Kai Zheng

Kai Zheng

Auto-generated profile, based on publicly available information

Kai Zheng

Department of Informatics, University of California, Irvine, Irvine, CA, USA

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
24 March 2021 in Journal of Medical Internet Research
Verified
Journal Article
Open access
JBJudith Borghouts
Judith Borghouts
EEElizabeth Eikey
Elizabeth Eikey
GMGloria Mark
Gloria Mark
CLCinthia De Leon
Cinthia De Leon
SSStephen M Schueller
Stephen M Schueller
DSDara H Sorkin
Dara H Sorkin
Background: Digital mental health interventions (DMHIs), which deliver mental health support via technologies such as mobile apps, can increase access to mental health support, and many studies have demonstrated their effectiveness in improving symptoms. However, user engagement varies, with regard to a user’s uptake and sustained interactions with these interventions.Objective: This systematic review aims to identify common barriers and facilitators that influence user engagement with DMHIs.Methods: A systematic search was conducted in the SCOPUS, PubMed, PsycINFO, Web of Science, and Cochrane Library databases. Empirical studies that report qualitative and/or quantitative data were included.Results: A total of 208 articles met the inclusion criteria. The included articles used a variety of methodologies, including interviews, surveys, focus groups, workshops, field studies, and analysis of user reviews. Factors extracted for coding were related to the end user, the program or content offered by the intervention, and the technology and implementation environment. Common barriers included severe mental health issues that hampered engagement, technical issues, and a lack of personalization. Common facilitators were social connectedness facilitated by the intervention, increased insight into health, and a feeling of being in control of one’s own health.Conclusions: Although previous research suggests that DMHIs can be useful in supporting mental health, contextual factors are important determinants of whether users actually engage with these interventions. The factors identified in this review can provide guidance when evaluating DMHIs to help explain and understand user engagement and can inform the design and development of new digital interventions.

16 April 2021 in Journal of Medical Internet Research
Verified
Journal Article
Open access
DSDara H Sorkin
Dara H Sorkin
EJEmily A Janio
Emily A Janio
EEElizabeth V Eikey
Elizabeth V Eikey
MSMargaret Schneider
Margaret Schneider
KDKatelyn Davis
Katelyn Davis
DMDana B Mukamel
Dana B Mukamel
Background: Accompanying the rising rates of reported mental distress during the COVID-19 pandemic has been a reported increase in the use of digital technologies to manage health generally, and mental health more specifically.Objective: The objective of this study was to systematically examine whether there was a COVID-19 pandemic–related increase in the self-reported use of digital mental health tools and other technologies to manage mental health.Methods: We analyzed results from a survey of 5907 individuals in the United States using Amazon Mechanical Turk (MTurk); the survey was administered during 4 week-long periods in 2020 and survey respondents were from all 50 states and Washington DC. The first set of analyses employed two different logistic regression models to estimate the likelihood of having symptoms indicative of clinical depression and anxiety, respectively, as a function of the rate of COVID-19 cases per 10 people and survey time point. The second set employed seven different logistic regression models to estimate the likelihood of using seven different types of digital mental health tools and other technologies to manage one’s mental health, as a function of symptoms indicative of clinical depression and anxiety, rate of COVID-19 cases per 10 people, and survey time point. These models also examined potential interactions between symptoms of clinical depression and anxiety, respectively, and rate of COVID-19 cases. All models controlled for respondent sociodemographic characteristics and state fixed effects.Results: Higher COVID-19 case rates were associated with a significantly greater likelihood of reporting symptoms of depression (odds ratio [OR] 2.06, 95% CI 1.27-3.35), but not anxiety (OR 1.21, 95% CI 0.77-1.88). Survey time point, a proxy for time, was associated with a greater likelihood of reporting clinically meaningful symptoms of depression and anxiety (OR 1.19, 95% CI 1.12-1.27 and OR 1.12, 95% CI 1.05-1.19, respectively). Reported symptoms of depression and anxiety were associated with a greater likelihood of using each type of technology. Higher COVID-19 case rates were associated with a significantly greater likelihood of using mental health forums, websites, or apps (OR 2.70, 95% CI 1.49-4.88), and other health forums, websites, or apps (OR 2.60, 95% CI 1.55-4.34). Time was associated with increased odds of reported use of mental health forums, websites, or apps (OR 1.20, 95% CI 1.11-1.30), phone-based or text-based crisis lines (OR 1.20, 95% CI 1.10-1.31), and online, computer, or console gaming/video gaming (OR 1.12, 95% CI 1.05-1.19). Interactions between COVID-19 case rate and mental health symptoms were not significantly associated with any of the technology types.Conclusions: Findings suggested increased use of digital mental health tools and other technologies over time during the early stages of the COVID-19 pandemic. As such, additional effort is urgently needed to consider the quality of these products, either by ensuring users have access to evidence-based and evidence-informed technologies and/or by providing them with the skills to make informed decisions around their potential efficacy.

5 May 2021 in JMIR mHealth and uHealth
Verified
Journal Article
Open access
SASeyed Amir Hossein Aqajari
Seyed Amir Hossein Aqajari
RCRui Cao
Rui Cao
ENEmad Kasaeyan Naeini
Emad Kasaeyan Naeini
MCMichael-David Calderon
Michael-David Calderon
KZKai Zheng
Kai Zheng
ARAmir Rahmani
Amir Rahmani
Background: Accurate, objective pain assessment is required in the health care domain and clinical settings for appropriate pain management. Automated, objective pain detection from physiological data in patients provides valuable information to hospital staff and caregivers to better manage pain, particularly for patients who are unable to self-report. Galvanic skin response (GSR) is one of the physiologic signals that refers to the changes in sweat gland activity, which can identify features of emotional states and anxiety induced by varying pain levels. This study used different statistical features extracted from GSR data collected from postoperative patients to detect their pain intensity. To the best of our knowledge, this is the first work building pain models using postoperative adult patients instead of healthy subjects. Objective: The goal of this study was to present an automatic pain assessment tool using GSR signals to predict different pain intensities in noncommunicative, postoperative patients. Methods: The study was designed to collect biomedical data from postoperative patients reporting moderate to high pain levels. We recruited 25 participants aged 23-89 years. First, a transcutaneous electrical nerve stimulation (TENS) unit was employed to obtain patients' baseline data. In the second part, the Empatica E4 wristband was worn by patients while they were performing low-intensity activities. Patient self-report based on the numeric rating scale (NRS) was used to record pain intensities that were correlated with objectively measured data. The labels were down-sampled from 11 pain levels to 5 different pain intensities, including the baseline. We used 2 different machine learning algorithms to construct the models. The mean decrease impurity method was used to find the top important features for pain prediction and improve the accuracy. We compared our results with a previously published research study to estimate the true performance of our models. Results: Four different binary classification models were constructed using each machine learning algorithm to classify the baseline and other pain intensities (Baseline [BL] vs Pain Level [PL] 1, BL vs PL2, BL vs PL3, and BL vs PL4). Our models achieved higher accuracy for the first 3 pain models than the BioVid paper approach despite the challenges in analyzing real patient data. For BL vs PL1, BL vs PL2, and BL vs PL4, the highest prediction accuracies were achieved when using a random forest classifier (86.0, 70.0, and 61.5, respectively). For BL vs PL3, we achieved an accuracy of 72.1 using a k-nearest-neighbor classifier. Conclusions: We are the first to propose and validate a pain assessment tool to predict different pain levels in real postoperative adult patients using GSR signals. We also exploited feature selection algorithms to find the top important features related to different pain intensities. International Registered Report Identifier (IRRID): RR2-10.2196/17783

14 September 2021 in Journal of Medical Internet Research
Verified
Journal Article
Open access
JBJudith Borghouts
Judith Borghouts
EEElizabeth V Eikey
Elizabeth V Eikey
GMGloria Mark
Gloria Mark
CLCinthia De Leon
Cinthia De Leon
SSStephen M Schueller
Stephen M Schueller
DSDara H Sorkin
Dara H Sorkin
Background: Mental health concerns are a significant issue among community college students, who often have less access to resources than traditional university college students. Mobile apps have the potential to increase access to mental health care, but there has been little research investigating factors associated with mental health app use within the community college population. Objective: This study aimed to understand facilitators of and barriers to mental health app use among community college students. Methods: A web-based survey was administered to a randomly selected sample of 500 community college students from April 16 to June 30, 2020. Structural equation modeling was used to test the relationships between the use of mental health apps, perceived stress, perceived need to seek help for mental health concerns, perceived stigma, past use of professional mental health services, privacy concerns, and social influence of other people in using mental health apps. Results: Of the 500 participants, 106 (21.2%) reported use of mental health apps. Perceived stress, perceived need to seek help, past use of professional services, and social influence were positively associated with mental health app use. Furthermore, the effect of stress was mediated by a perceived need to seek help. Privacy concerns were negatively associated with mental health app use. Stigma, age, and gender did not have a statistically significant effect. Conclusions: These findings can inform development of new digital interventions and appropriate outreach strategies to engage community college students in using mental health apps.

1 August 2016 in Yearbook of Medical Informatics
Verified
Journal Article
Closed access
JAJ. Abraham
J. Abraham
LNL. L. Novak
L. L. Novak
TRT. L. Reynolds
T. L. Reynolds
AGA. Gettinger
A. Gettinger
KZK. Zheng
K. Zheng
Summary: Objective: To summarize recent research on unintended consequences associated with implementation and use of health information technology (health IT). Included in the review are original empirical investigations published in English between 2014 and 2015 that reported unintended effects introduced by adoption of digital interventions. Our analysis focuses on the trends of this steam of research, areas in which unintended consequences have continued to be reported, and common themes that emerge from the findings of these studies.Method: Most of the papers reviewed were retrieved by searching three literature databases: MEDLINE, Embase, and CINAHL. Two rounds of searches were performed: the first round used more restrictive search terms specific to unintended consequences; the second round lifted the restrictions to include more generic health IT evaluation studies. Each paper was independently screened by at least two authors; differences were resolved through consensus development.Results: The literature search identified 1,538 papers that were potentially relevant; 34 were deemed meeting our inclusion criteria after screening. Studies described in these 34 papers took place in a wide variety of care areas from emergency departments to ophthalmology clinics. Some papers reflected several previously unreported unintended consequences, such as staff attrition and patients’ withholding of information due to privacy and security concerns. A majority of these studies (71%) were quantitative investigations based on analysis of objectively recorded data. Several of them employed longitudinal or time series designs to distinguish between unintended consequences that had only transient impact, versus those that had persisting impact. Most of these unintended consequences resulted in adverse outcomes, even though instances of beneficial impact were also noted. While care areas covered were heterogeneous, over half of the studies were conducted at academic medical centers or teaching hospitals. Conclusion: Recent studies published in the past two years represent significant advancement of unintended consequences research by seeking to include more types of health IT applications and to quantify the impact using objectively recorded data and longitudinal or time series designs. However, more mixed-methods studies are needed to develop deeper insights into the observed unintended adverse outcomes, including their root causes and remedies. We also encourage future research to go beyond the paradigm of simply describing unintended consequences, and to develop and test solutions that can prevent or minimize their impact.

28 May 2021 in Journal of Medical Internet Research
Verified
Journal Article
Open access
ENEmad Kasaeyan Naeini
Emad Kasaeyan Naeini
ASAjan Subramanian
Ajan Subramanian
MCMichael-David Calderon
Michael-David Calderon
KZKai Zheng
Kai Zheng
NDNikil Dutt
Nikil Dutt
ARAmir Rahmani
Amir Rahmani
Background: There is a strong demand for an accurate and objective means of assessing acute pain among hospitalized patients to help clinicians provide pain medications at a proper dosage and in a timely manner. Heart rate variability (HRV) comprises changes in the time intervals between consecutive heartbeats, which can be measured through acquisition and interpretation of electrocardiography (ECG) captured from bedside monitors or wearable devices. As increased sympathetic activity affects the HRV, an index of autonomic regulation of heart rate, ultra–short-term HRV analysis can provide a reliable source of information for acute pain monitoring. In this study, widely used HRV time and frequency domain measurements are used in acute pain assessments among postoperative patients. The existing approaches have only focused on stimulated pain in healthy subjects, whereas, to the best of our knowledge, there is no work in the literature building models using real pain data and on postoperative patients. Objective: The objective of our study was to develop and evaluate an automatic and adaptable pain assessment algorithm based on ECG features for assessing acute pain in postoperative patients likely experiencing mild to moderate pain. Methods: The study used a prospective observational design. The sample consisted of 25 patient participants aged 18 to 65 years. In part 1 of the study, a transcutaneous electrical nerve stimulation unit was employed to obtain baseline discomfort thresholds for the patients. In part 2, a multichannel biosignal acquisition device was used as patients were engaging in non-noxious activities. At all times, pain intensity was measured using patient self-reports based on the Numerical Rating Scale. A weak supervision framework was inherited for rapid training data creation. The collected labels were then transformed from 11 intensity levels to 5 intensity levels. Prediction models were developed using 5 different machine learning methods. Mean prediction accuracy was calculated using leave-one-out cross-validation. We compared the performance of these models with the results from a previously published research study. Results: Five different machine learning algorithms were applied to perform a binary classification of baseline (BL) versus 4 distinct pain levels (PL1 through PL4). The highest validation accuracy using 3 time domain HRV features from a BioVid research paper for baseline versus any other pain level was achieved by support vector machine (SVM) with 62.72% (BL vs PL4) to 84.14% (BL vs PL2). Similar results were achieved for the top 8 features based on the Gini index using the SVM method, with an accuracy ranging from 63.86% (BL vs PL4) to 84.79% (BL vs PL2). Conclusions: We propose a novel pain assessment method for postoperative patients using ECG signal. Weak supervision applied for labeling and feature extraction improves the robustness of the approach. Our results show the viability of using a machine learning algorithm to accurately and objectively assess acute pain among hospitalized patients. International Registered Report Identifier (IRRID): RR2-10.2196/17783