Design Thinking and a Human-Centered Approach to Explore the

Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health

Solution for University Students in India

Raman Saxena

Software Engineering Research Center (SERC) and Product Design and Management (PDM) Program, International

Institute of Information Technology (IIIT) Hyderabad, India

Keywords:

Smart Phones, Mobile Phones, Artiﬁcial Intelligence, Mental Health, University Students, Loneliness, Stress

and Anxiety, Personalized Mental Health Support, Design Thinking, Human-Cantered Design, Usability,

India.

Abstract:

Anxiety, stress and depression are the signiﬁcant mental health and well-being challenges being faced by the

university students, early one in three reporting signiﬁcant struggles. Academic pressure, family expectations,

a competitive environment, social isolation, ﬁnancial stress and stigma surrounding mental health contribute

to this issue. These challenges impacts students’ academic performance and social integration negatively,

which further impacts their mental health and well-being. Given high mobile phone usage among youths,

smartphones offer a unique, discreet avenue, for mental health support. By leveraging device sensors like

accelerometers, gyroscopes, GPS, proximity sensors, and biometric readers (e.g. heart rate, SpO2), a mobile

framework can analyze user activity pattens, social interactions, and screen time to detect early signs of mental

health concerns, such as stress, anxiety or loneliness. Integrating this data with trained mental health models

enhances predictive accuracy, enabling personalized help and therapeutic content like calming music, mind-

fulness exercises, or relaxation videos, Notiﬁcations, and chat bot conversations as a virtual buddy, tailored to

their preferences. The framework uses smartphones as an unobtrusive wellness companion, aiming to prevent

mental health deterioration while safeguarding, user privacy, thus empowering students with a personal tool

for mental health well-being.

1 INTRODUCTION

The prevalence of signiﬁcant mental health disorders,

especially stress, anxiety, and depression, is rising

manifold among college students. They face many

challenges, including academic pressure, social iso-

lation, and the stigma around seeking help (Hunt

and Eisenberg, 2010), all of which can affect their

ability to manage stress and anxiety. According to

the World Health Organization (WHO, 2020) report,

mental health conditions in this population are in-

creasing at a rapid pace, with nearly one in every three

students experiencing mental health difﬁculties. Ap-

proximately 40 of students in India reported symp-

toms of depression, and up to up to 70% of students

reported stress and anxiety (Dutta et al., 2023). Lone-

liness has been identiﬁed as a strong predictor of de-

pression and anxiety(Richardson et al., 2017), and

may negatively impact both academic performance

and social adjustment to the university environment

(Danneel et al., 2018). A survey in the UK found that

34% of students aged 18–24 felt lonely to some de-

gree (Venkatesh et al., 2015). The transition to digital

learning modes has increased the mental health bur-

den due to factors like increased screen time, time, re-

duced physical interaction, and the pressure of adapt-

ing to new technologies (Eisenberg et al., 2009). As

per the National Crime Records Bureau (NCRB) 2022

report, young adults aged 18-30 years accounted for

35% of all suicides in 2022; NCRB statistics also in-

dicate that, one student committed suicide every hour

(Reddy et al., 2018) (Kemp, 2023), highlighting the

urgent and critical need to address mental health con-

ditions among this population.

1.1 Design Thinking and Human-

Centered Design in Mental Health

and Well-Being

Design Thinking (DT) and Human-Centered Design

(HCD) are steadily more acknowledged for their re-

508

Saxena, R.

Design Thinking and a Human-Centered Approach to Explore the Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health Solution for University Students in India.

DOI: 10.5220/0013288200003928

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 20th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2025), pages 508-519

ISBN: 978-989-758-742-9; ISSN: 2184-4895

markable contributions to the design and develop-

ment of mental health well-being solutions that serve

users’ speciﬁc needs, particularly in navigating the

challenges of seeking mental health support. Empa-

thy is the core of the DT that enables designers to

gain mental health and well-being concerns from the

end user’s perspective. In their research, Brown and

Katz, 2011, illuminate DT’s effectiveness in foster-

ing creative problem-solving by prioritizing the emo-

tional experiences of users and addressing critical bar-

riers such as stigma, accessibility, and user engage-

ment. The close alignment of Human-centered de-

sign (UCD) and Design Thinking (DT) make sure

that the solutions are accessible, easy to use, and re-

sponsive, leading to higher trust and usability (IDEO,

2015). Improving the features and enhancing user en-

gagement and satisfaction with digital mental health

and well-being solutions, as well as iterative testing

and feedback loops, can contribute very effectively

(Torous et al., 2018).

1.2 Mobile Usage Behavior Among

Students and Mental Health

Mobile phones are the widely adopted technology

globally, playing a central role in our daily lives

across the population, and university students make

up a signiﬁcant portion of mobile phone users. Uni-

versity students living in hostels spend considerable

time using mobile phones for communication, brows-

ing the internet, social media, and entertainment. As

of early 2023, India had 1.10 billion active cellular

mobile connections, representing 77.0% of the total

population (IMRB, 2016). The Mobile Marketing As-

sociation, in collaboration with Kantar (IMRB, 2016),

reported that the average consumer spends 3 hours a

day on their smartphones, exceeding the time spent

on television and other forms of media (Carbonell

et al., 2012). Many young people admit to never turn-

ing off their smartphones, keeping them by their side

while they sleep, and compulsively checking them

throughout the day (Grifﬁths and Kuss, 2017). A

growing body of research investigates the relationship

between mobile phone usage and mental health out-

comes. Although mobile phones are valuable com-

munication and social connection tools, excessive or

maladaptive usage can contribute to mental health is-

sues. Many studies have reported that higher mobile

phone use, especially in the context of social media,

is linked with enhanced levels of anxiety, depression,

and stress among university students (Tarafdar et al.,

2015).

1.3 Mobile Phone, AI & Mental Health

While mobile phones can add to mental health prob-

lems, considering the limited availability of men-

tal health professionals, and the limited accessibil-

ity of mental health care in remote and rural areas

(Chandrashekar, 2018), they possess considerable po-

tential to provide adequate mental health interven-

tions (Ahmed et al., 2021). Mobile-based interven-

tions, such as mental health and well-being apps, have

gained popularity as a means of providing accessible

and scalable support to university students. Artiﬁcial

Intelligence (AI) is gradually being acknowledged as

a critical technology for diagnostic , assessment, in-

tervention, and real-time support in mental health and

well-being. Several empirical evidence indicate that

(AI)-driven applications and systems can reﬁne the

assessment and diagnosis of mental health disorders,

deliver real time and personalized interventions, and

help in mitigating the stigma associated with coping

and seeking mental health help. Research has proven

that AI-powered chat-bots, such as “Woebot”, effec-

tively lessen the symptoms associated with anxiety

and depression (Fitzpatrick et al., 2017). AI tools of-

fer a higher degree of personalization that traditional

therapeutic methods may struggle to match, particu-

larly in under served populations(Jacobson and Feng,

2022).

Integrating mobile phone technology and AI has

a very high potential to reshape the access and de-

livery of mental health treatment fundamentally. In

the past decade, mobile applications focused on men-

tal health have surged in popularity and are beginning

to be utilized in university environments as a valu-

able resource for supporting and addressing mental

health issues among college students. Indian univer-

sities, with limited availability of mental health re-

sources, can leverage the mobile-based mental health

and well-being interventions to bridge the gap be-

tween students and professional help, however, user

engagement, cultural relevance, and the quality of the

app content will decide the effectiveness and success

of these interventions (Firth et al., 2017). Tailoring

the system for diverse cultural contexts and individ-

ual differences is challenging (Torous and Roberts,

2018) More research is needed to explore how mobile

mental health interventions can be tailored to univer-

sity students’ speciﬁc needs and challenges in differ-

ent cultural contexts.

Design Thinking and a Human-Centered Approach to Explore the Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health

Solution for University Students in India

509

2 METHODOLOGY

Applying Design Thinking and Human-Centered Ap-

proach, this research project explored the potential for

a Mobile Phone (Smartphone) and Artiﬁcial (AI) en-

abled Mental Health and Well-being solution for uni-

versity students in India. This approach started with

engaging with all the stakeholders and understand

them using the mixed method approach including user

survey, interviews, expert interviews, and technology

review to answer the research question: Can Smart

Phone enabled with AI Support Mental Health Well-

Being Among University Students in India?

9The research study stared with identifying the

stakeholders including people (University Students,

Mental Health and Well-being’s Experts) and tech-

nology involved (Hardware and Software of Mobile

Phones and Artiﬁcial Intelligence, in the Eco-system.

After identifying the stakeholders, we designed

following four studies to ﬁnd answers for our primary

research question covering all the stakeholders:

• Study 1: Online survey with the university stu-

dents to understand their mobile phone usage be-

havior

• Study 2: In-person interviews with the university

students to understand their mobile phone usage

behavior

• Study 3: Technology reviews of the mobile phone

technology including hardware and software

• Study 4: Interviews with the Mental health &

well-being professionals and experts

2.1 Study 1- Online Survey with the

University Students

Study Objective: To understand mobile phone us-

age behaviors and patterns among university students

and its impact on their mental health and well-being.

Several studies highlight critical behavioral patterns

associated with mobile phone usage, including screen

time and dependency. Prolonged use is often linked to

stress, reduced academic performance, and decreased

physical activity due to excessive reliance on phones

can escalate stress and social isolation; social media

platform usage ampliﬁes comparison and self-esteem

issues, which can affect students’ mental health (Ke-

les et al., 2020). The survey also explored how they

cope with mental health challenges and their help-

seeking behavior from their peers, professionals, and

technology, as several studies highlight that students

rely on peer support, including friends, for informal

Online Survey

counseling and shared experiences to manage stress

(Rickwood et al., 2007).

Methodology: An online survey was created in Mi-

crosoft form and shared with in IIIT students as well

as on social media platform including LinkedIn, Face-

book and other online networks of students within

India. The online survey particularly suited for this

study considering university students being frequent

users of mobile phones and digital platforms, making

it an intuitive method for collecting data in a medium

they are comfortable with (Evans and Mathur, 2018).

Online surveys also provide a sense of anonymity,

encouraging honesty and reducing social desirability

bias (Tourangeau and Yan, 2007). This online survey

was complemented with qualitative methods (Study

2: In-person interviews) to understand the situation

comprehensively..

Sample Size: 286 (154 IIIT students and 132 students

from other universities across India.

2.2 Study 2 In-Person Interview with

University Student

Study Objective: To understand and analyze uni-

versity students’ mental health and well-being needs

and examine their primary challenges and awareness

of available resources through in-person interviews.

This study explores their potential acceptance of a

mobile and AI-enabled mental health solution. It

identiﬁes key motivators and barriers to using AI-

based mental health tools, their preferences around

personalization, privacy, data control, and discreet-

ness and engagement features that contribute to sus-

tained use. Additionally, identify effectiveness and

meaningful outcomes indicators from the user’s per-

spective..

Methodology: In-person interviews, a qualitative

research method for in-depth insights allowing re-

searchers to establish rapport and create a safe en-

vironment for open and honest discussions (Knox

and Burkard, 2009). As it takes place in person,

the researchers can use non-verbal cues that provide

additional context to verbal responses (Opdenakker,

2006).

Interview Guide: Our interview included open-

ended, exploratory questions covering areas such as

accessibility, personalization, and privacy concerns,

as well as desired outcomes and metrics for success.

The interview covered the following:

• Introduction and Warm-Up Questions to under-

stand university students’ existing mental health

and well-being management strategies, their fa-

In-person Interview Questionnaire

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

510

miliarity and experiences with digital mental

health and well-being tools and understand partic-

ipants’ views on the potential role of AI and mo-

bile technologies in enhancing mental health sup-

port, including any expectations or reservations.

• Potential Acceptance including their interest and

motivation, desired features, & identifying needs

related to customization & personalization.

• Identifying Barriers to Usage, including privacy

and security concerns, needs relating to data trans-

parency, discreetness and secrecy, and technical

accessibility, usability and ease of use, etc.

• Desired Outcomes and Success Metrics, includ-

ing User-centered success metrics ,engagement

and satisfaction, and long-term impact and behav-

ioral changes.

• Closing Thoughts to gather user-speciﬁed fea-

tures that would make an AI-driven mental health

app both effective and trustworthy and gather par-

ticipants’ suggestions for enhancing its relevance

and helpfulness.

Sample Size: 21 students from IIIT Hyderabad and

few other institutes/colleges around Hyderabad.

2.3 Study 3 Technology Review (Mobile

Phones and AI)

Study Objective: To explore and analyze the role

of mobile phone technology, including its hardware

components (e.g., sensors) and software applications,

in tracking and interacting with users’ behavioral

data. This study investigates how these technologi-

cal features including mobile phone hardware, built in

sensors and software can capture relevant behavioral

indicators, assess their accuracy in reﬂecting mental

health states, and evaluate their potential for support-

ing AI-enabled mobile phone based mental health so-

lutions tailored to individual needs.

Phone Devices: We used two phones’ models (one

Android phones and one IOS Phone) for conducting

the technology review:

• OnePlus 7, Model No. GM1901 - Majority of

university found to be using Android Phone

• Apple iPhone 13

2.4 Study 4-Interview with the Mental

Health Experts

Study Objective: This research aims to understand

essential behavioral indicators for assessing mental

Interview Questionnaire

health, focusing on patterns like sleep, activity, so-

cial interactions, and signs of anxiety, depression,

or stress. It will explore how mobile and digital

data— such as social media usage, activity levels, and

communication frequency—can contribute to a com-

prehensive mental health proﬁle. Additionally, the

study seeks expert opinions on optimal data collection

frequency and preferences for passive versus user-

reported data, ensuring minimal intrusiveness while

maintaining accuracy. Lastly, it will address ethical

considerations, such as privacy and data control mea-

sures to build patient trust and uphold ethical stan-

dards.

Methodology: In-person interviews with 3 mental

health and well-being experts including Counselor in

and around Hyderabad..

Interview Guide: The interview focused on gaining

insights into essential behavioral indicators for assess-

ing mental health and exploring how digital data can

aid in building a comprehensive mental health pro-

ﬁle for users. The process included introduction and

context setting and covered queries related to Behav-

ioral Indicators for Mental Health Assessment, The

potential of Mobile and Digital Data, Data Collection

Preferences, Privacy, Security, and Ethical Consider-

ations, and Engagement and Trust-building measures.

3 RESEARCH FINDINGS

Findings from four studies are explained below:

3.1 Findings: Research Study 1 - Online

Survey with Students

3.1.1 Compulsive Phone Checking

The survey reveals a solid attachment to mobile

phones, especially checking ﬁrst thing after waking

up and during social or academic activities, which

points to the potential for phone overuse or addiction.

This can disrupt focus and reduce face-to-face social

interactions, leading to further dependency on mobile.

3.1.2 Academic Distraction and Need for

Balance Usage

Mobile phone usage is a signiﬁcant distraction during

classes, which can undermine students’ ability to ab-

sorb the material and engage in learning, leading to

compromised academic performance and adding fur-

ther to mental health and well-being. While some

students perceive mobile phones as beneﬁcial to their

academic performance, most see them as distracting.

Design Thinking and a Human-Centered Approach to Explore the Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health

Solution for University Students in India

511

This implies the need for interventions or strategies to

promote balanced usage that minimizes negative im-

pacts while leveraging positive aspects, such as ac-

cessing learning resources.

3.1.3 High Academic & Personal Pressure

A signiﬁcant portion of students (60.75%) report feel-

ing overwhelmed by academic or personal pressures

frequently or consistently, suggesting mental strain is

expected, which could impact their well-being and

academic performance.

3.1.4 Concentration Challenges

Most (91.77%) of students ﬁnd it difﬁcult to focus,

with 48.12% reporting that they lose concentration

and focus frequently or consistently. This indicates

that many students struggle with concentration due to

stress or external pressures.

3.1.5 Inadequate Sleep and Disruption

The survey reveals that nearly two-thirds of the stu-

dents (63.29%) sleep less than the recommended

sleep of 7 Hours for young adults. This further con-

tribute to the stress, reduced concentration, and over-

all health issues. Many students (63.29%) reported

occasional or frequent sleep disruptions, which cor-

related to their reported stress levels, impacting their

ability to maintain healthy sleep patterns.

3.1.6 Burden and Sources of Stress, Anxiety,

and Depression

The survey result highlights a concerning appear-

ance of mental health issues among university stu-

dents, with a signiﬁcant 83% experiencing anxiety

or panic at varying levels, while only 19.22% re-

port rarely or never facing these emotions. Stress is

especially prevalent, impacting nearly 70% of stu-

dents, followed by anxiety at 56.7% and low self-

esteem at 51.4%. A signiﬁcant number of students

also express feelings of loneliness (47.88%), hope-

lessness (46.13%), and disconnection (44.66%), in-

dicating that many students struggle with a combina-

tion of emotional and mental challenges. 88% of stu-

dents indicating they often or occasionally feel men-

tally and emotionally drained, suggests a persistent

difﬁculty in maintaining resilience and overall well-

being. About sources of stress, academic pressure

was reported by 81%, whereas 57% indicated uncer-

tainty about the future.37% of them blamed bad per-

sonal relations as the cause of their stress.

3.1.7 Coping Mechanism and Help Seeking

Behavior

The survey data shows a strong inclination towards

taking charge of their mental well-being among uni-

versity students, with 60% of them choosing to man-

age their health on their own and avoiding seeking

out help from professional assistance, and only 6%

of students sought help from mental health profes-

sionals, due to concerns about stigma and judgment.

Instead, the survey data shows that the students rely

more on informal coping options, including, seek-

ing support from friends (67%), participating in hob-

bies (61.82%), and browsing social media (56.73%).

Other approaches include ignoring their problems

(43.64%), engaging in physical exercise (36.36%),

and practicing meditation (22.18%). The nearly even

split between those open to seeking help versus those

reluctant highlights an opportunity to tackle obstacles

to professional support, such as stigma, by encour-

aging accessible, stigma-free mental health services

speciﬁcally designed for students.

3.1.8 Comfort Discussing Mental Well-Being

Issues

This survey shows that merely 26.83% of students

feel at ease discussing mental health issues with their

families and close friends, while 32.93% do not, it

highlights to considerable barriers to openness about

mental health within student communities. This hes-

itation to talk about mental health may stem from

belief that family or friends may not comprehend

the issues, or elements such as elements such as

stigma, the fear of being judged, which are particu-

larly widespread among younger populations. This

indicates the necessity for enhancing a peer support

system, creating judgment-free environments and in-

troducing mental health education and support system

in schools and universities for a better mental health

well-being among university students.

3.1.9 Potential of AI Enabled Mobile Phone

Based Mental Health & Well-Being

Solution

64.51% showing interest in AI- driven mobile solu-

tions for mental health and well- being reveals their

positive attitude for such solutions or applications.

However, only 7% are currently utilizing the avail-

able mobile solutions, suggests a huge gap between

interest in using the AI driven mobile solutions and

its actual utilization. Many research studies has indi-

cated that young adults, including university students,

tend to show a high level of openness to AI-driven

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

512

solutions for mental health because of their perceived

convenience, accessibility, and potential for person-

alization but it also imply a necessity for targeted

awareness and trust-building efforts that tackle stu-

dents’ concerns, such as those related to effectiveness,

privacy, and personalization, to improve the adoption

and effectiveness of these solutions.

3.1.10 Conclusion

The survey ﬁndings highlight signiﬁcant concerns re-

garding university students’ mental health and well-

being, with a high prevalence of compulsive phone

checking, academic distractions, and social discon-

nection tied to mobile phone usage. Students report

frequent distractions in academic settings, impact-

ing concentration and leading to compromised aca-

demic performance. Additionally, inadequate sleep

and pervasive feelings of stress, anxiety, and emo-

tional exhaustion were reported, with academic pres-

sures and uncertainty about the future identiﬁed as

primary stress sources. While many students are

open to using AI-enabled mobile solutions for mental

health support, only a small%age currently use such

resources, indicating an opportunity for interventions.

Despite some students expressing comfort discussing

mental health issues with family or friends, a no-

table%age remain hesitant, citing stigma as a barrier.

The survey insights highlight a critical need for an ac-

cessible and adequate mental health support system,

that address the complex and interconnected issues

of stress, emotional fatigue and self-esteem, and of-

fer some accessible mental health and well- being so-

lution which is personal, accessible, can prevent and

addresses students’ academic, emotional, and social

well-being real-time, discreetly, efﬁciently and effec-

tively for university students.

3.2 Findings: Research Study 2 -

In-Person Interviews with Students

3.2.1 University Students’ Existing Mental

Health Strategies

Most participants relied on informal networks like

friends and family for mental health support. Many

expressed challenges in openly discussing mental

health issues due to stigma or lack of understand-

ing. While some students had tried digital mental

health tools, such as mood-tracking apps or mind-

fulness platforms, their engagement was often short-

lived due to a lack of personalization and perceived

ineffectiveness.

3.2.2 Potential Role of AI and Mobile

Students were optimistic about AI’s potential to pro-

vide tailored interventions and predictive insights.

Many viewed mobile technologies as a convenient

platform for daily mental health management. Con-

cerns about data misuse and the inability of AI to fully

understand human emotions were frequently men-

tioned.

“I don’t want to share my mental health situation

with everyone as I am worried if they make judgment

about me, so mostly I engage myself with my mobile

phone browsing internet, listening to music, watching

video and other material which helps me come out of

that state and provide a sense of relief.”

3.2.3 Desired Features for Engagement and

Personalization

Key Preferences

• Customized UI to suit individual preferences.

• AI-driven personalization that adapts to mood pat-

terns and habits.

• Gamiﬁed elements to maintain high engagement.

Motivators

• Features like rewards, reminders, and real-time

feedback were seen as crucial to sustaining usage.

Privacy and Security Concerns

• Strong apprehensions about how sensitive mental

health data would be stored and used.

• A clear demand for transparency in data collection

and usage policies.

Discreetness

• Many participants highlighted the need for apps

to be unobtrusive and provide options for anony-

mous usage.

Usability Challenges

• Complex navigation and overly technical inter-

faces were cited as deterrents.

3.2.4 Data Control and Transparency Needs

Students emphasized the importance of having com-

plete control over their data, including options to

delete or restrict access. Research conﬁrms that han-

dling sensitive mental health data securely is critical

(32). Transparency features, such as detailed reports

on how data is used, were considered essential for

building trust.

Design Thinking and a Human-Centered Approach to Explore the Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health

Solution for University Students in India

513

3.2.5 Desired Outcomes and Success Metrics

User-Centered Metrics

• Improved mood stability, reduced anxiety

episodes, and better sleep were mentioned as

tangible success indicators.

• Metrics like sustained app engagement and high

satisfaction rates were seen as proxies for effec-

tiveness.

Long-Term Impact

• Participants valued solutions that could foster pos-

itive behavioral changes over time, such as devel-

oping healthier coping mechanisms.

3.2.6 Suggestions for Effectiveness and

Trustworthiness

Features to be Included

• Real-time support during crises, such as guided

breathing exercises or escalation to human coun-

selors.

• Integration with wearable devices for a more

comprehensive assessment of behavioral

patterns.Trust-Building Measures.

Trust-Building Measures

• Explicit opt-in mechanisms for data sharing and

robust encryption protocols.

• AI explain-ability features to demystify how in-

sights and recommendations are generated.

3.2.7 Conclusion

The insights highlight a signiﬁcant gap in existing

mental health apps, particularly in personalization,

privacy, and sustained engagement. AI and mobile

technologies have the potential to bridge this gap by

offering user-centric, adaptive, and trustworthy solu-

tions. However, addressing ethical concerns, ensuring

transparency, and delivering measurable outcomes are

critical to their success.

3.3 Findings from Research Study 3 -

Mobile Phones Technology Review

3.3.1 Hardware Components

Smartphones are equipped with a wide range of hard-

ware, sensors, and software that enable them to mon-

itor a variety of usage behaviors and activities of the

user. These can provide insights into user proﬁles,

mental and physical health, and overall well- being.

3.3.2 Sensors

There are multiple sensors are embedded in Smart-

phones which can help in monitoring various physi-

cal and behavioral aspects of the use including Ac-

celerometer that measures movement and orientation,

detecting activities like walking, running, or shak-

ing the phone. It can track physical activity levels

and identify sedentary behavior or physical exertion.

Gyroscope detects angular rotation, improving move-

ment accuracy. It can track physical activities, ex-

ercise, or even postures as it’s used in gaming and

apps. Magnetometer (Compass), provides orientation

relative to the Earth’s magnetic ﬁeld. It is helpful in

navigation apps but can also give insights into daily

movement patterns. GPS track’s location data, which

can provide information on travel habits, preferred en-

vironments, and social behaviors (e.g., visiting parks

or gyms). It also can be used for tracking outdoor

physical activity. Proximity Sensor detecting when

the phone is near the user’s face or ear can help mon-

itor screen usage or phone habits (e.g. if the user

spends more time on calls). Ambient Light Sensor

adjusts the screen brightness based on environmental

lighting, but it can also track the lighting conditions

in which the phone is used, giving insights into sleep-

wake cycles.

Additional sensors include the barometer, which

measures atmospheric pressure to calculate altitude

and contributes to activity tracking by identifying stair

climbing or elevation changes, and the temperature

sensor, which tracks external or body temperature

in health-focused applications, though its presence is

less common. Heart Rate Sensor tracks heart rate and

can indicate stress levels. SpO2 Sensor (Oxygen Sat-

uration) It found in some health-focused smartphones,

to monitor blood oxygen levels and overall cardio-

vascular health. Fingerprint Scanner, offers biometric

security that can log usage patterns related to device

unlocking, providing insights into phone usage fre-

quency

3.3.3 Software and Apps

The smartphone’s software is where much of the data

processing takes place and includes, Operating Sys-

tem (iOS, Android) that facilitates data collection and

analysis, as well as AI/ML algorithms that track and

interpret behavior. Health Apps (e.g., Google Fit, Ap-

ple Health) integrates sensor data to track physical

activity, steps, heart rate, and more. These apps ag-

gregate data over time to provide health reports. Ac-

tivity Trackers (e.g., Fit-bit, Samsung Health) collects

movement and ﬁtness data (steps, calories burned, ex-

ercise routines) and can provide insights into physi-

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

514

cal health. Voice Assistants (e.g., Siri, Google As-

sistant) track voice commands and communication

habits, possibly analyzing tone and word choice to

monitor stress levels or emotional health. Sleep Mon-

itoring Apps (e.g., Sleep Cycle) use accelerometer

and microphone data to track sleep patterns, includ-

ing time spent in different sleep stages, detecting po-

tential sleep disorders like insomnia. Mental Health

Apps (e.g., Calm, Head-space) Analyze voice tone,

app usage patterns, or even physical behaviors (such

as tapping) to assess mental well-being, anxiety, or

mood changes and Usage Behavior Apps (e.g., Digi-

tal Well- being) tracks phone usage, screen time, app

usage frequency, and notiﬁcations to analyses user

habits and potential signs of addiction, stress, or pro-

ductivity levels.

3.3.4 Conclusion

Smartphones can track a variety of behavior that con-

tribute to building a user proﬁle and understanding

their mental and physical health:

Physical Activity: Sensors like the accelerometer,

gyroscope, and GPS track steps, running, cycling, and

general movement. This data can be used to assess

whether the user meets recommended activity levels

for good health.

Sedentary Behavior: Extended periods without

movement can be ﬂagged to indicate inactivity, which

may suggest poor physical health or risk factors for

lifestyle-related diseases.

Sleep Pattern: By monitoring screen time, ambient

light, and activity before bed, smartphones can esti-

mate sleep duration and quality, alerting users to po-

tential sleep disorders.

Social Interaction: Monitoring call logs, text mes-

sage frequency, and time spent on social media apps

can give insights into social engagement or isolation,

a factor in mental health.

Emotional State and Stress Levels: Using AI algo-

rithms, smartphones can analyze voice patterns, typ-

ing speed, or app usage to detect changes in mood

or stress. For example, frequent use of mental health

apps or certain language cues in text messages may

indicate emotional distress.

Dietary Habits: Some apps allow users to log their

meals, and smartphone cameras can identify food

items, helping track nutritional intake. This data can

be used to assess diet quality and its impact on health.

Cognitive Performance: Gaming apps or cognitive

training tools can track mental agility, memory, and

focus over time. Changes in performance could indi-

cate stress, fatigue, or cognitive decline.

3.4 Findings from Research Study

3.4.1 Most Crucial Behavioral Indicators to

Assess Mental Health and Well-Being

The most crucial behavioral indicators to assess men-

tal health and well-being drawn from the interviews

with the three mental health and well- being experts

are:

Sleep Patterns

• Sleep patterns including quality, duration, and

consistency, are primary indicators of mental

health well-being.

• Disruptions in sleep (insomnia), irregular sleep

schedules, and excessive sleep are indicators of

mental health issues such as depression, anxiety,

and stress.

Physical Activities

• Physical activities have a positive impact on men-

tal health and well-being, especially in reducing

depression and anxiety; a reduction in physical ac-

tivities or lack of motivation to engage in physical

activities indicates issues with mental health and

well-being.

Social Interactions

• Social interactions serve as protective factors for

mental health and frequency and quality. Any re-

duction in social engagements, withdrawal from

social events, or changes in patterns of commu-

nication, like withdrawing from communication,

are early signs of depression and anxiety.

Observable Signs of Anxiety, Depression, and

Stress

• Observable signs such as restlessness, mood

swings, irritability, or lack of focus, are signals for

increased stress and anxiety levels. Close mon-

itoring of daily behavior can help identify such

symptoms, which must be addressed.

3.4.2 Use of Digital Footprints to Assess Mental

Health

In the current age of digital society, where mobile

phones and the internet have become integral parts of

our daily lives, digital engagement behaviors and pat-

terns, including usage of social media, browsing pat-

terns, and communication frequency, are emerging as

valuable indicators of mental health and well-being.

3.4.3 Conclusion

The use of digital footprints in mental health assess-

ment offers promising beneﬁts, including early detec-

tion, personalized care, and enhanced understanding

Design Thinking and a Human-Centered Approach to Explore the Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health

Solution for University Students in India

515

of patient experiences. However, it also presents sig-

niﬁcant challenges related to privacy, ethical use, and

the risk of misinterpretation.

3.5 Conclusion from the Four Studies

The ﬁndings from the studies collectively under-

score the immense potential of employing a human-

centered approach to develop mobile and AI-enabled

just-in-time mental health solutions for university stu-

dents in India. The studies reveals that the univer-

sity students in India face signiﬁcant mental health

challenges, including stress, anxiety, emotional ex-

haustion, and low self-esteem, often rooted in aca-

demic pressure, uncertainty about the future, and per-

sonal relationships. Compulsive mobile phone usage

and inadequate sleep patterns adds to the same. The

limited willingness to seek professional help due to

stigma highlights the need for discreet and stigma-

free solutions. The study also revealed that even

though the compulsive use of mobile phone is con-

tributing to these challenges but the hardware, soft-

ware and various sensors built into the smart phones,

captures various factors that are similar to those used

by the mental health professional to assess and mental

health and well-being conditions and also offer ther-

apy.

AI-enabled smartphones offers potential to de-

liver personalized, adaptive, and real-time support

metal health assessment and support by capturing be-

havioral and environmental data, enabled through its

advance hardware and sensors. Students express op-

timism about mobile technologies and AI’s potential

to provide just-in-time mental-health support due to

stigma highlights the need for discreet and stigma-

free solutions.

A human-centered design of mobile and AI-

enabled just-in-time mental health solutions has the

potential to empower university students to man-

age their well-being effectively. By leveraging

smartphones’ capabilities, AI and applying human-

centered approach we suggest a human-centered

framework (Refer ﬁgure 1) and “MindMate” (Refer

ﬁgure 2), A Mobile Phone and AI-Enabled Just-in-

Time Mental Health & Well-being Solution that can

be made available on mobile phones and tailored to

the unique needs and concerns related to the men-

tal health and well-being of Indian university stu-

dents. This framework can address privacy, personal-

ization, and stigma, such solutions can deliver adap-

tive, empathetic, and accessible mental health and

well-being support.

4 SUGGESTED FRAMEWORK &

SOLUTION

Insights from the four studies led us to the follow-

ing framework (Refer to Figure 1) that emphasizes a

holistic, user-centered approach to creating an AI- en-

abled just-in-time (JIT) mental health solution that is

available on mobile phones and tailored to the unique

needs and concerns of Indian university students. In-

tegrating AI’s capabilities this framework offer effec-

tive, personalized, engaging and ethical support for

mental well-being.

4.1 Core Pillars of the Framework

4.1.1 Personalization

• Using AI, create dynamic user proﬁles based on

behavioural data (e.g., sleep patterns, physical ac-

tivity, social interactions).)

• Offer adaptive content delivery tailored to the

user’s mental health status (e.g., music, video,

content recommendations, suggest mindfulness

exercises, motivational messages, chat/ conversa-

tions with AI agent as Virtual buddy).

4.1.2 Apply Privacy & Ethical Use of User Data

Must implement a transparent data governance policy.

Use secure data encryption and anonymization. Ad-

here to consent-driven data collection practices and

local data protection regulations.

4.1.3 Offer Engagement and Gamiﬁcation

Integrate gamiﬁcation techniques to encourage moti-

vated engagements and sustained use. Offer rewards

to motivate users to encourage healthy behaviors like

following regular sleep, engaging in physical activity,

and with mental health resources.

4.2 Key Features

4.2.1 Real-Time Monitoring and Feedback

Using data inputs from the smartphone sensors moni-

tor critical indicators like:

• Sleep: Screen activity, ambient light levels.

• Physical Activity: Steps, movement patterns.

• Social Interaction: Communication frequency

and time spent on apps.

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

516

Figure 1: Framework for the mobile phone and AI-Enabled JIT Mental Health & Wellbeing Solution.

4.2.2 AI-Driven Insights

• Analyse digital footprints (browsing behaviour,

text sentiment analysis) to detect early signs of

mental health issues like depression/ anxiety.

• Predict and provide recommendations for poten-

tial risks using predictive analytics.

4.2.3 Behavioural Nudges

Send actionable notiﬁcations based on tracked pat-

terns (e.g., reminders to move after prolonged inac-

tivity, relaxation techniques during stressful periods).

4.2.4 Therapeutic Recommendations

Offers therapeutic recommendations such as music,

videos, games, or other content that can serve as sup-

port in preventing or offering some relief when any

stress, depression, or anxiety is tracked in the users.

4.2.5 Self-Help Tools

Offers Guided meditation, cognitive behavioral ther-

apy (CBT) exercises, and journaling features

4.2.6 Human Support Integration

• If needed, it links users to university counsellors

or helplines when critical thresholds are met.

4.3 “MindMate”, a Mobile and

AI-Enabled Just-in-Time Mental

Health & Wellbeing Solution

Considering the suggested framework above (Refer to

Figure 2), “MindMate” is conceptualized to empower

university students by fostering emotional resilience,

mental clarity, and personal growth through a seam-

less, AI-driven mobile experience. It transforms the

everyday smartphone into a supportive “AI-enabled

virtual buddy,” blending advanced AI with behavioral

insights, such as Sentiment Analysis and Social Sup-

port Theory to dynamically respond to users’ emo-

tional states and support mental well- being.

Figure 2: ”MindMate” Concept.

“MindMate” integrates AI with mobile hardware,

sensors, and software to subtly track user behaviours,

activities, and biometric data, predicting moments of

Design Thinking and a Human-Centered Approach to Explore the Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health

Solution for University Students in India

517

stress, anxiety, or low mood. During such moments,

it intuitively suggests curated content, whether mu-

sic, videos, reading material, or games contextually

(e.g., recommending some music tracks, etc., when

they browse music- a recommendation not forced)

that helps uplift and soothe the user. This tailored sup-

port respects each user’s autonomy and emotional pri-

vacy, gently reinforcing positive mental health habits

without overtly identifying triggers.

“MindMate” also offers a pro-active mode where

Virtual Buddy initiates Chat with the users when it

ﬁnds some abnormality in the user behaviour direct-

ing towards some stress, depression, anxiety etc. Pro-

viding just-in-time help to overcome the same.

In a context where mental health support must be

personal, empathetic, and culturally sensitive, “Mind-

Mate” aspires to redeﬁne student well-being by be-

coming a compassionate ally. By merging technology

with mindful, unobtrusive engagement, “MindMate”

aims to empower students to naturally cultivate re-

silience, self-care, and balance in their everyday lives.

4.3.1 Core Components of the “MindMate”

Figure 3: Core Components of the ”MindMate”.

4.3.2 ”MindMate” Workﬂow

Figure 4: ”MindMate” Workﬂow.

Core Components of the “MindMate”

”MindMate” Workﬂow

4.3.3 ”MindMate” Architecture

Figure 5: Core Components of the ”MindMate”.

5 CONCLUSIONS

This framework emphasises a holistic, user-centred

approach to creating an AI-enabled just-in-time

(JIT) mental health solution that is aligned to the

unique needs and concerns of Indian university stu-

dents. By integrating Design Thinking and Human-

centered principles with AI’s capabilities, Mind-

Mate, can provide effective, personalized, and eth-

ical support for mental well- being leading to

Improved Mental Health Outcomes- Early detec-

tion of mental health issues, Better engagement in

health-promoting behaviours; Enhanced Academic

Performance-Improved sleep and reduced stress lev-

els; Aggregated insights for universities to design tar-

geted mental health interventions.

6 FUTURE WORK

Based on our proposed framework and the concept

solution, we plan to build a prototype of “MindMate,”

an AI-enabled virtual buddy available on mobile to

support users in their mental health & well-being.

This prototype will then be tested for its usability

and functionality by inviting the university students.

Using a convenience sampling approach, we plan to

invite around 20 students between 18 and 26 years.

They will be asked to use this product for 4 to 5

weeks, and then will be interviewed about their expe-

riences, mental health well-being status, and sugges-

tions they have to make this solution better and more

impactful.

”MindMate” Architecture

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

518

ACKNOWLEDGMENTS

I extend my sincere thanks to all who in some way

or the other supported this study including the stu-

dent participants for their survey responses and in-

terviews and to the counselors and mental health ex-

perts for their invaluable insights. I would also like to

thank Prof. Raghu for sharing the CfP for ENASE25

and Prof. Karthik for helping me with some ques-

tions relate to Architecture and Latex, and my stu-

dents, MS & Honors Student’s Aditya and Dileep for

their help in formatting the paper. Their contributions

shaped the framework and solution, “MindMate,” and

enhanced my understanding of mental health method-

ologies, tools, and behavioral data for this study.

REFERENCES

(2020). Who report 2020. World Health Organization.

Ahmed, A., Ali, N., Aziz, S., Abd Alrazaq, A. A., Hassan,

A., Khalifa, M., and Househ, M. (2021). A review

of mobile chatbot apps for anxiety and depression and

their self-care features. Computer Methods and Pro-

grams in Biomedicine Update, 1:100012.

Carbonell, X., Chamarro, A., Grifﬁths, M., Oberst, U.,

Cladellas, R., and Talarn, A. (2012). Problematic

use of the internet and mobile phones in spanish ado-

lescents and young people. Annals of Psychology,

28(3):789–796.

Chandrashekar, P. (2018). Do mental health mobile apps

work: evidence and recommendations for designing

high-efﬁcacy mental health mobile apps. Mhealth, 4.

Danneel, S., Maes, M., Vanhalst, J., Bijttebier, P., and

Goossens, L. (2018). Developmental change in lone-

liness and attitudes toward aloneness in adolescence.

Journal of Youth and Adolescence, 47:148–161.

Dutta, G., Rajendran, N., Kumar, T., Varthya, S. B., and

Rajendran, V. (2023). Prevalence of depression among

undergraduate medical students in india: A systemic

review and meta-analysis. Cureus, 15(1).

Eisenberg, D., Downs, M. F., Golberstein, E., and Zivin,

K. (2009). Stigma and help seeking for mental health

among college students. Medical Care Research and

Review, 66(5):522–541.

Evans, J. R. and Mathur, A. (2018). The value of online

surveys: A look back and a look ahead. Internet Re-

search, 28(4):854–887.

Firth, J., Torous, J., Nicholas, J., Carney, R., Pratap, A.,

Rosenbaum, S., and Sarris, J. (2017). The efﬁcacy

of smartphone-based mental health interventions for

depressive symptoms: a meta-analysis of randomized

controlled trials. World Psychiatry, 16(3):287–298.

Fitzpatrick, K. K., Darcy, A., and Vierhile, M. (2017). De-

livering cognitive behavior therapy to young adults

with symptoms of depression and anxiety using a fully

automated conversational agent (woebot): a random-

ized controlled trial. JMIR Mental Health, 4(2):e7785.

Grifﬁths, M. D. and Kuss, D. (2017). Adolescent social

media addiction (revisited). Education and Health,

35(3):49–52.

Hunt, J. and Eisenberg, D. (2010). Mental health prob-

lems and help-seeking behavior among college stu-

dents. Journal of Adolescent Health, 46(1):3–10.

IDEO (2015). The Field Guide to Human-Centered Design.

IDEO.org.

IMRB (2016). Report by Mobile Marketing Association, in

collaboration with Kantar.

Jacobson, N. C. and Feng, B. (2022). Digital phenotyping

of generalized anxiety disorder: using artiﬁcial intel-

ligence to accurately predict symptom severity using

wearable sensors in daily life. Translational Psychia-

try, 12(1), 336.

Keles, B., McCrae, N., and Grealish, A. (2020). A sys-

tematic review: the inﬂuence of social media on de-

pression, anxiety and psychological distress in ado-

lescents. International Journal of Adolescence and

Youth, 25(1):79–93.

Kemp, S. (2023). Digital 23: India. Dataportal.

Knox, S. and Burkard, A. W. (2009). Qualitative research

interviews. Psychotherapy Research, 19(4–5):566–

575.

Opdenakker, R. J. G. (2006). Advantages and disadvantages

of four interview techniques in qualitative research.

Forum Qualitative Sozialforschung= Forum: Quali-

tative Social Research, 7(4):art–11.

Reddy, K. J., Menon, K. R., and Thattil, A. (2018).

Academic stress and its sources among university

students. Biomedical and Pharmacology Journal,

11(1):531–537.

Richardson, T., Elliott, P., and Roberts, R. (2017). Relation-

ship between loneliness and mental health in students.

Journal of Public Mental Health, 16(2):48–54.

Rickwood, D. J., Deane, F. P., and Wilson, C. J. (2007).

When and how do young people seek professional

help for mental health problems? Medical Journal

of Australia, 187(S7):S35–S39.

Tarafdar, M., Pullins, E. B., and Ragu-Nathan, T. S. (2015).

Technostress: negative effect on performance and

possible mitigations. Information Systems Journal,

25(2):103–132.

Torous, J. and Roberts, L. W. (2018). Assessment of risk as-

sociated with digital and smartphone health research:

A new challenge for institutional review boards. Jour-

nal of Technology in Behavioral Science, 3:165–169.

Torous, J., Wisniewski, H., Liu, G., and Keshavan, M.

(2018). Mental health mobile phone app usage, con-

cerns, and beneﬁts among psychiatric outpatients:

comparative survey study. JMIR Mental Health,

5(4):e11715.

Tourangeau, R. and Yan, T. (2007). Sensitive questions in

surveys. Psychological Bulletin, 133(5):859.

Venkatesh, B. T., Andrews, T., Mayya, S. S., Singh, M. M.,

and Parsekar, S. S. (2015). Perception of stigma to-

ward mental illness in south india. Journal of Family

Medicine and Primary Care, 4(3):449–453.

Design Thinking and a Human-Centered Approach to Explore the Potential of Mobile Phone and AI-Enabled Just-in-Time Mental Health

Solution for University Students in India

519