN) framework. We conclude the study with a normative judgment about the current application of AI to screen for depression on social media. Systematic review registration International Prospective Register of Systematic Reviews PROSPERO (registration number CRD42020187874).

scholarly journals A Systematic Review Evaluating Psychometric Properties of Parent or Caregiver Report Instruments on Child Maltreatment: Part 2: Internal Consistency, Reliability, Measurement Error, Structural Validity, Hypothesis Testing, Cross-Cultural Validity, and Criterion Validity

2020 ◽  
pp. 152483802091559
Author(s):  
Sangwon Yoon ◽  
Renée Speyer ◽  
Reinie Cordier ◽  
Pirjo Aunio ◽  
Airi Hakkarainen
Keyword(s):  
Systematic Review ◽  
Child Maltreatment ◽  
Psychometric Properties ◽  
Content Validity ◽  
Public Health Issue ◽  
Global Public Health ◽  
Measurement Instruments ◽  
Health Measurement ◽  
Patient Report Outcome ◽  
Psychometric Data

Aims: Child maltreatment (CM) is global public health issue with devastating lifelong consequences. Global organizations have endeavored to eliminate CM; however, there is lack of consensus on what instruments are most suitable for the investigation and prevention of CM. This systematic review aimed to appraise the psychometric properties (other than content validity) of all current parent- or caregiver-reported CM instruments and recommend the most suitable for use. Method: A systematic search of the CINAHL, Embase, ERIC, PsycINFO, PubMed, and Sociological Abstracts databases was performed. The evaluation of psychometric properties was conducted according to the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) guidelines for systematic reviews of patient-report outcome measures. Responsiveness was beyond the scope of this systematic review, and content validity has been reported on in a companion paper (Part 1). Only instruments developed and published in English were included. Results: Twenty-five studies reported on selected psychometric properties of 15 identified instruments. The methodological quality of the studies was overall adequate. The psychometric properties of the instruments were generally indeterminate or not reported due to incomplete or missing psychometric data; high-quality evidence on the psychometric properties was limited. Conclusions: No instruments could be recommended as most suitable for use in clinic and research. Nine instruments were identified as promising based on current psychometric data but would need further psychometric evidence for them to be recommended.

scholarly journals A Systematic Review Evaluating Psychometric Properties of Parent or Caregiver Report Instruments on Child Maltreatment: Part 1: Content Validity

2020 ◽  
pp. 152483801989845
Author(s):  
Sangwon Yoon ◽  
Renée Speyer ◽  
Reinie Cordier ◽  
Pirjo Aunio ◽  
Airi Hakkarainen
Keyword(s):  
Systematic Review ◽  
Child Abuse ◽  
Child Maltreatment ◽  
Psychometric Properties ◽  
Content Validity ◽  
Public Health Issue ◽  
Measurement Instruments ◽  
Health Measurement ◽  
Caregiver Report

Aims: Child maltreatment (CM) is a serious public health issue, affecting over half of all children globally. Although most CM is perpetrated by parents or caregivers and their reports of CM is more accurate than professionals or children, parent or caregiver report instruments measuring CM have never been systematically evaluated for their content validity, the most important psychometric property. This systematic review aimed to evaluate the content validity of all current parent or caregiver report CM instruments. Methods: A systematic literature search was performed in CINAHL, Embase, ERIC, PsycINFO, PubMed, and Sociological Abstracts; gray literature was retrieved through reference checking. Eligible studies needed to report on content validity of instruments measuring CM perpetrated and reported by parents or caregivers. The quality of studies and content validity of the instruments were evaluated using the COnsensus-based Standards for the selection of health Measurement INstruments guidelines. Results: Fifteen studies reported on the content validity of 15 identified instruments. The study quality was generally poor. The content validity of the instruments was overall sufficient, but most instruments did not provide high-quality evidence for content validity. Conclusions: Most instruments included in this review showed promising content validity. The International Society for the Prevention of Child Abuse and Neglect Child Abuse Screening Tool for use in Trial appears to be the most promising, followed by the Family Maltreatment–Child Abuse criteria. However, firm conclusions cannot be drawn due to the low quality of evidence for content validity. Further studies are required to evaluate the remaining psychometric properties for recommending parent or caregiver report CM instruments.

Application of Machine Learning Techniques in Healthcare

2020 ◽  
pp. 289-304 ◽  
Author(s):  
Sathya D. ◽  
Sudha V. ◽  
Jagadeesan D.
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Social Media ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Or Efficiency ◽  
Game Development ◽  
Learning Techniques ◽  
Media Services

Machine learning is an approach of artificial intelligence (AI) where the machine can automatically learn and improve its performance on experience. It is not explicitly programmed; the data is fed into the generic algorithm and it builds logic based on the data provided. Traditional algorithms have to define new rules or massive rules when the pattern varies or the number of patterns increases, which reduces the accuracy or efficiency of the algorithms. But the machine learning algorithms learn new input patterns capable of handling complex situations while maintaining accuracy and efficiency. Due to its effectual benefits, machine learning algorithms are used in various domains like healthcare, industries, travel, game development, social media services, robotics, and surveillance and information security. In this chapter, the application of machine learning technique in healthcare is discussed in detail.

scholarly journals Treatment Outcome Measurement Instruments for Port Wine Stains: A Systematic Review of Their Measurement Properties

Dermatology ◽  
2020 ◽  
pp. 1-17
Author(s):  
M. Ingmar van Raath ◽  
Sandeep Chohan ◽  
Albert Wolkerstorfer ◽  
Chantal M.A.M. van der Horst ◽  
Jacqueline Limpens ◽  
...  
Keyword(s):  
Systematic Review ◽  
Content Validity ◽  
Outcome Measurement ◽  
Measurement Properties ◽  
Minimal Important Change ◽  
Port Wine ◽  
Measurement Instruments ◽  
Health Measurement ◽  
Patient Reported

<b><i>Background:</i></b> A plethora of outcome measurement instruments (OMIs) are being used in port wine stain (PWS) studies. It is currently unclear how valid, responsive, and reliable these are. <b><i>Objectives:</i></b> The aim of this systematic review was to appraise the content validity and other measurement properties of OMIs for PWS treatment to identify the most appropriate instruments and future research priorities. <b><i>Methods:</i></b> This study was performed using the updated Consensus-Based Standards for the Selection of Health Measurement Instruments (COSMIN) methodology and adhered to PRISMA guidelines. Comprehensive searches in Medline and Embase were performed. Studies in which an OMI for PWS patients was developed or its measurement properties were evaluated were included. Two investigators independently extracted data and assessed the quality of included studies and instruments to perform qualitative synthesis of the evidence. <b><i>Results:</i></b> In total, 1,034 articles were screened, and 77 full-text articles were reviewed. A total of 8 studies were included that reported on 6 physician-reported OMIs of clinical improvement and 6 parent- or patient-reported OMIs of life impact, of which 3 for health-related quality of life and 1 for perceived stigmatization. Overall, the quality of OMI development was inadequate (63%) or doubtful (37%). Each instrument has undergone a very limited evaluation in PWS patients. No content validity studies were performed. The quality of evidence for content validity was very low (78%), low (15%), or moderate (7%), with sufficient comprehensibility, mostly sufficient comprehensiveness, and mixed relevance. No studies on responsiveness, minimal important change, and cross-cultural validity were retrieved. There was moderate- to very low-quality evidence for sufficient inter-rater reliability for some clinical PWS OMIs. Internal consistency and measurement error were indeterminate in all studies. <b><i>Conclusions:</i></b> There was insufficient evidence to properly guide outcome selection. Additional assessment of the measurement properties of OMIs is needed, preferentially guided by a core domain set tailored to PWS.

Sign in / Sign up

Export Citation Format

Share Document