Jonathan Jacob Marsh

Dr. Jonathan Jacob Marsh

orcid

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 May 2021 in Journal of Medical Internet Research
Verified
Journal Article
Open access
AGAziliz Le Glaz
Aziliz Le Glaz
YHYannis Haralambous
Yannis Haralambous
DKDeok-Hee Kim-Dufor
Deok-Hee Kim-Dufor
PLPhilippe Lenca
Philippe Lenca
RBRomain Billot
Romain Billot
CLChristophe LEMEY
Christophe LEMEY
Background: Machine learning systems are part of the field of artificial intelligence that automatically learn models from data to make better decisions. Natural language processing (NLP), by using corpora and learning approaches, provides good performance in statistical tasks, such as text classification or sentiment mining. Objective: The primary aim of this systematic review was to summarize and characterize, in methodological and technical terms, studies that used machine learning and NLP techniques for mental health. The secondary aim was to consider the potential use of these methods in mental health clinical practice Methods: This systematic review follows the PRISMA (Preferred Reporting Items for Systematic Review and Meta-analysis) guidelines and is registered with PROSPERO (Prospective Register of Systematic Reviews; number CRD42019107376). The search was conducted using 4 medical databases (PubMed, Scopus, ScienceDirect, and PsycINFO) with the following keywords: machine learning, data mining, psychiatry, mental health, and mental disorder. The exclusion criteria were as follows: languages other than English, anonymization process, case studies, conference papers, and reviews. No limitations on publication dates were imposed. Results: A total of 327 articles were identified, of which 269 (82.3%) were excluded and 58 (17.7%) were included in the review. The results were organized through a qualitative perspective. Although studies had heterogeneous topics and methods, some themes emerged. Population studies could be grouped into 3 categories: patients included in medical databases, patients who came to the emergency room, and social media users. The main objectives were to extract symptoms, classify severity of illness, compare therapy effectiveness, provide psychopathological clues, and challenge the current nosography. Medical records and social media were the 2 major data sources. With regard to the methods used, preprocessing used the standard methods of NLP and unique identifier extraction dedicated to medical texts. Efficient classifiers were preferred rather than transparent functioning classifiers. Python was the most frequently used platform. Conclusions: Machine learning and NLP models have been highly topical issues in medicine in recent years and may be considered a new paradigm in medical research. However, these processes tend to confirm clinical hypotheses rather than developing entirely new information, and only one major category of the population (ie, social media users) is an imprecise cohort. Moreover, some language-specific features can improve the performance of NLP methods, and their extension to other languages should be more closely investigated. However, machine learning and NLP techniques provide useful information from unexplored data (ie, patients’ daily habits that are usually inaccessible to care providers). Before considering It as an additional tool of mental health care, ethical issues remain and should be discussed in a timely manner. Machine learning and NLP methods may offer multiple perspectives in mental health research but should also be considered as tools to support clinical practice.

18 November 2022 in Schizophrenia Bulletin
Verified
Journal Article
Closed access
PDF file
JDJordan DeVylder
Jordan DeVylder
DADeidre Anglin
Deidre Anglin
MMMichelle R Munson
Michelle R Munson
ANAtsushi Nishida
Atsushi Nishida
HOHans Oh
Hans Oh
LFLisa Fedina
Lisa Fedina
Background & Hypothesis: Psychotic disorders are inequitably distributed by race in the United States, although it is not known whether this is due to assessment biases or inequitable distributions of risk factors. Psychotic experiences are subclinical hallucinations and delusions used to study the etiology of psychosis, which are based on self-report and therefore not subject to potential clinician biases. In this study, we test whether the prevalence of psychotic experiences (PE) varies by race and if this variance is explained by socioenvironmental risk factors.Study Design: Data on demographics, PE, and socioenvironmental risk factors were collected through the National Survey of Poly-victimization and Mental Health, a national probability sample of US young adults. Logistic regression analyses were used to determine whether PE prevalence varied by race/ethnicity and, if so, whether this was attenuated with inclusion of indicators of income, education, urban/rural living, discrimination, and trauma exposure.Study Results: Black and Hispanic respondents reported PE at significantly greater rates than White or “other” ethnoracial groups, with hallucinations more commonly reported by Hispanic respondents. PE were significantly associated with police violence exposure, discrimination, adverse childhood experiences, and educational attainment. These factors statistically explained ethnoracial differences in the likelihood of overall PE occurrence and of nearly all PE subtypes.Conclusions: Previously observed racial differences in psychosis extend beyond clinical schizophrenia, and therefore, are unlikely to be explained entirely by clinician biases. Instead, racial disparities in PE appear to be driven by features of structural racism, trauma, and discrimination.

16 March 2013 in Journal of Police and Criminal Psychology
Verified
Journal Article
Closed access
PDF file
MSMarina Sorochinski
Marina Sorochinski
MHMaria Hartwig
Maria Hartwig
JOJeffery Osborne
Jeffery Osborne
EWEugenia Wilkins
Eugenia Wilkins
JMJonathan Jacob Marsh
Jonathan Jacob Marsh
PGPär Anders Granhag
Pär Anders Granhag

28 May 2021 in Biological Psychiatry Global Open Science
Verified
Journal Article
Open access
HOHans Oh
Hans Oh
JSJason Schiffman
Jason Schiffman
JMJonathan Jacob Marsh
Jonathan Jacob Marsh
SZSasha Zhou
Sasha Zhou
AKAi Koyanagi
Ai Koyanagi
JDJordan DeVylder
Jordan DeVylder
Clinical reports from across the world have documented psychosis in the context of COVID-19 infection; however, there has yet to be a large-scale epidemiological study to confirm this association. We analyzed data from the Healthy Minds Study (N = 15,935; conducted between September and December 2020), which was administered online to students attending one of 28 colleges in the United States. Using multivariable logistic regression, we examined the associations between COVID-19 infection/severity and psychotic experiences over the past 12 months, adjusting for age, gender, race/ethnicity, and international student status as well as anxiety and depression. More than one fifth of the analytic sample reported COVID-19 infection, and about one in six students with COVID-19 infection reported psychotic experiences over the past 12 months. In weighted multivariable logistic regression models, COVID-19 infection was associated with significantly greater odds of having psychotic experiences (adjusted odds ratio 1.36, 95% CI 1.19–1.48). Compared with being asymptomatic, having moderate (adjusted odds ratio 1.85, 95% CI 1.03–3.31) or severe (adjusted odds ratio 1.76, 95% CI 1.11–2.77) symptoms was associated with significantly greater odds of having psychotic experiences. These associations became statistically nonsignificant when adjusting for depression and anxiety. Hospitalization was not significantly associated with psychotic experiences among students with COVID-19 infection. Psychotic experiences are associated with COVID-19 infections, though much of the association is attenuated when accounting for anxiety and depression. Findings based on this sample of college students should be replicated outside of the college context to determine whether psychosis is a neuropsychiatric symptom during and after COVID-19 infection.

12 June 2022 in Journal of American College Health
Verified
Journal Article
Closed access
HOHans Oh
Hans Oh
CMCaitlin Marinovich
Caitlin Marinovich
SJSamantha Jay
Samantha Jay
JMJonathan Jacob Marsh
Jonathan Jacob Marsh
SZSasha Zhou
Sasha Zhou
JDJordan E. DeVylder
Jordan E. DeVylder
The COVID-19 pandemic has dramatically altered the way of life in the United States, which may be linked to self-injurious behaviors.We conducted a secondary analysis of data from the Fall 2020 Coh...

5 April 2021 in Psychosis
Verified
Journal Article
Closed access
JMJonathan Jacob Marsh
Jonathan Jacob Marsh
ZNZui Narita
Zui Narita
FZFuhua Zhai
Fuhua Zhai
LFLisa Fedina
Lisa Fedina
JSJason Schiffman
Jason Schiffman
JDJordan DeVylder
Jordan DeVylder
Background: Psychotic experiences (PEs) occur in the general population, serving as indicators of vulnerability to psychotic disorders and other mental and physical concerns. Environmental stressors, including violence exposure and social disconnection are associated with PEs, although the exact nature of these relations is unknown.Methods: Data were drawn from an online community sample of 1,000 adults (476 male) residing in Baltimore or New York City. Binary logistic regression and mediation analyses were used to examine associations between PEs, violence exposure (adverse childhood experiences; intimate partner violence), and social disconnectedness (personal; neighborhood).Results: Personal social disconnectedness was significantly associated with delusional mood (p < .001). Neighborhood social disconnectedness was significantly associated with delusional mood (p < .01), delusions of reference and persecution (p < .001), delusions of control (p < .01), and hallucinations (p < .01). Both subtypes of violence exposure were found to have significant direct effects on PEs (p < .001). Personal social disconnectedness accounted for 6.72-7.87% of the total effect of violence exposure on PEs, whereas neighborhood social disconnectedness was found to account for 5.14-6.74% of the total effect.Conclusion: The results advance understandings of prominent environmental risks for psychosis, and offer new implications for public health consideration of psychosis.