scholarly journals A Convenient and Low-Cost Model of Depression Screening and Early Warning Based on Voice Data Using for Public Mental Health

2021 ◽  
Vol 18 (12) ◽  
pp. 6441
Author(s):  
Xin Chen ◽  
Zhigeng Pan
Keyword(s):  
Mental Health ◽  
Early Warning ◽  
Cost Model ◽  
Low Cost ◽  
Ethical Review ◽  
Algorithm Selection ◽  
Data Set ◽  
Public Data ◽  
Voice Data ◽  
The Voice

Depression is a common mental health disease, which has great harm to public health. At present, the diagnosis of depression mainly depends on the interviews between doctors and patients, which is subjective, slow and expensive. Voice data are a kind of data that are easy to obtain and have the advantage of low cost. It has been proved that it can be used in the diagnosis of depression. The voice data used for modeling in this study adopted the authoritative public data set, which had passed the ethical review. The features of voice data were extracted by Python programming, and the voice features were stored in the format of CSV files. Through data processing, a big database, containing 1479 voice feature samples, was generated for modeling. Then, the decision tree screening model of depression was established by 10-fold cross validation and algorithm selection. The experiment achieved 83.4% prediction accuracy on voice data set. According to the prediction results of the model, the patients can be given early warning and intervention in time, so as to realize the health management of personal depression.

Characteristic Latent Features for Analyzing Digital Mental Health Interaction and Improved Explainability (Preprint)

2021 ◽  
Author(s):  
Max-Marcel Theilig ◽  
Ashley A Knapp ◽  
Jennifer M Nicholas ◽  
Rüdiger Zarnekow ◽  
David C Mohr
Keyword(s):  
Mental Health ◽  
Data Science ◽  
Digital Health ◽  
A Priori ◽  
User Interaction ◽  
User Engagement ◽  
Behavioral Strategies ◽  
Data Set ◽  
Public Data ◽  
Latent Features

BACKGROUND Using mobile health technology has sparked a broad engagement of data science and machine learning methods to leverage the complex, assorted amount of data for mental health purposes. Despite many studies, there is a reported underdevelopment of user engagement concepts, and the desire for high accuracy or significance has shown to lead to low explicability and irreproducibility. OBJECTIVE To overcome such reasons of poor analysis input and facilitate the reproducibility and credibility of artificial intelligence applications, we aim to explore principal characteristics of user interaction with digital mental health. METHODS We generated five latent features based on previous research, expert opinions from digital mental health, and informed by data. The features were analyzed with descriptive statistics and data visualization. We carried out two rounds of evaluations with data from 12,400 users of IntelliCare, a mental health platform with 12 apps. First, we focused to proof concept and second, we assessed reproducibility by drawing conclusion from distribution differences. User data was drawn from both research trials and public deployment on Google Play. RESULTS Our algorithms showed advantages over commonly used concepts and reproduce in our public data set with different underlying behavioral strategies. These measures relate to the distribution of a user’s allocated attention, users’ circadian behavior, their consecutive commitment to a specific strategy, and users’ interaction trajectory. Because distributions between research trial and public deployment were similar, consistency was implied regarding the underlying behavioral strategies: psychoeducation and goal setting are used as a catalyst to overcome the users’ primary obstacles, sleep hygiene is addressed most regularly, while regular self-reflective thinking is avoided. Relaxation as well as cognitive reframing have increased variance in commitment among public users, indicating the challenging nature of these apps. The relative course of users’ engagement is similar in research and public data. CONCLUSIONS We argue that deliberate, a-priori feature engineering is essential for reproducible, tangible, and explainable study analyses. Our features enable improved results as well as interpretability, providing an increased understanding of how people engage with multiple mental health apps over time. Since we based the generation of features on generic interaction, these methods are applicable to further methods of study analysis and digital health.

scholarly journals Characteristic Latent Features for Analyzing Digital Mental Health Interaction and Improved Explainability

2020 ◽  
Author(s):  
Max-Marcel Theilig ◽  
Ashley Arehart Knapp ◽  
Jennifer Nicholas ◽  
Rüdiger Zarnekow ◽  
David Curtis Mohr
Keyword(s):  
Mental Health ◽  
Data Science ◽  
Digital Health ◽  
A Priori ◽  
Sensor Technology ◽  
User Engagement ◽  
Behavioral Strategies ◽  
Data Set ◽  
Public Data ◽  
Latent Features

Abstract Background: Using smartphones and wearable sensor technology has sparked a broad engagement of data science and machine learning methods to leverage the complex, assorted amount of data. Despite verified processes, there is a reported underdevelopment of user engagement concepts, and the desire for high accuracy or significance has shown to lead to low explicability and irreproducibility. To overcome these issues, we aim to analyze principal characteristics of everyday behavior in digital mental health. Methods: We generated five latent features based on previous research, expert opinions from digital mental health, and informed by data. The features were analyzed with descriptive statistics and data visualization. We carried out two rounds of evaluations with data from 12,400 users of IntelliCare, a mental health platform with 12 apps. First, we focused to proof concept and second, we assessed reproducibility by drawing conclusion from distribution differences. User data was drawn from both research trials and public deployment on Google Play. Results: Our algorithms showed increased rationale for the basic usage of apps with different underlying behavioral strategies. Measures of the distribution of user’s allocated attention, the user’s circadian behavior, their consecutive commitment to a specific strategy, and users’ interaction trajectory are perceived as transferable to the public data set. Because distributions between research trial and public deployment were similar, consistency was shown regarding the underlying behavioral strategies: psychoeducation and goal setting are used as a catalyst to overcome the users’ primary obstacles, sleep hygiene is addressed most regularly, while regular self-reflective thinking is avoided. Relaxation as well as cognitive reframing have increased variance in commitment among public users, indicating the challenging nature of these apps. The relative course of the engagement (learning curve) is similar in research and public data. Conclusions: The deliberate, a-priori engineered features were reproducible across app users from both data sets. These features led to improved results as well as increased interpretability, providing an increased understanding of how people engage with multiple mental health apps over time. Since we based the generation of features on generic interaction proxies, these methods are applicable to other cases in artificial intelligence and digital health.

scholarly journals Design a Chatbot for Chennai Corporation Using Logistic Regression Algorithm

2019 ◽  
Vol 8 (2S11) ◽  
pp. 2320-2323
Keyword(s):  
Logistic Regression ◽  
End User ◽  
Text Data ◽  
Data Set ◽  
Voice Input ◽  
The Public ◽  
Logistic Regression Algorithm ◽  
Voice Data ◽  
Online Sources ◽  
The Voice

Chatbots are the famous nowadays in business because of its service offered to the community at large. They provide support of 24*7 for business in terms of customercare, helpline, planning, analyzing and decision making. In this paper, a chatbot for Chennai corporation is proposed. This chatbot helps the citizens in providing the responses for their queries related to civic problems. There is no such system is available to handle the public grievances automatically. This system handles the public query and the relevant suggestion and responses will be given promptly. The chatbot receives the text or voice input and processed. The voice recognition module used to recognize the voice query and the voice to text convertor used to convert the voice data into text format. The matchmaking process used to match the input query with the available data set and the relevant responses is generated. If no match is for the query, the matchmaker will find the relevant response from online sources. The output channel equipped with the text to voice converter which converts the text data into voice and it will be delivered to the end user. The naïve bayers and logistic regression algorithm is implemented for classifying the query and the performance is compared. The result shows that the logistic regression algorithm outperform well with the precision and recall values.

scholarly journals Characteristic Latent Features for Analyzing Digital Mental Health Interaction and Improved Explainability

2020 ◽  
Author(s):  
Max-Marcel Theilig ◽  
Ashley Arehart Knapp ◽  
Jennifer Nicholas ◽  
Rüdiger Zarnekow ◽  
David Curtis Mohr
Keyword(s):  
Mental Health ◽  
Data Science ◽  
Digital Health ◽  
A Priori ◽  
Sensor Technology ◽  
User Engagement ◽  
Behavioral Strategies ◽  
Data Set ◽  
Public Data ◽  
Latent Features

Abstract Background: Using smartphones and wearable sensor technology has sparked a broad engagement of data science and machine learning methods to leverage the complex, assorted amount of data. Despite verified processes, there is a reported underdevelopment of user engagement concepts, and the desire for high accuracy or significance has shown to lead to low explicability and irreproducibility. To overcome these issues, we aim to analyze principal characteristics of everyday behavior in digital mental health. Methods: We generated five latent features based on previous research, expert opinions from digital mental health, and informed by data. The features were analyzed with descriptive statistics and data visualization. We carried out two rounds of evaluations with data from 12,400 users of IntelliCare, a mental health platform with 12 apps. First, we focused to proof concept and second, we assessed reproducibility by drawing conclusion from distribution differences. User data was drawn from both research trials and public deployment on Google Play. Results: Our algorithms showed increased rationale for the basic usage of apps with different underlying behavioral strategies. Measures of the distribution of user’s allocated attention, the user’s circadian behavior, their consecutive commitment to a specific strategy, and users’ interaction trajectory curve are perceived as transferable to the public data set. Because distributions between research trial and public deployment were similar, consistency was shown regarding the underlying behavioral strategies: psychoeducation and goal setting are used as a catalyst to overcome the users’ primary obstacles, sleep hygiene is addressed most regularly, while regular emotional exposure is avoided. Relaxation as well as cognitive reframing have increased variance in commitment among public users, indicating the challenging nature of these apps. The relative course of the engagement (learning curve) is similar in research and public data. Conclusions: The deliberate, a-priori engineered features were reproducible across app users from both data sets. These features led to improved results as well as increased interpretability, providing an increased understanding of how people engage with multiple mental health apps over time. Since we based the generation of features on generic interaction proxies, these methods are applicable to other cases in artificial intelligence and digital health.

Prevalence of Self-Reported Depression Symptoms and Perceived Anxiety Among Community-Dwelling U.S. Adults Reporting Tinnitus

2020 ◽  
Vol 5 (4) ◽  
pp. 959-970
Author(s):  
Kelly M. Reavis ◽  
James A. Henry ◽  
Lynn M. Marshall ◽  
Kathleen F. Carlson
Keyword(s):  
Mental Health ◽  
Fold Increase ◽  
Community Dwelling ◽  
Self Report ◽  
Depression Symptoms ◽  
Community Based Interventions ◽  
Data Set ◽  
Health Distress ◽  
Validated Questionnaire ◽  
Mental Health Distress

Purpose The aim of this study was to examine the relationship between tinnitus and self-reported mental health distress, namely, depression symptoms and perceived anxiety, in adults who participated in the National Health and Nutrition Examinations Survey between 2009 and 2012. A secondary aim was to determine if a history of serving in the military modified the associations between tinnitus and mental health distress. Method This was a cross-sectional study design of a national data set that included 5,550 U.S. community-dwelling adults ages 20 years and older, 12.7% of whom were military Veterans. Bivariable and multivariable logistic regression was used to estimate the association between tinnitus and mental health distress. All measures were based on self-report. Tinnitus and perceived anxiety were each assessed using a single question. Depression symptoms were assessed using the Patient Health Questionnaire, a validated questionnaire. Multivariable regression models were adjusted for key demographic and health factors, including self-reported hearing ability. Results Prevalence of tinnitus was 15%. Compared to adults without tinnitus, adults with tinnitus had a 1.8-fold increase in depression symptoms and a 1.5-fold increase in perceived anxiety after adjusting for potential confounders. Military Veteran status did not modify these observed associations. Conclusions Findings revealed an association between tinnitus and both depression symptoms and perceived anxiety, independent of potential confounders, among both Veterans and non-Veterans. These results suggest, on a population level, that individuals with tinnitus have a greater burden of perceived mental health distress and may benefit from interdisciplinary health care, self-help, and community-based interventions. Supplemental Material https://doi.org/10.23641/asha.12568475

Text Messaging as a Mental Health Treatment Intervention: A Corpus-Based Study

2019 ◽  
Author(s):  
Megan Partch ◽  
Cass Dykeman
Keyword(s):  
Mental Health ◽  
Mental Health Treatment ◽  
Text Messaging ◽  
Low Cost ◽  
Statistical Significance ◽  
Text Message ◽  
Treatment Session ◽  
Text Messages ◽  
Health Treatment ◽  
Emotional Tone

Mental health treatment providers seek high-impact and low-cost means of engaging clients in care. As such, text messaging is becoming more frequently utilized as a means of communication between provider and client. Research demonstrates that text message interventions increase treatment session attendance, decrease symptomology, and improve overall functioning. However, research is lacking related to the linguistic make up of provider communications. Text messages were collected from previously published articles related to the treatment of mental health disorders. A corpus of 39 mental health treatment text message interventions was composed totaling 286 words. Using Linguistic Inquiry and Word Count (LIWC) software, messages were analyzed for prevalence of terminology thought to enhance client engagement. Clout, demonstrating the writer’s confidence and expertise, and positive Emotional Tone were found to be at a high level within the corpus. Results demonstrated statistical significance for five linguistic variables. When compared with national blog norms derived from Twitter, Clout, Emotional Tone, and use of Biological terminology were found to be at higher rates than expected. Authenticity and Informal terminology were found at significantly lesser rates.

scholarly journals Positioning for success: building capacity in academic competencies for early-career researchers in sub-Saharan Africa

2019 ◽  
Vol 6 ◽  
Author(s):  
C. Merritt ◽  
H. Jack ◽  
W. Mangezi ◽  
D. Chibanda ◽  
M. Abas
Keyword(s):  
Mental Health ◽  
Capacity Building ◽  
Low Cost ◽  
Early Career ◽  
Sub Saharan Africa ◽  
Training Experience ◽  
African Countries ◽  
Early Career Researchers ◽  
Academic Competencies ◽  
Sub Saharan

Background. Capacity building is essential in low- and middle-income countries (LMICs) to address the gap in skills to conduct and implement research. Capacity building must not only include scientific and technical knowledge, but also broader competencies, such as writing, disseminating research and achieving work–life balance. These skills are thought to promote long-term career success for researchers in high-income countries (HICs) but the availability of such training is limited in LMICs. Methods. This paper presents the contextualisation and implementation of the Academic Competencies Series (ACES). ACES is an early-career researcher development programme adapted from a UK university. Through consultation between HIC and LMIC partners, an innovative series of 10 workshops was designed covering themes of self-development, engagement and writing skills. ACES formed part of the African Mental Health Research Initiative (AMARI), a multi-national LMIC-led consortium to recruit, train, support and network early-career mental health researchers from four sub-Saharan African countries. Results. Of the 10 ACES modules, three were HIC-LMIC co-led, four led by HIC facilitators with LMIC training experience and three led by external consultants from HICs. Six workshops were delivered face to face and four by webinar. Course attendance was over 90% and the delivery cost was approximately US$4500 per researcher trained. Challenges of adaptation, attendance and technical issues are described for the first round of workshops. Conclusions. This paper indicates that a skills development series for early-career researchers can be contextualised and implemented in LMIC settings, and is feasible for co-delivery with local partners at relatively low cost.

Sign in / Sign up

Export Citation Format

Share Document