The Feasibility and Acceptability of AI-Based eGuide for Healthcare Centers in Oman

Abstract

The rapid advancement of artificial intelligence (AI) in healthcare delivery has introduced innovative tools to improve patient care, streamline administrative processes, and bridge accessibility gaps. This study assesses how end-users perceive the practicality and usability of a proposed AI-enabled eGuide within Omani healthcare facilities, addressing cultural, linguistic, and regulatory requirements unique to the Sultanate. Through a mixed-methods framework combining stakeholder analysis, technological readiness assessment, and socio-cultural adaptation strategies, the research identifies the operational, economic, and ethical viability of the system. The current research results suggest that regulatory alignment, stakeholder engagement, and proper localization of AI-based eGuides will significantly enhance patient navigation after being tested on a wider dataset or real-world healthcare environments, reduce healthcare delivery bottlenecks, and increase patient satisfaction. Furthermore, digital literacy disparities, data privacy compliance, and infrastructure variability challenges need to be planned strategically and handled with care. This study offers a roadmap for policymakers and healthcare administrators to adopt AI-enabled eGuide systems that are both technically feasible and socially acceptable within the Omani healthcare ecosystem.

1. Introduction

With the advancements of digital transformation, Artificial Intelligence (AI) and Machine learning integration are playing key roles in driving digital transformation by integrating clinical and administrative workflows in healthcare centers around the world [1]. AI is the creation of machines that perform tasks requiring human intelligence. AI-based decision support [2,3], patient triage [4], and virtual guidance systems [5] have demonstrated potential in improving access to healthcare, reducing waiting times, and enhancing patient engagement [6]. In Oman, healthcare centers operate within a mixed public–private framework regulated by the Ministry of Health (MoH) [7]. While the country has invested in telemedicine [8], electronic health records (EHR) [9], and hospital management systems [10], there remains a significant gap in patient-centered digital navigation tools, particularly in Arabic–English bilingual environments. An AI-based centralized eGuide could address this gap by offering automated, culturally aware, and linguistically adapted healthcare navigation support to patients. The current study investigates the acceptability and feasibility of the mobile application implementation for the centralized and AI-based eGuide for healthcare centers in Oman for the purpose of technical readiness assessment and perceptions of stakeholders by aligning the regulatory frameworks. The rapid digital transformation of the healthcare sector is reshaping patient engagement, service delivery, and clinical decision-making worldwide. Artificial Intelligence (AI)-driven technologies, particularly conversational agents and e-guides, are emerging as transformative tools to address gaps in accessibility, efficiency, and patient-centered care [11,12]. AI-based e-guides, intelligent systems, and patient-facing digital assistants have been increasingly adopted in developed healthcare systems to streamline administrative tasks, improve patient education, and enhance clinical triage. Their potential to reduce waiting times, optimize healthcare resource allocation, and deliver personalized support makes them highly relevant to the evolving demands of healthcare in the Middle East [13,14].

Oman, under its Vision 2040, has prioritized healthcare digitalization as a national objective to achieve equitable, accessible, and high-quality care. The Ministry of Health (MoH) has already implemented integrated hospital information systems (Al-Shifa) [15] and launched the Shifa mobile application [16], which offers various services related to personal health records [15,16]. Building upon these digital foundations, an AI-based e-guide represents the next frontier in enhancing patient experience while alleviating the burden on healthcare providers. The availability of nationwide 5G connectivity and secure data infrastructures further strengthens the feasibility of deploying such intelligent systems in Oman [17].

Globally, evidence suggests that AI-powered healthcare assistants are generally acceptable to patients when designed with user-centric features such as ease of use, reliability, and cultural/linguistic localization [18,19]. Studies indicate positive outcomes in chronic disease self-management, mental health support, and administrative efficiency, with users expressing trust and satisfaction when interventions are transparent and contextually adapted [20,21]. Similar initiatives have been taken in the GCC region and neighboring country of Oman, Saudi Arabia where the Sehhaty system has been implemented and the other similar initiative has been taken by the UAE’s government is the implementation of the AI health assistants for the upgrading of the level of patient adoption are the encouragement demonstration and these initiative have paved the way for Oman to implement its tailored solution [12].

Nevertheless, for ensuring ethical compliance, some challenges may arise, especially in ethical compliance, safeguarding personal data, and addressing variations in digital literacy among populations [13]. Oman’s Personal Data Protection Law, stated in Royal Decree 6/2022, establishes a robust legal framework that emphasizes the need for lawful processing, security, and informed consent in AI-driven health applications [14]. Moreover, alignment with international standards such as WHO’s guidance on AI ethics and ISO/TS 82304-2 for digital health applications can further reinforce trust and regulatory compliance [12,15]. Given these circumstances, the purpose of this research is to investigate user perceptions toward adopting an AI-supported e-guide, focusing on its ease of use, practicality, and readiness for implementation in Omani healthcare settings. Such an evaluation not only informs technical and organizational readiness but also addresses ethical, socio-cultural, and regulatory considerations. This study aims to examine how AI-driven e-guides can integrate into Oman’s healthcare ecosystem, evaluate their acceptance among diverse patient groups, and highlight pathways for sustainable and culturally aligned digital health.

In Oman, primary healthcare centers experience substantial patient influx and administrative workload, often leading to longer waiting times, repeated informational queries, and pressure on front-desk staff. These challenges align with national digital transformation priorities under Oman Vision 2040, which highlight the importance of smart healthcare technologies and patient-centric service improvements. Motivated by these needs, this study investigates the feasibility and acceptability of an AI-based eGuide system developed specifically for healthcare centers in Oman.

Unlike general chatbots or survey-based studies, this work provides a scientifically grounded methodology by integrating technical system design, expert-validated query formulation, and empirical usability testing. The findings not only establish the system’s practicality within real healthcare workflows but also contribute to the broader scientific understanding of AI adoption and human–AI interaction in healthcare settings

As this work presents an early stage feasibility prototype that does not handle personal data, security and PDPL-compliance mechanisms are noted only as future deployment requirements rather than components of the current system. This study focuses exclusively on quantitative feasibility and acceptability assessment; qualitative exploration is planned as a separate future study.

2. Literature Review

AI-based e-guides in which patient-facing virtual assistants combining conversational AI with knowledge bases and clinical integrations are categorized under the WHO’s digital-health interventions portfolio. The World Health Organization (WHO) emphasizes that feasibility and acceptability should be assessed in terms of benefits, risks, safety, equity, and ethical concerns [12,19]. Oman’s Vision 2040 prioritizes world-class healthcare, digitalization, and equitable access. The Ministry of Health (MoH) has invested in the national HIS (Al-Shifa) [16] and the Shifa patient app, with strong 5G infrastructure enabling real-time services [17,22]. It is important to clarify that the proposed eGuide prototype is not intended to replace or compete with existing Ministry of Health platforms such as the Shifa mobile app or Al-Shifa HIS. This study focuses solely on evaluating the feasibility of an independent AI-driven informational prototype, and therefore, a formal comparative analysis with national systems is outside the scope of this work.

AI chatbots are now widely deployed for navigation, self-management, and administrative tasks. Recent systematic reviews highlight benefits in accessibility and engagement, but also concerns in diagnostic reliability [20,23]. Triage and symptom-checking tools have variable accuracy, sometimes underperforming expert clinicians, necessitating careful scoping of AI use [14].

The two models that are considered central and well-understood technology adoption frameworks are UTAUT and the technology acceptance frameworks. Most of the researchers have proposed effort expectancy, risk perception, trust, and performance expectations as some of the strong prediction variables, especially in the context of patient adoption [18,19]. Studies on older adults demonstrate that carefully designed chatbots can support autonomy and self-care [24]. In Middle Eastern contexts, patient satisfaction and trust are pivotal for continued use [19]. Arabic natural language processing (NLP) challenges remain, but domain-adapted large language models show promising results for healthcare-specific queries [19,25]. A critical and important acceptability mediator is the Digital health literacy [26].

Telemedicine has remained one of the top priorities for countries like Oman where the authorities have established technical and regulatory foundations. Telepsychiatry studies, telephone-based consultation usability evaluations, and clinical guidelines illustrating feasibility [22,27,28,29]. Oman’s counterparts, such as Saudi Arabia-based studies present e-services and portals uptake in patients [30,31].

The 5G services connectivity of Oman has ensured reliable communication with low latency [32]. The integration pathways can be established with the Al-Shifa app and Al-Shifa [16], which would be aligned with the standards of HL7FHIR [33]. Data protection is governed by Oman’s Personal Data Protection Law (Royal Decree 6/2022), with compliance extended to 2026 [34]. WHO, OECD, and ISO/TS 82304-2 standards provide ethical and technical frameworks [19,35].

The use cases of education and administration, including FAQs, reminders, scheduling, and appointments, are suggested as a priority while restricting unsupervised triage due to safety risks [26,28,36]. Arabic-localized interfaces, bilingual support, and voice-based services improve accessibility and trust [24]. Rigorous frameworks such as RE-AIM, CFIR, and NASSS provide multidimensional evaluation of AI e-guide interventions [26,37]. There is a need for Oman-specific evidence on digital literacy, cultural acceptability, and longitudinal adoption. Further, strengthening Arabic NLP and ensuring compliance with PDPL will be critical for sustainable deployment.

3. Methods

3.1. Survey Design

A cross-sectional survey [38] has been designed to evaluate the feasibility [39,40,41] and acceptability [42,43,44,45] of the mobile application of AI AI-based centralized eGuide for healthcare centers in Oman. This has been performed by administering a survey individuals using a developed application in Muscat city of Sultanate of Oman.

3.2. Setting

Oman is one of the Gulf Cooperation Council (GCC) member countries of the six council countries: Saudi Arabia, Bahrain, Qatar, Kuwait, and the United Arab Emirates [46]. Oman is located on the southeastern coast of the Arabian Peninsula, strategically situated at the mouth of the Persian Gulf [47]. It shares borders with the United Arab Emirates to the northwest, Saudi Arabia to the west, and Yemen to the southwest. The Sultanate has a long coastline bordering the Arabian Sea and the Gulf of Oman, which has historically facilitated its role as a maritime trading power [48,49]. The country’s topography is diverse, featuring vast, barren desert plains in the interior, rugged mountain ranges such as the Al Hajar Mountains in the north, and a fertile coastal plain known as Salalah in the southern Dhofar region, which is influenced by a monsoon climate [50].

3.3. Study Population and Sampling Strategy

The study participants have been recruited from the students, academic staff, and administrative staff of various colleges and universities. For inclusion criteria, we have selected smartphone/tablet owners and able to speak and understand Arabic and English. The various disciplines (background of participants) were selected, focusing on the computer science-related disciplines and others. A total of 426 participants have been contacted and were included in the samples from Muscat governorate of the Sultanate of Oman. Due to unavailability, non-cooperation, and incomplete participation, researchers have been forced to exclude 30 participants, and hence, the remaining (n = 396) participants constituted the final sample size of the current study.

3.4. Inclusion Criteria

The participants were contacted by the researcher and requested to participate, and their consent was obtained before obtaining their responses. The participants who agreed to participate were included in the study if they possessed a mobile phone or tablet, a signed consent form, and agreed that they could understand Arabic and English well. The app was installed on their devices, and they were asked app-related questions.

3.5. Exclusion Criteria

Participants who refused to sign a form were excluded. Furthermore, some of the participants were also excluded due to inconsistent timing, non-cooperation, and incomplete information. In this study, a total of 30 participants were excluded.

3.6. Measurements

The measurement instrument used to evaluate the feasibility and acceptability of the AI-based eGuide system was adapted from previously validated tools in health technology assessment and user-acceptance research, including studies referenced in [51,52,53]. The adaptation process followed a structured procedure to ensure conceptual relevance, cultural appropriateness, and clarity within the Omani healthcare context. The participants were requested to provide demographic information, including their age, gender, education level, and other parameters. The other additional parameters were also assessed using the application. The perception of the user about the eGuide application, experiences such as likeability [54,55], and the usefulness during the use of the application were also part of information gathering [56]. For the questionnaire, a five-point Likert scale [57,58] has been used for the measurement of the questions. A value of plus one (+1) has been allocated for the Strongly Disagree, and the value of plus five (+5) has been allocated for the Strongly Agree. The sum of all questions was scored, and after analysis, it was converted to a percentage. A threshold or cutoff point has been set for 70%, which means the score is low according to less acceptable or it is less feasible, whereas the score of more than 70% indicates a high level of acceptability and highly feasible [59]. For the engagement, also called acceptability and likability, a total of ten items were measured by employing a Likert scale. Five questions were measured with reversed scoring, and analysis for the acceptability was performed by merging all the questions so that a total score of 50 could be obtained, which becomes 100%. The same rule has been applied for the threshold of 70%. The same procedure has been applied for feasibility. A pilot study has been performed on 55 individuals, where each subject had to complete the survey, and then each participant was discussed for every item for the sake of clarity and understanding, so that the expected intention of the data collection and interpretation can be obtained. The overall (general) Cronbach alpha [59] for the acceptability sample obtained was 0.822, and for the feasibility obtained, it was 0.891.

3.6.1. Questionnaire Adaptation Process

The questionnaire items were initially reviewed and mapped against the objectives of this study. Items related to perceived usefulness, ease of use, clarity of system responses, trust in AI, and overall acceptability were retained. Several questions were modified linguistically to align with terminologies familiar within Omani healthcare centers, such as clinic timings, referral, and service navigation. To ensure cultural and contextual suitability, the adapted items were reviewed independently by two healthcare professionals, one expert in digital health, and two AI/NLP specialists. The feedback of the experts has mainly contributed to the enhancement of clarity, alignment, and relevance, along with the information needs a patient.

3.6.2. Pilot Testing for Face and Content Validity

As for the clarity and relevance of the questionnaire items, fifty-five individuals were selected for the pilot study. For this purpose, the participants were requested to identify any ambiguity or unnecessary complexity. The unfamiliar terminology identification was also requested to find out if any. The pilot focused on face validity (do items appear appropriate?) and content validity (do items adequately cover the intended domains?). Feedback from the pilot resulted in minor revisions to item wording, response options, and formatting to improve accessibility.

3.6.3. Reliability Assessment (Internal Consistency)

Internal consistency reliability was evaluated using Cronbach’s alpha, which demonstrated acceptable to excellent levels across the main constructs, such as the feasibility scale: α = 0.822 and acceptability scale: α = 0.891. These values exceed the commonly accepted threshold of 0.70, indicating sufficient internal consistency for a feasibility-oriented exploratory study.

3.6.4. Limitations of Current Psychometric Validation

Because the purpose of this study was a preliminary feasibility assessment, the sample size and scope were not sufficient for conducting advanced psychometric analyses, such as exploratory factor analysis, confirmatory factor analysis, and convergent validity testing, such as AVE values and discriminant validity testing, such as Fornell–Larcker criterion. These procedures require larger and more diverse datasets. The absence of such tests is now acknowledged both here and in the Limitations section.

3.6.5. Acceptability Threshold Justification

A threshold of 70% agreement was used to classify feasibility and acceptability outcomes. This value is consistent with thresholds widely applied in telehealth feasibility studies, technology acceptance evaluations, and early stage digital health usability assessments.

Citations have been added to support the use of this benchmark; however, we emphasize that such thresholds are not universal, and future research may refine empirically grounded cut-off values based on larger, more representative samples. While full factor analysis was not feasible at this stage, we conducted an item-level descriptive review.

3.7. Procedure

A total of 396 participants from academic staff, students, and general staff received and installed the centralized eGuide for the healthcare centers in Oman. The participants were asked to use this app during their daily life and report after four to six weeks. The data of installation and other parameters about the participants were recorded. The eGuide application is designed to be used in emergencies, normal life/daily routine, doctor’s appointments, hospital searching, electronic patient records, blood bank searches, reservation systems, and other health information is made available as it pertains to the closed area of the Sultanate of Oman. The various facilities are given for the participants to use and comment on the application. The overall structure of the app is given in Figure 1. The various screenshots are also illustrated in Figure 2.

System Testing Duration and User Interaction Procedure

It is important to clarify that the short interviews referenced during the prototype refinement stage were not part of a formal qualitative data collection process. These informal conversations were conducted solely to assess clarity, usability, and navigation flow during early system development. They were not designed for systematic qualitative analysis or reporting.

The AI-based eGuide system was introduced to participants for a short-term feasibility evaluation over a period of four to six weeks. To assess immediate usability was the main purpose of this phase. The purpose of such a time period for selection was selected by the researchers intentionally so that the objectives of the early stage prototype assessment can be achieved, rather than achieving the long-term adoption of measuring behavioral change for the sake of the system’s clarity and the initial perception of respondents for the acceptability of the system.

The participants were engaged in a dialogue and were instructed on the eGuide interaction during the testing period. The participants were instructed for a set of predefined tasks, simulating the healthcare inquiries, various types of information retrieving, service availability, clinic or doctor timing, locations of blood banks, various referral steps, and other navigational needs. These tasks were designed to approximate real-world use cases; however, it is acknowledged that participants were university affiliates, such as students, academic staff, and administrative personnel, and they were not interacting with the system under genuine healthcare needs or clinical conditions. The interactions of the users reflect simulated rather than organic or real-time behavior for information seeking.

A controlled environment has been used for the system deployment, and preliminary feasibility, including the interaction frequency, query logs, and session duration, has not been recorded systematically. The absence of detailed usage analytics is recognized as a limitation, as it restricts the ability to evaluate user engagement patterns or compare intended versus actual usage behaviors.

Furthermore, because the system was not deployed within a functioning healthcare center, factors that typically influence real-world use, such as patient stress, waiting times, environmental distractions, or urgent care needs, were not present. As a result, the ecological validity of the study is limited, and the conclusions should be interpreted as initial insights into usability and acceptability, rather than definitive evidence of long-term adoption or real-world performance.

The eGuide is given to users to check the application’s various aspects and information available in text, animation, graphics, animations, videos, maps, and tabular information. Efficacy and other parameters were asked to use and judge these parameters so that they can provide honest opinions [60] about the centralized eGuide of healthcare centers. The eGuide application uses a lightweight NLP-based intent recognition and rule-based matching mechanism to map user queries to hospital and blood bank information in the knowledge base, and the application has both offline and online in which the application is capable of saving internet data size. After the specified time, the users were asked to fill out the questionnaire with the help of data collectors and other supporting assistants. The data was collected through questionnaire administration and the online forms created in Microsoft Forms. Some of the participants were personally contacted, and their data was collected.

The AI-based eGuide system was developed as an independent research prototype designed to evaluate the feasibility of conversational support for basic healthcare information needs. The system does not integrate with any Ministry of Health hospital systems or national digital platforms; instead, it operates as a standalone application optimized for early stage usability testing. The system can run on 4G/5G networks, but the present study did not measure or evaluate network performance characteristics.

3.8. Query Formulation and Expert Validation

For the checking and assurance of eGuide system’s queries as technically feasible and clinically meaningful, a well-structured and multi-step process has been employed. For this step, the medical professionals and technical experts were consulted for an evaluation of a brief pilot research.

3.8.1. Initial Identification of Common Patient Queries

The service utilization reports review from the Ministry of Health of Oman has been used for the initial query pool. The other services were also studied in this regard, are the primary healthcare center’s data on patient flow, logs of the digital services, frequently asked questions, and other documents of customer portals. This mechanism helped researchers to understand various activities, like service navigation, operational timings, booking, searching for a doctor, appointments, referral processes, and other common and general information needs for healthcare.

3.8.2. Medical Expert Consultation

For the assurance of clinical accuracy and relevance, the set of queries was reviewed and verified by the medical practitioners, such as a healthcare administrator and two general practitioners from the local healthcare centers. The clinical phrasing of the questions was the result of the feedback from these experts and ensured actual patient communication patterns, the elimination of medical inappropriateness, and helped to discard the irrelevant queries. This also helped align queries and the reflection of the real-world information needs. The validation of the experts also helped to categorize service support groups and clinical information.

3.8.3. Technical Expert Evaluation

After medical validation, the refined query set was evaluated by two AI and health-informatics experts. The expert’s evaluation also helped to identify the compatibility of the queries with NLP processing, along with the structure and clarity of the queries and questions. The redundant and overlapping queries have been merged or removed in order to standardize the AI eGuide system with technical standards. The standardization of linguistic patterns in queries and the optimal performance of the chatbot were guided by the technical team.

3.8.4. Final Validation and Pilot Testing

For the final validation of the experts’ reviews and guidance, a small pilot study has been conducted to evaluate user comprehension and confirm that the queries selected are easy to understand and reflect the real-world environment, along with real-world issues. After the small pilot study, minor refinements have been performed on the wording.

3.8.5. Final Query Set

A balanced mix of the queries related to operations, healthcare information, administration balance, and questions on service navigation was used to create the final set of queries. The validation from the experts and approach based on the experts’ reviews ensured that the clinically grounded queries are technically implementable and that these queries are aligned with the real-world user needs, resulting in the relevance and reliability of the current study’s findings.

3.9. Data Analysis

All statistical analyses were conducted using IBM SPSS Statistics (Version 29). An AI-based centralized eGuide for healthcare centers in Oman has been tested and checked for descriptive statistics. Various statistical tests were also conducted to confirm the application’s feasibility and accessibility. The level of significance has been achieved at 0.05 by obtaining the 95% confidence level.

The purpose of the analysis was to evaluate demographic differences in the acceptability and feasibility of the AI-based eGuide system and to determine whether user perceptions varied significantly across age, gender, and education level

3.9.1. Data Preparation and Transformation

Responses collected using a 5-point Likert scale ranging from strongly agree to strongly disagree were transformed into binary categories to facilitate categorical analysis, such as Likert response strongly agree is equal to highly acceptable/feasible to strongly disagree to less acceptable/feasible. This dichotomization approach is consistent with prior research in digital health feasibility and technology acceptance evaluations.

3.9.2. Descriptive Statistics

Descriptive statistics (frequencies, percentages, and 95% confidence intervals) were generated for demographic variables, acceptability categories, and feasibility categories.

3.9.3. Inferential Analysis

To assess associations between demographic variables and binary feasibility/acceptability outcomes, Pearson’s Chi-Square tests (χ²) were applied separately for age groups, gender, and education levels. For each chi-square test, χ² statistic, degrees of freedom, p-value, and Cramer’s V effect size are reported. Because multiple chi-square tests were conducted, the Bonferroni correction was applied to control for Type I error inflation. The adjusted significance threshold was set at:

$${\alpha }_{\mathrm{adjusted}}={\displaystyle \frac{0.05}{ \mathrm{number} \mathrm{of} \mathrm{tests} }}.$$

3.9.4. Reliability and Validity

Internal consistency for the feasibility and acceptability items was evaluated using Cronbach’s alpha. Confidence intervals for proportions were calculated using the Wilson method.

3.10. Technical Architecture

It is important to clarify that the eGuide prototype used in this study does not collect, transmit, or store any personal or clinical data. All queries submitted during testing were anonymous and not linked to individual identities. Because the system operates solely as a text-based research prototype without backend storage of user information, the implementation of clinical-grade encryption, secure authentication, or regulated data-handling mechanisms is not applicable at this stage. These measures will be incorporated only when the system progresses to later development phases involving real patient data or integration with the Ministry of Health digital environments.

The interface used in this study represents a fixed-layout prototype designed solely for feasibility testing. Cross-platform compatibility, responsive layouts, and detailed UI/UX engineering are planned for future stages of development and were not part of the present study’s scope.

4. Results and Discussion

For the purpose of assessing the acceptability and the feasibility of the eGuide for healthcare centers in Oman, a response rate of the participants has been achieved as 93% (396 out of 426), as shown in

Table 1. Demographic details of the respondents.

Gender	Frequency	Cumulative Percentage
Male	285	71.97
Female	111	28.03
Total	396	100%
Marital status	Frequency	Cumulative percentage
Married	101	25.50
Single	295	74.50
Total	396	100%
Age group	Frequency	Cumulative percentage
20–29 years	172	43.43
30–39 years	108	27.27
40–49 years	102	25.76
50 years and above	14	03.54
Total	396	100
Education	Frequency	Cumulative percentage
Intermediate or less	00	00.00
Bachelor (14 years)	269	67.93
Master (16 years)	112	28.28
MBA/M.Phil/PhD	15	03.79
Total	396	100%

. Male participants have been dominant, as the percentage was nearly 72%, while the female participants were 28%. The ratio of the married respondents has remained nearly a quarter, whereas most of the participants were married. A total of 172 respondents were between twenty to twenty-nine years of age, whereas only 14 participants were above 50 years old and above. The reason for the zero respondents as the only university students were our target; in this case, only university students were contacted data were collected. The demographic details of respondents are given below in Table 1.

4.1. AI-Based eGuide App Acceptability

Most of the participants (nearly 75%) of the overall participants strongly agreed and recommended the eGuide app as good for use, as they explained it is not complex to use, and no further training or support is required for the usage of this app. It is expected that most users can use this application without any prior knowledge or training.

Table 2. The acceptability (usability) of the eGuide among the respondents.

Item	Strongly Disagree n (%)	Disagree n (%)	Neutral n (%)	Agree n (%)	Strongly Agree n (%)
I think that I would like to use this App frequently.	4 (1.11)	22 (5.56)	4 (1.11)	114 (28.89)	252 (63.33)
I found the App unnecessarily complex.	198 (50.00)	101 (25.56)	13 (3.33)	48 (12.22)	35 (8.89)
I thought the App was easy to use.	31 (7.78)	26 (6.67)	26 (6.67)	141 (35.56)	171 (43.33)
I think that I would need the support of a technical person to use it.	202 (51.11)	110 (27.78)	17 (4.44)	31 (7.78)	35 (8.89)
I found the various functions in this App well integrated.	9 (2.22)	0 (0.00)	36 (10.00)	167 (42.22)	181 (45.56)
I thought there was too much inconsistency in this App.	158 (40.00)	154 (38.89)	35 (8.89)	35 (8.89)	13 (3.33)
I imagine that most people would learn to use this App very quickly.	13 (3.33)	31 (7.78)	17 (4.44)	132 (33.33)	202 (51.11)
I found the App very cumbersome to use.	180 (45.56)	145 (36.67)	13 (3.33)	35 (8.89)	22 (5.56)
I felt very confident using the App.	26 (6.67)	9 (2.22)	26 (7.78)	149 (37.78)	181 (45.56)
I needed to learn a lot of things before I could get going with it.	150 (37.78)	127 (32.22)	17 (4.44)	57 (14.44)	44 (11.11)

explains the acceptability of the application among the respondents.

As per the collected data, the usability assessment of the AI-based eGuide system demonstrates that the application was generally well received among participants, with a strong majority reporting favorable responses across multiple usability indicators. Notably, over 90% of respondents either “agreed” or “strongly agreed” that they would like to use the application frequently, indicating a high potential for adoption [58] and integration into daily healthcare workflows. This finding aligns with previous studies emphasizing that perceived ease of use and usefulness are critical determinants of technology acceptance in healthcare [60,61]. A positive response has been received from the participants, and more than 78% responses were reported as the application is easy to use. This has given evidence that the AI-based eGuide is built to standards and has achieved an intuitive tool for digital health.

A high level of acceptance has been achieved in the overall discussion and the context of the user confidence [62], as well as their willingness to adopt the technology [63], along with functional integration [64]. Nonetheless, concerns related to technical support requirements and consistency should be addressed through ongoing training, user support, and iterative design improvements [65].

4.2. AI-Based eGuide App Feasibility

The eGuide is a new platform for the people of Oman, and most of the participants strongly agree that the app is very useful in normal situations as well as in emergencies.

Table 3. Frequency and Percentage Distribution on the Feasibility of eGuide apps.

Item	Strongly Disagree n (%)	Disagree n (%)	Neutral n (%)	Agree n (%)	Strongly Agree n (%)
I like the eGuide video, which teaches me more about healthcare practices.	0 (0.00)	0 (0.00)	9 (2.22)	174 (37.78)	213 (60.00)
I like the eGuide pointing out the critical areas in healthcare delivery that require extra attention.	0 (0.00)	0 (0.00)	22 (5.56)	207 (52.22)	167 (42.22)
I like the eGuide showing the areas in patient care that require extra effort in the workflow chart.	0 (0.00)	0 (0.00)	31 (7.78)	207 (52.22)	158 (40.00)
I like the eGuide reminding me to use additional healthcare tools (e.g., checklists, guidelines) to improve care.	22 (5.56)	4 (1.11)	71 (20.00)	145 (36.67)	145 (36.67)
I like the eGuide user manual; it makes the system easy to use.	36 (10.00)	18 (4.44)	101 (25.56)	92 (23.33)	145 (36.67)
I like the eGuide rewarding system that acknowledges successful completion of a healthcare task.	71 (20.00)	13 (3.33)	40 (10.00)	145 (36.67)	107 (30.00)

illustrates the responses of the subjects regarding the feasibility of the app.

The feasibility assessment of the AI-based eGuide system focused on participants’ perceptions of the practicality and added value of its features. Results reveal strong endorsement for the eGuide’s capacity to deliver educational and supportive functions in healthcare practice. For example, 97% of respondents agreed or strongly agreed that the instructional videos or a slight introduction enhanced their knowledge of healthcare practices [66], highlighting the platform’s ability to serve as an effective educational tool [67]. This indicates that the AI-based eGuide provides actionable insights that enhance clinical decision-making and workflow efficiency. These types of features comply with and are also aligned with the recommendations of the World Health Organization’s [12,19,22] for the digital decision support systems, which emphasize the overall guidance and evidence-based practices for the help and guidance of the healthcare workers. Prompts for additional healthcare tools or best practices, also called the reminder function, were also positively rated by more than 73% of participants. This suggests that the AI-based eGuide could contribute significantly to reinforcing compliance with clinical protocols, which is particularly valuable in busy healthcare environments where staff may overlook certain steps. Similarly, the user manual and integrated support materials were considered useful, with nearly 60% of respondents affirming that the system was easy to navigate due to these aids. For the reduction of the adoption barriers, this reinforces the instructional support and requirement for the built-in training [68].

Interestingly, while most responses were favorable, a proportion of participants expressed neutrality or disagreement, particularly regarding the system’s reward feature. Approximately 30% of respondents strongly agreed that positive reinforcement (such as acknowledgements for successful task completion) was valuable, while 20% strongly disagreed. The importance of gamification strategies in healthcare settings is also suggesting the variability [69]. Many of the professionals have positive and negative reactions. Some professionals understand the technology as positive motivation, whereas others perceive it as unnecessary and not relevant to the clinical texts. This shows the rewarding mechanism for the professionals, especially in Oman’s healthcare sector. The overall feasibility results show that the AI-based eGuide can support healthcare delivery and well well-positioned in the context of the targeted feedback, education, and workflow schedules.

5. Demographic Properties vs. Acceptability and Feasibility

For the improvement of efficiency, accessibility, and patient engagement, mobile health applications are typically integrated into healthcare delivery [69,70]. The acceptability is the major factor for the successful adoption of these new technologies, where the users are expecting and willing to adopt any new technologies or apps [42,43,44,45]. The other estimated factor is the feasibility and practicality in real-world contexts and apps [39,40,41,45]. Demographic characteristics, including age, gender, occupation, and level of education are the main influences for the feasibility and the acceptability, and this is also emphasized by the literature. These associations will help policymakers to tailor interventions for their diverse populations by ensuring equity in implementing digital health.

5.1. Age as a Determinant of Acceptability and Feasibility

Age is one of the most consistent predictors of mobile app usage patterns. Younger adults and adolescents often demonstrate higher levels of acceptability, perceiving health apps as intuitive and aligned with their everyday use of digital technology [71,72,73,74,75]. By contrast, older adults may encounter barriers such as reduced digital literacy, vision impairments, or resistance to technology adoption. These barriers can negatively influence feasibility, as older users may require additional technical support, training, or age-sensitive design features [76]. A substantial proportion of the workforce of the health care professionals in the Sultanate of Oman includes older age professionals and middle-aged categories. The usability in older demographics can be enhanced by implementing simplified navigation, structured orientation, and larger size of the fonts in the AI-based eGuide system for healthcare centers in the Sultanate of Oman.

5.2. Gender Differences in Mobile App Engagement

The acceptability is also affected by gender, and gender plays an important role in the acceptability of the eGuide implementation. Women are generally considered as more proactive in health-seeking behaviors and may perceive mHealth apps as valuable tools for self-care, preventive health, and caregiving roles [77,78,79,80]. On the other hand, men may demonstrate lower engagement unless the app directly relates to performance, fitness, or disease management. Feasibility challenges may also emerge if gender-specific health needs are overlooked in app design [81,82,83,84]. For instance, in the Omani context, where cultural norms shape gendered access to healthcare resources, ensuring that AI-based eGuide systems address both men’s and women’s unique health concerns is crucial for equitable adoption.

5.3. Educational Attainment and Digital Literacy

For the purpose of acceptability and feasibility, the level of education plays a vital role as a highly qualified user is more likely to perceive the digital platforms or digital health tools as more useful, as something they can use easily, and as something they can easily integrate with the other activities in their daily routine [85,86,87,88]. On the other hand, those with lower education levels may encounter challenges in understanding technical jargon, interpreting health data, or troubleshooting errors. Digital literacy, closely linked with education, further moderates’ feasibility. Without adequate digital literacy, even technically sound apps may face low adoption rates. The policymakers of the Sultanate of Oman must ensure that simple instructions are included, multilingual support is available, and culturally adapted manuals are available to bridge the educational disparities.

5.4. Occupation and Professional Role

Occupational background and the professional role are also linked with the feasibility of the mobile apps. The medical professional and healthcare specialist who are already trained know a technology background then can adopt AI-based eGuide more easily. [89,90,91,92]. Administrative staff or community health workers, however, may require more comprehensive training before they can integrate such tools effectively. In Oman’s healthcare centers, feasibility could be enhanced by offering differentiated training modules based on job roles. The decision support functions will definitely help the clinicians while performing advanced decisions, whereas the non-clinical staff will get simplified instructions and the guides available for the daily workflow applications.

5.5. Cultural and Contextual Factors

Cultural norms strongly shape perceptions of digital health. In some societies, health applications are readily embraced as symbols of modernization, while in others, skepticism persists due to trust issues, privacy concerns, or unfamiliarity with AI. In Oman, cultural expectations surrounding healthcare delivery, such as high regard for face-to-face consultation and reliance on hierarchical medical authority, may influence app acceptability. The local infrastructure is also a factor for the feasibility, including the availability of the internet in remote areas and an interface that works in Arabic languages. A well-designed application may have limited potential if these contextual factors are ignored.

5.6. Intersections of Demographic Characteristics

The demographic factors are important, and they may not act in isolation. With limited education and qualifications, an older adult having less experience will compound the barriers, resulting in a reduced feasibility and acceptability level. At the same time, a person who is young and educated may show a higher level of acceptability, but they may not agree if the cultural values are missing or are not reflected in the application. In this case, to avoid exacerbating health inequalities, various interventions must be considered in relation to intersectionality.

5.7. Implications for AI-Based eGuide Systems in Oman

Keeping in view the feasibility assessment of the eGuide systems, the findings suggest that the overall acceptability has proven high in user experiences. More strategies for digital learning can be incorporated to enhance the targeted program learning, especially for individuals who need older staff. Another strategy would be incorporated for the gender sensitive features, which may reflect the health priorities for both genders. The other main factor is the Omani culture-based contents that may align with the language preference and healthcare practice in the Sultanate of Oman. Equitable infrastructure is the last strategic investment for ensuring a high score for the feasibility in underserved, rural, and remote areas.

By implementing the above strategies, the sustainability of Oman’s digital health interventions can be improved, and the adoption can be improved. With the help of the association between the characteristics of the population and mobile app acceptability, the importance can be inferred for the digital solutions to be provided to the population. The demographic characteristics like gender, age, occupation, and education will play an important role in practical engagement in digital technologies and shaping perceptions, especially in the cultural context of the country like Oman.

6. Overall Acceptability and Feasibility Against Demographic Characteristics

The different age groups and dimensions of feasibility and acceptability help to analyze the perceptions and the intervention policy. The p-values suggested a statistical test (likely a chi-square test) was used to determine if the observed differences between age groups are statistically significant. Table 4 illustrates the AI-based eGuide’s overall acceptability and feasibility against the demographic characteristics for the age group.

Table 4. AI-based eGuide’s overall acceptability and feasibility among the participants by demographic characteristics (age group) (N = 396).

Age Group	Less Acceptable n (%)	Highly Acceptable n (%)	p Value	Less Feasible n (%)	Highly Feasible n (%)	p Value
20–29 years	15 (23.43)	49 (76.57)	0.415	21 (23.86)	67 (76.14)	0.013
30–39 years	22 (17.89)	101 (82.11)		14 (10.85)	115 (89.14)
40–49 years	8 (7.69)	96 (92.31)		16 (18.0)	73 (82.00)
50–57 years	9 (9.38)	87 (90.62)		21 (25.93)	60 (74.07)
58 and above	0 (0.00)	9 (100.00)		0 (0.00)	9 (100.00)
Total	54 (13.64)	342 (86.36)		72 (18.18)	324 (81.82)

Table 4. AI-based eGuide’s overall acceptability and feasibility among the participants by demographic characteristics (age group) (N = 396).

Age Group	Less Acceptable n (%)	Highly Acceptable n (%)	p Value	Less Feasible n (%)	Highly Feasible n (%)	p Value
20–29 years	15 (23.43)	49 (76.57)	0.415	21 (23.86)	67 (76.14)	0.013
30–39 years	22 (17.89)	101 (82.11)		14 (10.85)	115 (89.14)
40–49 years	8 (7.69)	96 (92.31)		16 (18.0)	73 (82.00)
50–57 years	9 (9.38)	87 (90.62)		21 (25.93)	60 (74.07)
58 and above	0 (0.00)	9 (100.00)		0 (0.00)	9 (100.00)
Total	54 (13.64)	342 (86.36)		72 (18.18)	324 (81.82)

6.1. Explanation and Discussion of Findings (By Age)

6.1.1. Interpretation of Key Findings for Accessibility

The vast majority of respondents across all age groups found the intervention Highly Acceptable (86.36% overall). The differences between age groups, while visually present (e.g., 90.62% in the 50–57 group vs. 100% in the 58+ group), are not statistically significant (p = 0.415). This means we cannot confidently say that age influences the perceived acceptability of the intervention, and the variation seen could be due to random chance. The high overall acceptability is a strong positive indicator (which is also achieved in our eGuide for health centers of Oman) for the intervention’s potential uptake, as acceptability is a key predictor of implementation success [51].

6.1.2. Interpretation of Key Findings for Feasibility

Feasibility perceptions have proven to be more changing and varying as compared to the acceptability by age group (p = 0.013). The interventions seem to be highly feasible among most of the groups, and the group of 50 to 57 years is completely realized as an outlier. The other age group, such as 30 to 39, proved as highly feasible with 89.14% and the last group, with 100% cannot be perceived as acute.

6.1.3. Discussion and Implications

It can be observed and the critically insightful the divergence between feasibility and acceptability. The agreement of the people or population in the intervention’s goals can be called highly acceptable, or they may have some doubts about the practical execution, which can be estimated as variable feasibility. A common challenge has been considered for the implementation of science [52]. A low feasibility perception among the group of 50 to 57 years of age requires further investigation, along with the indicators that this age group has unique characteristics, which must be investigated, such as the peak career demands, the age pressure, children’s education, mortgages, and others. This age group has another different aspect, which must be investigated carefully is the caring for their aging parents, and this phenomenon is also called sometimes the sandwich generation [93].

6.1.4. Implications for Implementation

The policy makers and the program designers of the Sultanate of Oman must consider the following strategies or steps for the efficient implementation to ensure a successful rollout, so that more people can benefit from the implementation of the AI-based eGuide. A qualitative research focus group of people between 50 and 57 years old needs to be conducted of as this group, as this group had various perception barriers in the feasibility study such as cost, required skills, and others. Another strategy is the tailoring of targeted strategies for the mitigation of the identified barriers to this group. These mitigations may involve the availability of flexible scheduling. Financial benefits and subsidies, simplification of the participation steps, or any other benefits that can save time and money.

6.1.5. Limitations

One of the groups is the 58+ years of age, and the data for this group is not generalizable, and feasibility findings are unstable as the sample size of this group is too small, resulting in 100% feasibility findings, which can be understood as a data limitation. Furthermore, the data does not provide context on the nature of the intervention itself, which is crucial for a fully nuanced interpretation.

6.2. Explanation and Discussion of Findings (By Gender)

The provided data analyzes differences in the perception of an intervention between male and female respondents along two critical dimensions for implementation science: acceptability and feasibility. Table 5 illustrates the AI-based eGuide’s overall acceptability and feasibility against the demographic characteristics by gender.

Table 5. AI-based eGuide overall acceptability and feasibility among the participants by gender (N = 396).

Gender	Less Acceptable n (%)	Highly Acceptable n (%)	p Value	Less Feasible n (%)	Highly Feasible n (%)	p Value
Male	39 (18.93)	167 (81.07)	0.126	35 (17.07)	170 (82.93)	0.623
Female	13 (6.84)	177 (93.16)		26 (13.61)	165 (86.39)
Total	52 (13.13)	344 (86.87)		61 (15.40)	335 (84.60)

Table 5. AI-based eGuide overall acceptability and feasibility among the participants by gender (N = 396).

Gender	Less Acceptable n (%)	Highly Acceptable n (%)	p Value	Less Feasible n (%)	Highly Feasible n (%)	p Value
Male	39 (18.93)	167 (81.07)	0.126	35 (17.07)	170 (82.93)	0.623
Female	13 (6.84)	177 (93.16)		26 (13.61)	165 (86.39)
Total	52 (13.13)	344 (86.87)		61 (15.40)	335 (84.60)

6.2.1. Interpretation of Key Findings for Accessibility

For acceptability, A large majority of both genders found the intervention Highly Acceptable (Overall: 86.87%). However, a noticeable, though not statistically significant (p = 0.126), difference exists between genders. As a matter of fact, female respondents reported higher acceptability, with 93.16% rating it as Highly Acceptable, whereas the male respondents were somewhat less likely to do so, with 81.07% giving a Highly Acceptable rating. This suggests a trend where women may be more receptive to the intervention’s concept or goals, but the difference is not large enough to be considered conclusive beyond chance variation in this sample.

6.2.2. Interpretation of Key Findings for Feasibility

For Feasibility, perceived feasibility was also overwhelmingly positive and showed no statistically significant difference by gender (p = 0.623). Both men (82.93%) and women (86.39%) largely viewed the intervention as Highly Feasible. The proportions of those who found it Less Feasible were also similar (17.07% for males vs. 13.61% for females). This indicates that men and women in this sample largely agree on the practical implementability of the intervention, perceiving similar levels of logistical, resource-based, or procedural barriers.

6.2.3. Discussion and Implications

The observed trend in acceptability (higher among females), even if not significant, aligns with a body of research suggesting that women often express more positive attitudes toward health-seeking behaviors and psychosocial interventions [94,95]. The fact that this trend did not extend to feasibility is insightful; it suggests that while women might be more receptive to the idea, both genders similarly evaluate its practical execution.

6.2.4. Implications for Implementation

To ensure successful rollout and its implementation, the Sultanate of Oman and the program designers must perform the following steps so that effective implementation is possible, and so an increasing number of people in Oman can benefit from this implementation. Broad, non-gender-specific communication strategies can be effectively employed since no significant gender gap exists; this will promote both the value (acceptability) and practicality (feasibility) during the process of the intervention. Another implication is the monitoring of the trend, which means that while it is not significant here, the acceptability trend should be monitored with a larger sample. If the trend holds, it might be beneficial to investigate the underlying reasons, whether they are related to the intervention’s content, marketing, or delivery method, to better engage male participants. The other necessary action is the focus on other demographics because the lack of gender-based differences suggests that implementers should focus their tailoring efforts on other demographic factors that showed significant variation (e.g., age, as often seen in other analyses) or on structural barriers that affect all genders equally.

Table 4, Table 5 and Table 6 indicate varying levels of digital literacy, age-related technology preferences, and differences in exposure to AI-based tools. To address these barriers, several actionable implementation strategies are proposed, including targeted digital literacy training where healthcare centers should provide structured and short digital literacy sessions for the elderly and low-literate Omani people. Another implementation strategy is digital assistants to help on-site support staff with training on the usage of the eGuide system. A step-by-step QR code-based tutorials and guidelines in both languages, Arabic and English, are also recommended. This will help with the mitigation of the usability gaps identified among older and less tech-savvy populations.

The eGuide must provide multilingual and culturally aligned content, which can support Arabic-speaking users, along with common expatriate language content, such as Urdu, Hindi, and Bengali. Culturally familiar terms, icons, and examples must be validated through community consultations.

To accommodate older age users, high contrast themes, larger font, and simplified paths must be implemented. In case those who do not read confidently, voice-enabled interactions must be made available. In addition to these activities, the healthcare centers must have steps for the awareness and adoption campaigns, in which the healthcare centers run on-site posters and digital screens for the introduction of eGuide systems. The healthcare center must run social media campaigns about the use of the system and the characteristics of the system to ensure unified, consistent communication targeted at a diverse audience.

6.2.5. Limitations

The main limitation is the inability to generalize the observed trend in acceptability due to its lack of statistical significance. Furthermore, the data treats gender as a binary (Male/Female) variable, which may not capture the full spectrum of gender identities and their potential unique perspectives. The context and nature of the intervention itself are also unknown, which is crucial for a deeper interpretation of these perceptions.

6.3. Explanation and Discussion of Findings (Education)

As the data reveal, the education of a particular individual has a statistically significant relationship with the feasibility and acceptability of the specific interventions. Table 6 illustrates the AI-based eGuide’s overall acceptability and feasibility against the demographic characteristic of level of education.

Table 6. AI-based eGuide overall acceptability and feasibility among the participants by gender (N = 396).

Level of Education	Less Acceptable n (%)	Highly Acceptable n (%)	p Value	Less Feasible n (%)	Highly Feasible n (%)	p Value
Less than high school	18 (81.82)	4 (18.18)	0.000	13 (59.1)	9 (40.9)	0.038
High school	26 (9.70)	242 (90.30)		40 (14.18)	242 (85.82)
College or university up to bachelor	8 (7.84)	94 (92.16)		8 (9.09)	80 (90.91)
College or university up to master/PhD	0 (0.00)	4 (100.0)		0 (0.00)	4 (100.0)
Total	52 (13.13)	344 (86.87)		61 (15.40)	335 (84.60)

Table 6. AI-based eGuide overall acceptability and feasibility among the participants by gender (N = 396).

Level of Education	Less Acceptable n (%)	Highly Acceptable n (%)	p Value	Less Feasible n (%)	Highly Feasible n (%)	p Value
Less than high school	18 (81.82)	4 (18.18)	0.000	13 (59.1)	9 (40.9)	0.038
High school	26 (9.70)	242 (90.30)		40 (14.18)	242 (85.82)
College or university up to bachelor	8 (7.84)	94 (92.16)		8 (9.09)	80 (90.91)
College or university up to master/PhD	0 (0.00)	4 (100.0)		0 (0.00)	4 (100.0)
Total	52 (13.13)	344 (86.87)		61 (15.40)	335 (84.60)

6.3.1. Interpretation of Key Findings for Accessibility

There is an extremely strong and statistically significant relationship between education level and acceptability (p < 0.001). The trend is clear and linear, such as Low Education, which means the group with Less than high school education stands out dramatically, with only 18.18% finding the intervention highly acceptable and an overwhelming 81.82% finding it less acceptable. This perception shifts radically with higher education; Acceptance rates jump to 90.30% for high school graduates, 92.16% for those with a bachelor’s degree, and reach 100% for those with a master’s or PhD. This indicates that the intervention, in its current form, is perceived very poorly by the least educated group but is almost universally accepted by those with more formal education.

6.3.2. Interpretation of Key Findings for Feasibility

A similar, though slightly less pronounced, statistically significant relationship exists for feasibility (p = 0.038). Again, the “Less than high school” group is the outlier, with only 40.9% finding the intervention highly feasible and 59.1% perceiving significant practical barriers. Perceived feasibility improves markedly with education level, with over 85.82% of high school graduates and 90.91% of college-educated individuals finding it highly feasible. The most educated group again reported 100% feasibility.

6.3.3. Discussion and Implications

The equity gap for the implementation has been highlighted as critical in findings. Education has played a favorable role, as the educated group perceived the technology as highly acceptable and feasible, whereas the population with a lower level of education had a lower perception of the technology. This can be called a major concern because this group (with lower level of education and lower perception) is also most targeted by social programs and is the most vulnerable in the public health system. This disparity can be explained by several theoretical frameworks. The intervention of health literacy, in which the concept of intervention literacy is crucial. Individuals with lower education may struggle to understand the intervention’s purpose, instructions, or benefits, leading to rejection [96,97]. The materials or communication strategy may be too complex. Another implication is the cognitive-affective barriers, which means the Lower acceptability and feasibility may stem from a history of negative experiences with similar systems, mistrust of authorities, or a perceived lack of self-efficacy to engage with the intervention successfully [21].

6.3.4. Implications for Implementation

To avoid exacerbating health inequities, a one-size-fits-all approach is untenable. The intervention requires urgent, targeted adaptation, such as co-designing with the community. Co-designing with the community means the intervention and its materials must be redesigned in partnership with representatives from the low-education demographic. This ensures cultural appropriateness, simplicity, and relevance [11]. Simplification of the communication is inevitable, which means all informational materials must be simplified using plain language, visual aids, and clear, step-by-step instructions to overcome literacy barriers.

Building trust and self-efficacy is another implication where the implementation of the eGuide should involve trusted community champions (not just clinical staff) to promote the intervention and provide direct, hands-on support to build participants’ confidence. Structural barriers must be addressed, meaning implementers must audit the intervention for hidden barriers (e.g., technology requirements, cost, timing, location) and create solutions (e.g., phone-based access, subsidies, childcare).

System-level challenges relate to infrastructure readiness, technology integration, staff capability, and organizational policy alignment. The following strategies address these concerns in actionable detail, including infrastructure requirements and deployment readiness, in which the healthcare center must ensure stable Wi-Fi connectivity and redundancy for eGuide kiosks. Tablet-based meetings must be held, and cloud services must be obtained. Use of FHIR/HL7-compliant APIs to retrieve patient information must be implemented where appropriate.

Concrete privacy measures are now included in the form of end-to-end encryption for user queries and system responses. Role-based access control for backend administrative dashboards and consent notices informing users about data usage and logging must be incorporated. This clarifies how privacy obligations will be operationalized and alignment with Oman’s Personal Data Protection Law and MoH cybersecurity guidelines.

For the training of the staff and organizational readiness, a structural training module is proposed in which training receptionists, nurses, and administrators are taught to use and promote the eGuide. The creation of the AI help desks is proposed to assist the users in implementation. Monitoring and evaluation are proposed in the form of a monthly or quarterly system performance evaluation that must be shared with IT and clinical stakeholders.

6.3.5. Limitations

The very small sample size in the extreme groups (Less than high school and Master/PhD) means the precise percentages (e.g., 80%, 100%) should be interpreted with caution, as they can be unstable. However, the strength and consistency of the trend across all levels strongly support the overall conclusion of a significant educational divide.

6.4. Implementation Roadmap

To ensure structured deployment of the eGuide system, a phased implementation roadmap is introduced. The roadmap given in Table 7 provides clear steps for real-world implementation.

Table 7. Implementation roadmap.

Phase	Activities	Stakeholders	Expected Outcome
Phase 1: Preparation (0–3 Months)	Infrastructure assessment, procurement of kiosks/tablets, API compatibility review	MoH IT, PHC Admin, Vendor	System deployment readiness
Phase 2: Prototype Integration (3–6 Months)	Integration with Al-Shifa HIS/Shifa app (limited features), pilot testing in 1–2 healthcare centers	MoH IT, Clinical Staff, Pilot Patients	Initial evaluation of interoperability
Phase 3: Community Co-Design & Training (6–9 Months)	Workshops with patients and caregivers; staff training modules; UI refinement	Community reps, Nurses, Admin Staff	Inclusive and user-centered refinement
Phase 4: Deployment (9–12 Months)	Full rollout in selected centers; placement of kiosks; communication campaigns	MoH Admin, IT Teams	Active system use and adoption
Phase 5: Monitoring & Optimization (12–18 Months)	Analytics collection, feedback loops, AI model updates, security reviews	IT Analysts, Clinical Leads	Improved performance and long-term sustainability

Table 7. Implementation roadmap.

Phase	Activities	Stakeholders	Expected Outcome
Phase 1: Preparation (0–3 Months)	Infrastructure assessment, procurement of kiosks/tablets, API compatibility review	MoH IT, PHC Admin, Vendor	System deployment readiness
Phase 2: Prototype Integration (3–6 Months)	Integration with Al-Shifa HIS/Shifa app (limited features), pilot testing in 1–2 healthcare centers	MoH IT, Clinical Staff, Pilot Patients	Initial evaluation of interoperability
Phase 3: Community Co-Design & Training (6–9 Months)	Workshops with patients and caregivers; staff training modules; UI refinement	Community reps, Nurses, Admin Staff	Inclusive and user-centered refinement
Phase 4: Deployment (9–12 Months)	Full rollout in selected centers; placement of kiosks; communication campaigns	MoH Admin, IT Teams	Active system use and adoption
Phase 5: Monitoring & Optimization (12–18 Months)	Analytics collection, feedback loops, AI model updates, security reviews	IT Analysts, Clinical Leads	Improved performance and long-term sustainability

7. Discussion

This study report summarizes the findings of the demographic analysis of the acceptability and feasibility of the AI-based eGuide system designed for healthcare centers in Oman. The system aims to streamline patient guidance, enhance decision support, and improve operational efficiency. A total of 396 participants were included, with demographic breakdowns across age, gender, and education.

Acceptability Findings: Overall, 86.87% of participants rated the eGuide system as highly acceptable. Across age groups, acceptability was consistently high, ranging from 90.62% (50 to 57 years) to 100% (58+ years). Women reported slightly higher acceptability (93.16%) than men (81.07%). Educational attainment strongly influenced acceptability, as participants with postgraduate education reported 100% acceptability, while those with less than high school education reported only 81.82% acceptability.

Interpretation: Acceptability is broadly strong, reflecting positive perceptions of the system. However, educational disparities indicate that lower-literacy users may feel excluded, necessitating targeted support.

Feasibility Findings: Overall, 81.82% of participants considered the system highly feasible for integration into their work. Feasibility varied significantly by age (p = 0.013). While younger groups (30 to 39 years: 89.14%) rated feasibility highly, middle-aged staff (50 to 57 years) struggled (25.93%). Gender differences in feasibility were minimal (men: 82.93%, women: 86.39%), with no statistical significance (p = 0.623). Education again played a decisive role (p = 0.038). Participants with bachelor’s or postgraduate degrees demonstrated very high feasibility (90–100%), whereas those with less than high school education reported only 40.9%.

Interpretation: Feasibility challenges are not equally distributed. Age-related barriers may reflect workload or resistance to workflow changes, while educational disparities highlight the role of digital literacy.

Key Implications for Implementation:

• Targeted Training Programs in which specific orientation for middle-aged professionals and those with limited education.
• User-Centered Design is required as simplified navigation, multilingual support, and visual guides tailored for diverse literacy levels.
• Policy Integration is in need of time as inclusive measures to ensure equitable adoption across demographics.
• Infrastructure Support is direly required to ensure adequate technical infrastructure, especially in rural or resource-limited settings.

7.1. Interpretation of Key Findings

Results of the current study display that the users perceived the AI-based eGuide as helpful, reliable, and intuitive health-related information. As per the reporting of the perception, the understanding of the various activities has been improved, including referral processes, clinic timings, and the areas where the patients are involved. The results indicate that the interactive system supported by AI can replace or help human staff in automating various repetitive information tasks, and high-volume data is involved. Participants’ positive attitudes toward interacting with AI suggest increasing readiness in Oman for integrating intelligent digital health tools into routine care. This aligns with global trends showing growing patient acceptance of chatbot-based solutions in healthcare administration and telehealth environments.

7.2. Significance of the Study

The demonstration of this study shows the significance of the AI-based eGuide which has a meaningful role in patient service delivery with the help of digital tools and technologies, especially in the cases of health care centers where the patient inquiries and administrative load are high. Empirical evidence has been presented for a healthcare context and is considered crucial and important, as simulation-based AI healthcare research has remained limited for countries with high income. The query identification and validation of the framework, with expert validation along with participation evaluations, have led this study to find out the gaps in the literature in the context of the healthcare service navigation, duly supported by the AI tools. This study shows AI-powered tools and scientific understanding can reduce waiting times, support operational efficiency, and improve patient satisfaction.

7.3. Contribution to the Scientific Community

Several contributions can be made from the current study for a broader scientific community, including the methodological contribution. The methodological contribution includes the development of a query, its validation, and the dual inputs of the experts from technical and medical backgrounds. This model can be employed as a reference model for the upcoming AI-based patient information systems.

The empirical contribution is understood as the other contribution, in which the study presented rare evidence from the Middle East, particularly the region of Oman, where very limited studies are available, especially pertaining to the AI-based guidance systems. These regional studies play a very important and vital role in the understanding of healthcare and AI-based healthcare in a global context. The findings of this study also have conceptual and theoretical contributions that reinforce that AI and human interaction display clarity of response, trust, and simplification of design, which may influence the acceptability for users in healthcare contexts.

7.4. Contribution to the Healthcare Community

The strong potential transferability can be seen in the methodology and framework, along with system design, even though the current study is based on the Omani healthcare centers. Several components of the study, such as the NLP-based query handling, expert-validated question set, and chatbot–patient interface, are universally applicable and can be adapted to healthcare settings in other countries. The contextual elements, such as alignment with Oman’s Ministry of Health workflows, can be replaced with local administrative processes in different regions. This makes the AI-based eGuide generalizable to hospitals in developing countries facing similar administrative bottlenecks, clinics with limited manpower, telehealth service platforms, and multilingual or multicultural healthcare environments where structured informational support is needed. In this case, the current study not only addresses the local and regional challenges but also provides a model which can be replicated for the global healthcare systems.

7.5. Contribution to Policy Making

The significance of this study also applies to healthcare organizations and policymakers. This study demonstrates that front desk staff load can be reduced and patient flow management can be improved, as well as accurate real-time response to common inquiries through the implementation of the AI-based eGuide systems. The patient satisfaction level can also be enhanced through the AI-based eGuide. The results suggest that the integration of AI tools with other national digital health strategies is also aligned with the Oman Vision of 2040.

7.6. Implications for Future Digital Health Systems

The current study lays a foundation and indicates the effectiveness of developing, designing, and deploying more advanced health care systems that are duly supported and assisted by AI and other advanced technologies. Some of the upcoming systems can be for the integration with the national health care record system, multilingual support, predictive analytics, and others. The strong user acceptability demonstrated here suggests that healthcare stakeholders are ready to adopt more sophisticated AI systems.

8. Conclusions

As of its first stage, the AI-based eGuide will possess characteristics that will be more feasible for healthcare professionals and other stakeholders in Oman, as the preliminary acceptability and feasibility study suggests. The absence of gender-related disparities suggested that most of the demographic challenges will be resolved through design improvements, training, and long-term strategies for the assurance of the successful and sustainable adoption into the Sultanate of Oman’s health care system. This study provides empirical evidence on the feasibility and acceptability of an AI-based eGuide system tailored for healthcare centers in Oman. With an overall high level of acceptance (86.87%) and feasibility (81.82%), the findings highlight the system’s potential to enhance patient navigation, improve service efficiency, and support Oman’s national vision for digital health transformation. The results confirm that gender differences are minimal, suggesting that equitable adoption across men and women is attainable. However, age and education emerged as critical determinants: middle-aged professionals expressed lower perceptions of feasibility, and individuals with lower educational attainment demonstrated markedly reduced levels of both acceptability and feasibility. These findings underscore the importance of implementing targeted strategies to address demographic disparities. Tailored training for middle-aged staff, simplified design features for users with limited literacy, and culturally adapted communication will be essential for equitable integration. Furthermore, sustained stakeholder engagement, regulatory alignment with Oman’s Personal Data Protection Law, and investment in digital infrastructure are crucial to mitigate barriers related to trust, privacy, and accessibility. While this study validates the promise of AI-enabled eGuides in strengthening healthcare delivery, it also reveals the risk of deepening health inequities if vulnerable groups are overlooked. Policymakers and system designers must therefore adopt a proactive, inclusive approach to ensure that digital transformation benefits the widest possible spectrum of users. Future research should expand beyond academic and healthcare staff to include patients from diverse socioeconomic backgrounds, conduct longitudinal assessments of sustained use, and explore advanced features such as personalized health recommendations through Arabic natural language processing.

In conclusion, the proposed AI-based eGuide system will be acceptable and feasible within the healthcare ecosystem of the Sultanate of Oman. By implementing the thoughtful adoption and real-world testing and validation, the proposed system will serve as a transformative tool that can improve healthcare acceptability and surely empower potential users, making this technology a meaningful contribution to Oman’s Vision 2040 goals of high-level and high-quality healthcare.

9. Limitations and Future Work

The current study presented valuable insights regarding the acceptability and feasibility of the AI-based eGuide system in the context of the Sultanate of Oman’s healthcare. Along with the insights, various limitations must be acknowledged regarding the current study. The following limitations will help to shape the findings and interpretations along with the identified directions where further research is required.

9.1. Limitations

Sampling bias can be understood as the primary limitation of this study, as the participants of this study were exclusively selected from the university environment, such as students, administrators, and academic staff. The population of this study is skewed towards the young and educated population, which does not represent the diverse population of patients or the individuals working in hospitals and healthcare centers. The limited representation of older adults (only 14 participants aged ≥ 50) and the exclusion of individuals under 20 restricts the ability to understand how various age groups, especially elderly patients and those with lower digital literacy, would interact with and perceive the eGuide system. A comparative evaluation with existing Ministry of Health platforms, such as Shifa App and Al-Shifa HIS, was not conducted because eGuide is an exploratory prototype developed independently for feasibility testing. Integration or comparison with national systems requires separate studies and MoH-level technical collaboration, which are planned for future research but are beyond the scope of the present work

Therefore, the findings should be interpreted as preliminary indicators of feasibility rather than generalizable outcomes for the broader patient population.

9.1.1. Prototype-Level Evaluation

The study evaluated an early prototype of the AI-based eGuide, focusing primarily on usability, response clarity, and participant acceptability. The system was not tested in real clinical environments where factors such as waiting time pressure, multilingual needs, patient stress levels, and environmental noise could influence user experience. The real-world performance of this technology may not be fully reflected, and its technological validity may not be fully established, due to the lack of real-time evaluation in this study

The advanced security features and encryption standards, such as authentication, audit logs, security messages, and document security, may not be fully understood as this phase of the prototype has not stored any personal or health-related confidential information at its current status. These requirements fall under the Personal Data Protection Law and will surely be addressed during the phases where real-world deployment will be prepared.

9.1.2. Limited Technical Depth in Live Deployment

Although the study included detailed system architecture, NLP workflow, and query-processing mechanism, the AI model was not deployed with live integration to healthcare information systems such as electronic health records and appointment management systems. Consequently, the current findings do not capture system behavior under operational load, data security constraints, or interoperability challenges.

9.1.3. Scope of Queries and Functional Coverage

The set of queries that were validated through the consultation of the experts can be understood as potential patient queries. The emergency-related queries, clinical questions, and interactions in various languages can be included in the upcoming phase and were not included in this phase, resulting in the narrowed scope of the current system’s functional assessment.

9.2. Future Work

The sampling issues will be addressed through future studies in which the participants will be recruited or selected from the original healthcare centers and hospitals of Oman. The participants will be elderly individuals with different demographic characteristics, patients suffering from chronic diseases, patients or people with varying educational backgrounds, and diverse populations. This will add a more generalizable and a more comprehensive understanding of the user acceptability around the healthcare system user spectrum. A critical next step involves deploying the eGuide system in actual healthcare settings to assess real-time interaction patterns, environmental influences, and operational constraints. Under realistic situations, user behavior can be investigated through real-time field testing, which will help to find out the real potential barriers.

9.2.1. Real-World Deployment and Field Testing

The actual healthcare setting is considered a critical step for the investigation of real-time interaction patterns, environmental factors, and real-time operational constraints. For understanding practical implementation barriers, field testing is unavoidable, and wider user behavior must be investigated under certain realistic conditions.

9.2.2. Technical Enhancements and System Integration

Subsequent development phases will focus on integrating the eGuide with electronic health records, enabling multi-language support such as Arabic–English, expanding the knowledge base to cover more complex queries, enhancing NLP capabilities for dialectal Arabic, and strengthening data privacy and cybersecurity measures aligned with healthcare regulations.

Robust data logging modules, real-world technical validation, and the cybersecurity controls will be incorporated once the proposed system passes beyond the prototype phase. Then, the system will be eligible for integration into real-world clinical environments. The proposed model has not been checked or evaluated for battery consumption or battery-based performance as this falls outside the scope of the feasibility study.

9.2.3. Longitudinal and Comparative Studies

A longitudinal design will be employed to evaluate the changes over time in user acceptance, which may occur due to the advancements of AI. Comparisons with other regional studies and country studies can establish the cross-cultural variations in the adoption of the AI eGuide activities and navigational tools.

9.2.4. Evaluation of Impact on Healthcare Workflows

The operational benefits provided by the eGuide system will be investigated, including the reduction of the administrative workload, reduction of patients’ waiting time, along patient satisfaction and service navigation. These assessments will help quantify the system’s effectiveness and guide decision-making for large-scale deployment.