Smart ICT MED, mHealth development to basic illness symptoms

Introduction

Mobile applications have long been a challenge for software developers, but their evolution over the decades has transformed the industry from an exclusive realm to a thriving commercial sector, attracting major hardware and software companies (1-3). In the early days, digital pager technology dominated the market until the emergence of mobile phones in the early 1990s, which significantly impacted pager sales (4). Nowadays, the use of smart mobile devices, such as tablets and cell phones, is steadily increasing, compelling mobile device manufacturers to continuously improve their hardware and innovate with advanced technologies (5,6).

Everyday, technology especially in the field of digital health, plays a vital role in addressing public health challenges worldwide. The use of technology, such as mobile applications, has become increasingly prevalent due to their ability to accept and verify essential information, ensuring efficient access to necessary tools (7,8). The utilization of technology databases and health resources is crucial in meeting the requirements of service providers and improving overall usage (9-11).

In Thailand, the accessibility of location-based internet services via mobile phones has significantly increased, as indicated by the Thai language’s usage statistics in the year 2022 (12). The widespread usage of mobile platforms such as iOS and Android emphasizes the significance of taking into account the distinct requirements associated with these platforms, which are widely accepted and utilized across various activities.

Additionally, the ability of mobile applications to provide healthcare services can have a positive impact by allowing individuals to manage their health conditions and seek appropriate care (13). However, challenges remain, such as the need to educate individuals to avoid relying solely on service providers and seek appropriate self-treatment knowledge (14). By adopting innovative systems like the Triagist mobile application system, which categorizes patients based on illness severity codes, it becomes possible to streamline the healthcare process (11).

When considering the readiness of the Thai population to use information technology, it becomes evident that they are highly prepared for its adoption. Therefore, implementing advancements in accessing medical through various models becomes more feasible. Developing a system that can be utilized on smartphones would greatly assist the majority of the population in accessing different systems and public health services, serving as a tool for self-care (self-medication) and providing access to various health information for prevention. This aligns with the current efforts of government agencies and regulatory organizations that aim to develop and implement applications to aid in accessing services and facilitate the public sector in almost every domain (13,15). Considering budget cost-effectiveness, the expenditure per individual access unit (node destitution) remains relatively low. Therefore, developing a system for use in any government mission is a viable approach.

Considering the attention given to medical services and the numerous inquiries that are often made, it is evident that there are challenges in terms of self-treatment. Many Thai individuals seek assistance from service providers without having the necessary knowledge to treat themselves (14). According to a report from institutions providing daily care services in various hospitals, over 60% of cases in emergency rooms are unnecessary visits (16), which impacts the quality of service and adds strain to hospitals, especially community hospitals, public hospitals, and major healthcare centers where high patient volumes are managed daily. This inefficient utilization of resources results in a waste of the healthcare budget, as many symptoms can be self-diagnosed.

Hence, this study establishes the foundation for developing a system to detect common, non-severe illnesses. The tool, envisioned as a mobile app, targets general users, particularly in Thailand, aiming to significantly impact healthcare and alleviate Thailand’s public health burden. The research addressed two main questions: (I) How can the application system be designed to ensure usability across all age groups and user-friendly navigation? (II) What functionalities should be incorporated into the system, and to what extent should disease sequence analysis content be integrated?

The development of medical diagnosis systems

The development of U-Health integrates ubiquitous IT into healthcare, transforming traditional hospital-based care into a universal, daily life value. It supports preventive health care by analyzing lifestyle, nutrition, and other factors beyond mere disease identification. Kwon introduced a systematic and intelligent medical diagnosis expert system designed to mimic real expert assistance. This system leverages a knowledge base and reasoning engine, employing rule-based and case-based reasoning to offer accurate health assessments and personalized care recommendations. Such systems are pivotal in advancing self-diagnosis and proactive health management, particularly in resource-constrained settings (17).

Moreover, the advent of mobile applications has significantly influenced various fields, including healthcare and education. Kim and Mun explore the design and development of a self-diagnostic mobile application for monitoring learning progress in non-face-to-face (non-F2F) practice learning environments. Utilizing the ADDIE model (Analysis, Design, Development, Implementation, and Evaluation), their study demonstrates how mobile applications can effectively support self-diagnosis and enhance learning outcomes in remote education settings. The usability evaluation of the developed mobile application revealed high satisfaction scores among users, underscoring the importance of such tools in maintaining engagement and ensuring the continuity of practical learning during the coronavirus disease 2019 (COVID-19) pandemic (18).

Using a system for assessing illness symptoms demonstrates the crucial role of information technology in facilitating self-treatment. This research highlights the feasibility of these systems in providing basic guidance for managing non-serious conditions, underscoring their potential to enhance self-care and promote health literacy. By empowering users to make informed health decisions, these technologies significantly contribute to overall well-being.

The increasing demand for mobile health has driven its rapid development. Healthcare authorities support mobile health to reduce treatment costs, improve advanced care through digitalization, and expand services to include people with disabilities, the elderly, socially isolated individuals, and the general public. Mobile health services can be categorized based on their purpose and application scope, as shown in Table 1.

Methods

Group symptoms for diagnosis

The development of a diagnostic system offering user guidance incorporates the use of guidelines primarily derived from the diagnosis of 54 symptom groups. These guidelines encompass common illnesses that can be assessed by considering symptoms. This step includes tasks related to processing the flow of symptoms. The diagnosis is performed in a specific order based on disease groups, as shown in Table 2 based on the Clinical Drug Information (26-29).

Diagnosis regulation

In this research, for analysis the patient condition which focus on gathering information related to the assessment of various types of illnesses and their treatment guidelines, organized by groups of symptoms. A sequence of symptom assessments is obtained for detailed study. The information has been collected from multiple reputable sources, including:

• Manual for the Rational Use of Medicines According to the National List of Essential Medicines (26);
• Wolters Kluwer Clinical Drug Information (27);
• A Handbook of Medical Diagnosis for Students (28);
• Symptom to Diagnosis: An Evidence-Based Guide (29).

Despite drawing on information from several reliable sources as mentioned, concerns may persist regarding the completeness of the current medical information and knowledge, highlighting the need for ongoing enhancement and verification from diverse sources.

Importantly, the syndromes addressed by the system are common and exhibit straightforward or uncomplicated symptom patterns, falling outside the realm of complex symptom analysis. Consequently, the analysis based on the data sources provided is deemed adequate for the basic diseases referenced.

Selection expert

The criteria for expert selection, potential biases, and their involvement level in the validation process were determined through purposive sampling. We targeted general doctors, pharmacists, and nurses with over 10 years of experience in hospitals. In the initial round, we identified 12 doctors, 10 pharmacists, and seven nurses from university hospitals, pharmacy faculties, and general ward nursing services. Subsequently, we randomly selected individuals from each group to serve as experts. After extending invitations, four doctors, five pharmacists, and five nurses agreed to participate.

Research operation to system development

The development of the Smart ICT MED mobile application was driven by the objective of addressing the challenges encountered in the pre-hospital process, specifically in the domain of self-diagnosis. This development adopts an experimental research method (specifically the experimental research design) in combination with the development cycle of the Waterfall Methodology. The designed application follows an Evolutionary Model and is developed according to the following stages (30).

Step 1: collecting data on various types of symptom assessment and treatment guidelines for different symptom categories

During this stage, data will be collected on symptom groups and assessment criteria to develop an adapted and applied symptomatic assessment. Wolters Kluwer Clinical Drug Information, A Handbook of Medical Diagnosis for Students, Symptom to Diagnosis: An Evidence-Based Guide, and the expertise of Tanpichat et al. (26-29) were utilized to gather information on correct patient assessment and advice on the initial treatment of symptoms. A team of experts from the Faculty of Medicine, Nursing, and Pharmacy will provide input and validate the sequence of initial treatment and drug group symptoms for initial relief. Planning analysis will be conducted to design the assessment, which will involve developing questions about symptoms based on various sources such as information from textbooks, academic books, and input from experts to ensure assessment accuracy. The results of the symptom assessment will be provided to the user, and the evaluation order will be established to ensure accuracy.

Step 2: analysis the functional operations of the application

In this step, the analysis diagram for designing the symptom discrimination assessment, which was outlined in the previous step, is utilized. The application’s operation is planned by designing a draft with specifications for various functions that benefit users. It should be accessible to users of all ages and compatible with their devices. The draft design is then reviewed by system experts for verification. Additionally, the results obtained from real-world problem conditions and opinion surveys are analyzed to further inform the design process.

Step 3: designing the user interface and system expectation

Implementing the front-end design system to enhance user experience and usability. This involves ensuring the application is accessible to users of all group. Conducting interviews and using open-ended questions to identify characteristics and functions that are suitable and easy to use. The design is tailored to meet specific requirements and incorporates appropriate graphic design across all dimensions. Testing with users is conducted to measure performance and assess the system’s usability across different functions and components. Thus, this stage looks at the difficulties and important aspects of making mobile application interfaces, especially focusing on how people interact with computers (31,32).

To ensure the design meets the actual needs of the users, an evaluation form will be used at this stage to gather information on their requirements and purposes for using the system. The collected results will be analyzed and utilized as a guideline for future system development.

Step 4: application development

System development is guided by system requirements derived from design studies. The user interface prioritizes user experience and usability (31,32), accommodating users of all ages. The development process adheres to disease analysis as outlined in relevant guidelines and professional expertise. Initial prototypes are created, reviewed, and refined in collaboration with medical professionals to enhance functionality and perfect the software model (30).

Step 5: testing

After obtaining version one of the application, testing is conducted to evaluate the accuracy of data, diagnostic flows, and other functions such as data collection for the local area and user data management stored in the backend system. User feedback and application prototypes are utilized to gather feedback from experts and a sample of real users, and heuristic evaluation and usability were employed. The target users, randomly selected from the population who will be using the system, provide their opinions and express their needs. The results of the analysis from the collected feedback are summarized and used to make improvements to the application.

Step 6: implementation

The modified application will undergo another round of technical testing to ensure its functionality. Once validated, it will be made available to the general users. The application can be accessed through the iOS and Android application stores, allowing users from hospitals nationwide to test and utilize it in real-world scenarios.

Diagnostic process

The diagnostic process employs standard programming techniques to ensure that the logic and flow of symptom analysis are consistent and reliable. This involves creating algorithms and decision trees that guide the diagnosis based on user-reported symptoms.

The diagnostic process integrates with standard programming practices through expert flow analysis, ensuring consistency and reliability in the evaluations. The step of analysis patient condition based on the 54 symptom groups with the analysis regulation (26-29). The application employs this logic to diagnose based on the decision process for all symptom groups. The example of algorithm for fever diagnosis decision tree technique was computed into the Eq. [1] as:

D(s)={f(s)D(T(c,y,n))ifcurrentnodeisleafotherwise[1]

Here D is the DiagnoseFever function, s are the symptoms, f is the model, and T (c, y, n) represents traversing the decision tree based on the condition c and child nodes y and n.

Thus, the equation means that the DiagnoseFever function will traverse the decision tree based on the symptoms (s) provided by the user. The tree is structured to ask specific questions (conditions) and make decisions based on the user’s answers, leading to a final diagnosis.

Meanwhile, the user interaction for asking questions and getting responses is represented into the Eq. [2] as a function:

Q:C→{"yes","no"}[2]

Q represents the function that asks questions. C represents the set of all possible conditions or questions. The function returns either “yes” or “no” based on the user’s response. This means that for each condition or question (C) posed by the application, the user provides a response of either “yes” or “no” These responses guide the diagnostic process through the decision tree, ultimately leading to a diagnosis or recommendation to each symptom group.

Validation of decision/diagnosis accuracy

To ensure the accuracy and reliability of the application’s diagnostic decisions, the application was validated by cross-referencing its outputs with established Diagnosis Regulations (26-29). This process involved medical doctors comparing the Smart ICT application’s diagnostic results with these regulations. The analysis flow of the algorithm was verified as correct for all 54 symptom groups. This process is critical to building trust in the application and ensuring it delivers reliable health assessments to its users.

System architecture

The Smart ICT MED application architecture diagram, Figure 1 illustrates the interrelationships between users and medical professionals as they interact with the centralized health information server through various authentication methods. Users can log into social media to access personal health services on their smart devices via their smartphones. Meanwhile, specialists and administrators utilize specific credentials to manage records related to diagnostic sequence management and to prepare preliminary recommendations, which can be increased, decreased, improved, and modified. The central server is responsible for recording and processing related information to operate the system, ensuring a streamlined health management experience, achieving maximum efficiency, and not causing harm to users.

Figure 1 System architecture.

Statistical analysis

The study utilized quantitative methods for its statistical analysis, focusing on data derived from the 54 grouped symptoms which were used to assess the application’s diagnostic performance. The development and validation process involved experimental design, with data collected through both systematic testing and user feedback. Quantitative analysis was primarily applied to measure the effectiveness of the application’s features, such as symptom assessment accuracy, user satisfaction, and function utilization. These metrics were gathered from a wide user base and reviewed to evaluate the accuracy and reliability of the application.

While predominantly quantitative, the research also integrated qualitative feedback through usability testing and expert interviews. This approach enabled the study to refine the application’s interface and functionalities, enhancing the user experience. However, the main focus remained on quantitative data to validate the symptom assessment flow and ensure reliability across different symptom categories. This balance allowed for a robust analysis that addressed both statistical validity and practical usability, ensuring the application met its intended healthcare objectives.

Ethical statement

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). In this research, ethical approval was granted by the Chiang Mai University Research Ethical Committee, which issued an ethical exemption (No. CMUREC 64/005). Informed consent was obtained from all participants prior to their involvement in the study.

Results

Result of collecting data on various types of symptom assessment and treatment guidelines for different symptom categories

The outcome of this step involves 54 general disease groups, primarily focused on common illnesses suitable for symptom assessment. Tasks related to the flow sequence of assessment for each symptom group are performed. These symptom groups are presented in Table 2 and encompass a total of 54 basic symptom groups (see Table 2).

During the assessment sequence analysis, an assessment design is developed based on the synthesis of empirical evidence and knowledge gathered from various sources, including information searches, textbooks, and academic books (26-29). The developed design undergoes scrutiny by six experts specializing in medicine and pharmacology to ensure safety, suitability, and accuracy of the assessment sequence. Their opinions and reviews contribute to achieving the correct assessment of symptom groups. Once the assessment sequence is established, the results of the symptom assessment can be provided to the user, ensuring accurate and reliable assessments for each symptom group.

Result of analysis of the functional operations of the application

Based on the findings of the expert investigation, it was determined that the designed assessment sequence was appropriate, taking into account expert consensus. While the assessment sequence is a preliminary symptom assessment (screening) and may not include detailed questions, some questions may require medical test results. Consequently, certain messages related to requesting measurement results using specific tools may be omitted.

Furthermore, the experts emphasized the need to adjust the text to avoid being overly academic, except for terms with specific meanings in the medical context. The language should be adapted for real-world use by users who may not possess medical knowledge. Therefore, the questions are presented in a logical order and have been refined and adjusted accordingly to ensure efficient utilization.

In summary, the amendments made to the assessment sequence include:

• Avoiding overly academic language or terminology;
• Adjusting the order of questions that are difficult to understand or may be misinterpreted;
• Removing messages or questions that require results from specific tools not available to users;
• Rearranging the order of questions to consider variables or factors;
• Adding preliminary instructions that cover the initial treatment of symptoms;
• Incorporating drug information for certain symptom groups (e.g., back pain, toothache, itchy scalp, abdominal pain, diarrhea, fever, etc.), emphasizing that over-the-counter options are available.

Result of designing the user interface and system expectation

To ensure the developed system meets users’ needs, respondents were invited to participate in the project. The invitation was announced, and the recruitment process took 1 month, during which interested individuals were randomly selected. A total of 94 evaluation forms were received, both online and offline, providing brief yet valuable information. At this stage, the results of the assessment design analysis scheme for discriminating illnesses, summarized in the previous step, were obtained. These functionalities are expected to be developed. Data collection from a user group has been conducted through questionnaires and experiences related to using mobile applications as a tool for providing advice when feeling unwell. The data analysis includes both online and offline assessments.

The result shows that the application operation is planned by designing a draft with specifications for various functions that benefit users of all ages, ensuring compatibility with their devices. The design is reviewed by system experts, and the results from surveys and opinion interviews conducted under real problem conditions are analyzed. The researchers have outlined a preliminary plan for the system requirements, including the following predictive functions:

• Preliminary symptom assessment function;
• Advice function;
• Self-monitoring conditions;
• Display function for pharmacy maps and nearby hospitals.

Demographic a respondent

Table 3 presents the findings from the 94 respondents who evaluated their expectations regarding system design and user interfaces.

System function expectations

Table 4 shows the anticipated proportions of system functions. A majority of respondents agreed that the nearby hospital search function offers easy access. Additionally, it is expected that the developed system will include a referral function to hospitals to reduce waiting times. In Thailand, accessing public health services often involves significant waiting periods.

Furthermore, the function for notifying or monitoring a patient’s illness indicates a need for reporting user medical information to document events or symptoms during illness. However, the function for sharing personal health information with other applications remains controversial, as personal information is considered sensitive and should not be widely shared.

Conversely, the evaluators highly approve functions related to forwarding illness information to other applications, including both general applications and those used by government agencies.

Interface expectation

To evaluate the performance of the graphic user interface, heuristic evaluation and usability principles were applied and revised for clearer communication with respondents. Table 5 shows the proportion results of interface expectations of respondents. All respondents agreed that the system should prioritize reliability over beauty. Most of the respondents agreed that the system should focus on convenient design over screen beauty, font size and number of buttons on the screen affect usability, the system should use messages that are easier to understand than graphics, and the background color affects the readability or interpretation of a text. However, most of the respondents disagreed that the system should prioritize pictures over text, the system should have beautiful graphics accompanied by text, and the system should be colorful, with lots of colors.

The system design illustrates the sequence of operations from initialization to implementation across relevant components, including instructions, subscription access, and other IDs. The user interface design is available on Figma and can be accessed at https://www.figma.com/file/alDxrq3A6bLlgy9Apl40f6/Medicine-UI-App?node-id=0%3A1.

The example showcases a hierarchy of disease assessment with five symptomatic examples (provided to demonstrate the flow and order of user interface alignment). The user interface layout and sequence for other syndromes are not shown in this report, as the focus is on the designed flow. The internal display does not include any graphical representation; it solely presents the text from the flow.

The results obtained from designing functions and user interfaces for Android and iOS, which have a similar display format, differ only in the language used for development. The results were obtained by collecting user requirements, considering the flow of use, order of symptom inquiry, and standard development techniques.

Result of application development

The medical diagnosis application was designed to assist apprehensive individuals who, due to fear or the feeling of wasting time by visiting a doctor when experiencing health problems, prefer to search the internet for potential causes and find a diagnosis. To further simplify matters, Smart ICT MED offers its users, free of charge, a presumptive medical diagnosis service guided by their answers to various questions. The responses are “yes” and “no”. Finally, to make sure that the diagnosis is as accurate as possible (Figures 2-4). Meanwhile, Figure 5 shows the back-end interface used for adding and modifying analysis criteria and providing initial suggestions, all connected to the application as shown in Figure 2, which serves to evaluate and provide advice.

Figure 2 An example showing the usage to ask for symptoms, bring information to evaluate and give advice.

Figure 3 User interface related to consent form and for collecting user information.

Figure 4 Using the emergency function for calling the relevant authorities and symptom assessment results and the use of functions to connect to the map and display the information shared to the Line application.

Figure 5 The emergency function back-end screen (A) is utilized for managing the diagnostic flow, which involves adding, modifying analysis criteria, and providing initial suggestions to the user (B,C).

Application features

Assessment function by condition, function properties after modification:

• Show the sequence of questions to ask the user’s symptoms. and can summarize the answers or important points.
• There is a menu to search by keyword. and pictures according to the symptoms of illness (to make it easier to find).
• Link the assessment results to the back-end system to save the data (user name, gender, age, results, recommendations, symptom status, symptom monitoring).
• In the event that the assessment results are considered urgent for treatment Allows the system to present a nearby phone number for assistance (urgent cases such as tel. 1669, etc.).
• The decision-making structure shall be based on the guidelines for the analysis of symptom sequences from working professionals. Can increase symptoms that may be caused by emerging diseases in the future through the back-end system.
• Show map, hospital, or advice about medical facilities as an alternative to users. By showing the results as a nearby map and displayed by the nearest distance by showing the distance.
• There is a back-end where the flow sequence can be adjusted or the rule of symptom assessment can be modified for future improvement. Which was carried out in accordance with the advice of medical experts and information system specialists.

Result of testing

The results of this step involve testing the system, covering both technical and functional aspects, and including user testing. Participants were invited to test and verify the correctness of their application. System development was based on system requirements derived from design study results. The user interface was designed with behavioral and usability evaluations in mind, leading to the creation of Prototype version 1. This prototype was then presented to medical experts for review, adjustments, and efficiency improvements until it reached a comprehensive condition.

To test the user interface design, the questions were updated based on usability criteria to reflect the expectations outlined in Table 6. The goal of this update was to verify whether the final design met these expectations. However, only 70 participants from the initial group responded to the questions and provided feedback. The evaluation questions were based on Ben Shneiderman’s 8 Golden Rules of Interface Design, consisting of 8 questions rated on a scale from 1 to 5 (with 1 being strongly disagree and 5 being strongly agree).

Potential challenges that might be encountered during implementation, measures for data security, and methods for monitoring the application’s effectiveness and safety after deployment were considered. Every scenario was meticulously planned and tested until all errors or oversights were addressed. A reevaluation of the analysis flow against reference sources was conducted.

Additionally, the functional design of the system was evaluated using a manual testing approach again. In the final phase before user release, experts confirmed that the system includes the necessary features to support various disease groups. Testing revealed that the system primarily follows simple guidelines for symptom assessment across 54 symptom groups, focusing on common conditions identifiable by their symptoms. It strictly adhered to guidelines for analyzing symptom sequences sourced from all relevant data (26-29).

To ensure proper usage and error prevention among system users, the application is connected to a back-end system. This back-end system can check inquiry data for symptom analysis and display results with recommendations. This thorough initial review helps confirm that the system is functioning as intended (Back-end link: https://smart-med.system-on.cloud/beta-test-v1d/authen).

Result of implementation

Dissemination

In response to the challenge of influencing healthcare and innovating new healthcare resources, the promotion and dissemination of the system’s utilization have led to a notable increase in its visibility and recognition on social media platforms. The system’s usage has led to a notable increase in its visibility and recognition. Upon releasing the system on the application store and promoting it through social media channels, particularly on the Facebook page “Smart ICT MED”, the response was overwhelmingly positive. This can be attributed to a substantial number of people being exposed to information about the system and the project. This proactive approach is effective in extending the system’s usage among healthcare workers and the general public, with the Facebook page boasting around 600 followers since its official launch in May 2021 (Figure 6) while Figure 7 shows visitor numbers surged during the second wave of the COVID-19 outbreak.

Figure 6 Data reflecting the awareness campaign for the Smart ICT MED application on Facebook offers intriguing insights (https://www.facebook.com/smartictmed). No permission is required for the publication of this image.

Figure 7 The Facebook page reached over 181,356 visitors within a span of 4 months, from July 2021 to October 2021 (screen display retrieves from the Facebook page). No permission is required for the publication of this image.

The project’s content and system utilization were promoted through three short video clips, which collectively garnered over 132,000 views. This greatly increased health awareness and interest in the system. The audience analysis showed that the content resonated most with individuals aged 25–34 years, particularly those interested in health and hygiene, with more males than females engaging with the content.

These statistics confirm widespread awareness and usage of the system for primary healthcare. The data reveals substantial user engagement across various regions, supporting the project’s goal of providing accessible healthcare information. This demonstrates the success of using social media to raise health awareness, reducing the reliance on hospitals for information. These findings clearly show that users can easily access the project’s outputs.

Chiang Mai University also supported this research by promoting it through their media channels, which have over 100,000 followers, mostly students and members of the public. This broad outreach effectively conveyed self-care guidelines and raised awareness about primary health issues.

Application publication

Regarding downloads, the Google Play Store registered 875 users for the system from 2021 until its termination in 2022 which was due to new publisher regulations, after which it was resubmitted. Additionally, distribution through the Google Play Store allowed the system to reach 711 user accounts directly through download links from January 2021 to October 26, 2021 (Figure 8).

Figure 8 Application release to Google play and Direct link install via APK download. No permission is required for the publication of this image. APK, Android package kit.

As for the Apple Play Store, statistics indicate 5,819 users downloaded the system from June 2021 to October 26, 2021 (Figure 9).

Figure 9 The report over 5,819 installations during 4 months (June 2021 to October 26, 2021). No permission is required for the publication of this image.

A version update was released on August 1, 2021, under the account of the Faculty of Pharmacy, Chiang Mai University. The update aimed to present the system as a health service, necessitating publication solely through a health service agency. However, initial attempts to publish through the Apple Play Store encountered several challenges, resulting in a delay of over 3 months. Following Apple’s approval, the system was officially launched, without participating in the Beta version, in July (Figure 10).

Figure 10 The report over 11,944 user who enroll into back-end system of application installation via all application release channel (count number of all installation) into the back-end section. No permission is required for the publication of this image.

Nationwide dissemination

Efforts to disseminate information about the system’s usage have reached 87 target hospitals nationwide, including primary and secondary institutions, through the distribution of brochures demonstrating its operation and via invitations. Furthermore, Figure 11 demonstrates how users on social media have shared information across various online publications, such as the Khaosod newspaper and the Vanguard Newspaper at Chiang Mai University, which has a following of over 100,000, with the goal of raising awareness and creating an impact [links: (I) https://www.khaosod.co.th/pr-news/news_6922107; (II) https://www.naewna.com/relation/639304; (III) https://www.khaosod.co.th/pr-news/news_6922107; (IV) https://www.matichon.co.th/publicize/news_3214940; and (V) https://thaiinnovation.center/2022/03/smart-ict-med/].

Figure 11 Disseminate information over social media, flyer and newspaper (in Thai). No permission is required for the publication of this image.

Results of application use

Based on the system usage report, which analyzes the connection between the application and the back-end system, noteworthy data was gathered from 11,033 users. The findings, detailed in Table 7, and corresponded to Figure 12. The analysis of symptom function usage within a healthcare application indicates distinct patterns in user engagement, highlighting which functions are frequently or infrequently used in response to various symptoms. For the symptom of fever, the “Share Symptom Condition” function was used 1,778 times, indicating a high level of engagement among users who wish to communicate and share their condition with others. This is further evidenced by the “Find Hospital” function, which was used 1,828 times, suggesting that users experiencing fever are actively seeking medical facilities. Additionally, the “Use the Monitoring Function” was utilized 277 times, showing a moderate but significant engagement in monitoring health status. Similarly, for the symptom of headache, the “Share Symptom Condition” function was used 629 times, while the “Use the Monitoring Function” was accessed 559 times. This indicates that users with headaches also tend to share their condition and monitor their symptoms frequently, though not as intensively as those with fever.

Figure 12 Comparing the percentage of male and female cases for each symptom and the usage of different application functions by symptoms corresponded to the Table 7.

In contrast, the analysis shows that the “Call Emergency” function is rarely used across all symptoms. For headache, it was used only four times, and for fever, just five times. The symptom of cough saw minimal engagement with the “Call Emergency” function, used only once, and the “Find Hospital” function, used 87 times. For sore throat symptom, the “Call Emergency” function was not used at all, and the “Share Symptom Condition” function was used 72 times.

Similarly, for runny nose, the “Call Emergency” function was also not used, and the “Use the Monitoring Function” was engaged only 99 times. The symptom of fever stands out with the highest overall engagement across all functions, particularly in sharing symptom conditions and finding hospitals. This suggests that fever is perceived as a more severe symptom, warranting more communication and search for medical assistance.

On the other hand, the “Call Emergency” function is the least utilized across all symptoms, indicating that users may prefer other means of emergency contact or that emergencies are relatively rare in the context of the analyzed symptoms. The “Use the Monitoring Function” and “Share Symptom Condition Function” are more frequently used for symptoms like headache and fever, underscoring the importance of these functions for symptoms perceived as serious or requiring ongoing monitoring. The much higher engagement for symptoms like fever compared to runny nose or sore throat could indicate the perceived severity or the need for assistance with these symptoms. This insight can guide the prioritization of resources and functionalities within the application. The very low usage of the “Call Emergency” function might suggest that it is not frequently needed or that users prefer other means of emergency contact. This could warrant a review of the function’s accessibility or an assessment of user preferences and behaviors in emergencies. Understanding why certain functions are more popular can help tailor the application to better meet user needs. For example, the high use of the “Share Symptom Condition” function for fever indicates a significant need for communication and information sharing among users experiencing this symptom. Enhancing features that facilitate symptom sharing and hospital finding could improve user satisfaction and application effectiveness. In conclusion, analyzing the frequency of function use by symptom type provides valuable insights into user behavior and needs, allowing for targeted improvements and better user support within healthcare applications.

Discussion

The evolution of mobile applications in healthcare represents a significant advancement in addressing public health challenges worldwide. The Smart ICT MED mobile application, developed using the Waterfall methodology (30), is an excellent example of how technology can be leveraged to enhance healthcare delivery and promote self-diagnosis among the general population, particularly in Thailand. This discussion will explore the findings of this study, highlight its significance, and suggest areas for future improvement.

The initial step in developing the Smart ICT MED application involved collecting data on various symptom groups and their associated treatment guidelines. This comprehensive approach ensured that the application covered a broad spectrum of common illnesses, making it a valuable tool for preliminary symptom assessment. By categorizing symptoms into 54 groups, the application provides users with a structured and systematic way to identify potential health issues and seek appropriate advice (26-29). This feature is particularly beneficial in resource-constrained settings where access to healthcare professionals may be limited.

The methodology employed in this study also included rigorous testing and validation processes involving healthcare professionals from diverse backgrounds, including doctors, pharmacists, and nurses. This multi-disciplinary approach ensured the reliability and accuracy of the symptom assessment sequences (26-29), ultimately enhancing the application’s credibility and user trust. Furthermore, the application’s design, which focuses on user experience and accessibility, makes it suitable for users of all age groups and educational backgrounds (31,32).

The results of the functional operation analysis and user interface design reveal that most users found the application intuitive and easy to use. The high satisfaction scores among users, particularly in the areas of usability and interface design, underscore the importance of incorporating user feedback into the development process (32). The inclusion of functions such as finding nearby hospitals, tracking illness conditions, and sharing personal health information with other applications demonstrates the application’s potential to integrate seamlessly into users’ daily lives and support their healthcare needs.

Despite the positive reception, the study identified some areas for improvement. For instance, the “Call Emergency” function was hardly used across all symptoms, indicating a potential gap in user engagement or a preference for alternative emergency contact methods (31,32). Logically, if the symptoms addressed by the application are not severe, this function may not be essential. This finding suggests that the application could benefit from more intuitive design elements or clearer instructions for using this function.

Additionally, while users appreciated the application’s ability to provide preliminary symptom assessments, there was a general reluctance to rely solely on the application for medication advice, indicating a need for further integration with professional medical consultation services.

The dissemination of the Smart ICT MED application through social media and other digital platforms significantly increased its visibility and user base. The high engagement rates on platforms such as Facebook highlight the effectiveness of digital marketing strategies in promoting healthcare applications (12,13). Obviously, this dissemination strategic will help people take better care of their health and find ways to heal themselves.

The application’s implementation phase demonstrated its practical utility in real-world settings. The significant number of downloads and active users indicates a strong demand for digital health tools that facilitate self-diagnosis and healthcare management (9,11). The frequent use of the “Share Symptoms” and “Find Hospital” functions for symptoms like fever and headache highlights the application’s role in helping users find suitable medical facilities. It also allows users to prepare in case their symptoms worsen by recording and tracking them. This information can then be forwarded to a doctor or hospital for further diagnosis.

Therefore, the Smart ICT MED application represents a significant advancement in digital health technology, offering a practical tool for self-diagnosis and preliminary health assessment. Developing this system is not novel innovations or inventions; rather, it involves appropriately applying information technology within the health context.

However, ongoing efforts to enhance user engagement, integrate professional medical consultations, and streamline the publication process are essential for the application’s continued success and wider adoption.

Future research should also explore the long-term impact of such applications on healthcare outcomes and their potential to reduce the burden on healthcare systems by promoting self-care and early intervention.

Limitation

While this research primarily focuses on the development and initial implementation of the SMART ICT MED application, it does not include a long-term impact study. The goal of this application is to promote self-care and early intervention through reliable symptom assessment. However, assessing its sustained impact on healthcare outcomes and its capacity to reduce healthcare system burdens would require further longitudinal research beyond the current study’s scope. Future studies could explore these areas to better understand the application’s long-term benefits and potential for wider healthcare system integration.

Conclusions

The SMART ICT MED mobile application represents a significant advancement in digital health technology, providing an effective tool for self-diagnosis and preliminary health assessment. Developed with rigorous data collection and validation by medical experts, the application offers comprehensive symptom analysis across 54 common illness groups, making it a reliable resource for managing general, non-serious conditions in Thailand. Its user-friendly interface, designed to cater to users of all age groups and educational backgrounds, enhances accessibility and usability, which has been reflected in high user satisfaction scores.

The application’s dissemination through social media and digital platforms has significantly increased its visibility and user base, demonstrating the effectiveness of digital marketing strategies in promoting healthcare tools. Despite initial challenges in application store publication, the application’s outreach extended to 87 hospitals nationwide, supported by social media campaigns and media coverage. This widespread adoption underscores the application’s potential to positively impact public health by facilitating self-care and reducing the burden on healthcare systems.

While the SMART ICT MED application is not a substitute for professional medical consultation, it serves as an initial point of contact for healthcare, helping to reduce unnecessary hospital visits and promoting informed health decisions among users. The study highlights the importance of ongoing efforts to enhance user engagement, integrate professional medical consultations, and streamline the publication process to ensure the application’s continued success and wider adoption.

Acknowledgments

The research aligns with Strategy 3 of Funder, which aims to foster an inclusive, equitable society through the use of digital technology. The author extends gratitude to the director of the 81 hospitals for promoting and endorsing the use of digital health to support health organizations. In addition, this research was calculated the Social Return on Investment (SROI) to support Chiang Mai University’s policy.

Footnote

Data Sharing Statement: Available at https://mhealth.amegroups.com/article/view/10.21037/mhealth-24-38/dss

Peer Review File: Available at https://mhealth.amegroups.com/article/view/10.21037/mhealth-24-38/prf

Funding: This study was supported by the Digital Development Fund for Economy and Society, Thailand Contract (No. 1021/63) (dated 27 May 2020).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://mhealth.amegroups.com/article/view/10.21037/mhealth-24-38/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). In this research, ethical approval was granted by the Chiang Mai University Research Ethical Committee, which issued an ethical exemption (No. CMUREC 64/005). Informed consent was obtained from all participants prior to their involvement in the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Ford R, Wierdermann J. The App & Mobile Case Study Book. Indexed ed. Los Angeles: Taschen America LLC; 2011.
2. Lane ND, Miluzzo E, Lu H, et al. A survey of mobile phone sensing. IEEE Commun Mag 2010;48:140-50. [Crossref]
3. Székely A, Talanow R, Bágyi P. Smartphones, tablets and mobile applications for radiology. Eur J Radiol 2013;82:829-36. [Crossref] [PubMed]
4. Panhale M. Beginning hybrid mobile application development. 1st ed. New York: Apress; 2016.
5. Yi WJ, Jia W, Saniie J. Mobile sensor data collector using Android smartphone. In: 2012 IEEE 55th International Midwest Symposium on Circuits and Systems (MWSCAS). IEEE; 2012:956-9.
6. Pryss R, Reichert M, Herrmann J, et al. Mobile crowd sensing in clinical and psychological trials--a case study. In: 2015 IEEE 28th International Symposium on Computer-Based Medical Systems. IEEE; 2015:23-4.
7. Ahn Y, Bae J, Kim HS. The development of a mobile u-Health program and evaluation for self-diet management for diabetic patients. Nutr Res Pract 2016;10:342-51. [Crossref] [PubMed]
8. Ramey L, Osborne C, Kasitinon D, et al. Apps and Mobile Health Technology in Rehabilitation: The Good, the Bad, and the Unknown. Phys Med Rehabil Clin N Am 2019;30:485-97. [Crossref] [PubMed]
9. Thinnukool O, Khuwuthyakorn P, Wientong P, et al. Non-Prescription Medicine Mobile Healthcare Application: Smartphone-Based Software Design and Development Review. Int J Interact Mobile Technol 2017;11:130-46. [Crossref]
10. Thinnukool O, Khuwuthyakorn P, Wientong P, et al. Type 2 Diabetes Mobile Application for Supporting for Clinical Treatment: Case Development Report. Int J Online Biomed Eng 2019;15:21-38. [Crossref]
11. Sutham K, Khuwuthyakorn P, Thinnukool O. Thailand medical mobile application for patients triage base on criteria based dispatch protocol. BMC Med Inform Decis Mak 2020;20:66. [Crossref] [PubMed]
12. Kemp S. Digital 2022: Thailand. 2022. [Accessed 2023 Jun 9]. Available online: https://www.datareportal.com/reports/digital-2022-thailand
13. National Electronics and Computer Technology Center. Category: innovation-mobile-application. 2018. [Accessed 2023 Jun 9]. Available online: https://www.nectec.or.th/innovation/innovation-mobile-application/
14. Tangcharoensathien V, Witthayapipopsakul W, Panichkriangkrai W, et al. Health systems development in Thailand: a solid platform for successful implementation of universal health coverage. Lancet 2018;391:1205-23. [Crossref] [PubMed]
15. Department of Empowerment and Development of Persons with Disabilities. National Development Plan for Persons with Disabilities, 5th ed, B.E. 2560-2564. Ministry of Social Development and Human Security; 2560. Available online: https://dep.go.th/th/law-academic/knowledge-base/disabled-person-situation
16. National Institute of Emergency Medicine. Criteria for Emergency Severity Triage and Emergency Operation Standards 2011. Bangkok: National Institute of Emergency Medicine; 2013. Available online: https://www.niems.go.th/1/Ebook?group=55
17. Kwon PJ, Kim H, Kim U. A study on the web-based intelligent self-diagnosis medical system. Adv Eng Softw 2009;40:402-6. [Crossref]
18. Kim S, Mun HJ. Design and development of a self-diagnostic mobile application for learning progress in non-face-to-face practice learning. Appl Sci 2021;11:10816. [Crossref]
19. Müller R, Klemmt M, Koch R, et al. "That's just Future Medicine" - a qualitative study on users' experiences of symptom checker apps. BMC Med Ethics 2024;25:17. [Crossref] [PubMed]
20. Bashshur RL, Shannon GW, Smith BR, et al. The empirical foundations of telemedicine interventions for chronic disease management. Telemed J E Health 2014;20:769-800. [Crossref] [PubMed]
21. Piwek L, Ellis DA, Andrews S, et al. The Rise of Consumer Health Wearables: Promises and Barriers. PLoS Med 2016;13:e1001953. [Crossref] [PubMed]
22. Supramaniam P, Beh YS, Junus S, et al. Exploring mHealth app utilization for diabetes self-management: survey insights from a northern district in Malaysia. BMC Public Health 2024;24:3542. [Crossref] [PubMed]
23. Wang K, Varma DS, Prosperi M. A systematic review of the effectiveness of mobile apps for monitoring and management of mental health symptoms or disorders. J Psychiatr Res 2018;107:73-8. [Crossref] [PubMed]
24. Santo K, Singleton A, Rogers K, et al. Medication reminder applications to improve adherence in coronary heart disease: a randomised clinical trial. Heart 2019;105:323-9. [Crossref] [PubMed]
25. Carter MC, Burley VJ, Nykjaer C, et al. Adherence to a smartphone application for weight loss compared to website and paper diary: pilot randomized controlled trial. J Med Internet Res 2013;15:e32. [Crossref] [PubMed]
26. Wolters Kluwer. Clinical Drug Information, Inc. (Lexi-Drugs). Wolters Kluwer Clinical Drug Information, Inc.; 2020. Available online: https://www.wolterskluwer.com/
27. Herrick JB. A Handbook of Medical Diagnosis for Students. RareBooksClub; 2012. Available online: https://dokumen.pub/oxford-handbook-of-clinical-diagnosis-oxford-medical-handbooks-9780199679867-019967986x.html
28. Stern SD, Cifu AS, Altkorn D. Symptom to Diagnosis: An Evidence-Based Guide. McGraw-Hill Educ; 2019. Available online: https://accessmedicine.mhmedical.com/book.aspx?bookID=2715
29. Tanpichat W, Tangkiatkamjai M, Misawasdi P, et al. Rational Drug Use in Community Pharmacy: RDU Pharmacy. Academic Working Group for Promoting Rational Drug Use in Community Pharmacy; 2017. Available online: https://dmsic.moph.go.th/index/download/730
30. GeeksforGeeks. Evolutionary Model - Software Engineering. Available online: https://www.geeksforgeeks.org/software-engineering-evolutionary-model/
31. Manzano-Monfort G, Paluzie G, Díaz-Gegúndez M, et al. Usability of a mobile application for health professionals in home care services: a user-centered approach. Sci Rep 2023;13:2607. [Crossref] [PubMed]
32. Saha D, Mandal A, Pal SC. User interface design issues for easy and efficient human computer interaction: an explanatory approach. Int J Comput Sci Eng 2015;3:127-35.