Page 127 - AIH-2-2

Artificial Intelligence in Health Transgender healthcare

individuals through AI in medical tourism. SmartPLS 4 their working experience, with 14% having 0 – 4 years of
GmbH software was employed to integrate AI into medical experience, 30.7% having 5 – 11 years of experience, 39.3%
tourism by assessing relationships between constructs using having 12 – 21 years of experience and 16% having 21 –
the partial least squares structural equation modeling (PLS- 30 years of experience.
SEM) program. Discriminant validity was tested using The combined impacts of medical and risk factors,
the Fornell-Larcker criterion, ensuring that constructs represented by values ranging from 0.2 to 0.5, indicate
were distinct and measured appropriately. Relationships the strength of the relationships between these factors
between constructs were analyzed to determine the impact and the performance of the AI-based health system’s
of AI on healthcare and medical tourism.
internal model. These values reflect the system’s ability to
To evaluate the reliability of constructs, metrics such as improve healthcare delivery, particularly in addressing
Cronbach’s alpha, composite reliability (rho_a and rho_c), cultural disparities. In addition, international medical
and the average variance extracted (AVE) were used. standards emphasize the importance of integrating
Cronbach’s alpha assessed internal consistency, composite AI to ensure safe and reliable medical destinations for
reliability measured indicator loadings, and AVE quantified patients.
variance captured relative to measurement error. All
constructs met the thresholds for reliability (Cronbach’s Discriminant validity values of 0.6, 0.5, 0.8, and 0.7
alpha ≥0.7, AVE ≥0.5), confirming the robustness of the serve as prospective measurements for factors influencing
measurement model. AI in medical tourism, with long-lasting benefits for
transgender individuals in improving health outcomes.
Discriminant validity was further validated using the The results of this study show that the Fornell-Larcker
heterotrait-monotrait ratio (HTMT), which compares Criterion is between 0.5 and 0.9. In AI-supported medical
between-construct and within-construct correlations. tourism, the study evaluates two model formats. Both
HTMT values below 0.9 confirmed that constructs were models demonstrate acceptable fit indices and meet the
distinct and free from multicollinearity. These findings discrepancy unweighted least squares criteria, a metric
ensure that the structural model is reliable for exploring used to assess model fit in PLS-SEM. This indicates that
AI’s role in medical tourism. the models effectively capture the relationships among
4. Results and discussion constructs and provide reliable insights into the role of
AI in medical tourism. The results of the tests show a
There were 46.1% of female respondents and 43.9% of male correlation of 1.000 between all items, indicating a strong
respondents as mentioned in Table 1 in this study. The age relationship between the components of the AI health
distribution of the respondents was as follows: 33.6% were system. This finding is consistent with the path coefficient
between 20 and 29 years old, 23.4% were between 30 and values, which range between 0 and 1, bolstering the notion
39 years old, 22% were between 40 and 49 years old, and that an improved AI health system significantly lowers
the remaining 21% of the respondents were in the 50–65 the possible hazards associated with medical travel and
age groups. As for the education qualifications of the tourism as depicted in Table 2.
study population, 27% had a bachelor’s degree, 29% had
a master’s degree, and 12.3% had an intermediate level As shown in Table 3, the reliability values derived from
of education; the remaining 38% of the participants had PLS-SEM for the AI system (0.796), medical travel and
postgraduate degrees, suggesting that they were better risk factors (0.844), attitude (0.742), behavioral intention
qualified to provide insights on the use of AI in medical (0.814), destination image (0.879), and medical tourism
travel. Their education qualifications correlated with (0.642) indicate significant relationships between these
constructs and their respective indicators, reflecting
internal consistency within the model. In addition,
Table 1. Path coefficients
the r-value of 0.343 for AI health suggests a moderate
Path Path coefficients positive correlation with other variables in the model.
Medical tourism -> AI system 0.290 These findings demonstrate that higher-quality AI tools
AI system -> Destination image 0.578 are strongly associated with improved outcomes in
Medical travel and risk factors -> Attitude 0.525 medical travel and tourism, as they enhance healthcare
delivery and patient satisfaction while addressing
Behavioral intention -> AI system 0.603 associated risk factors and destination perception. This
Destination image -> Attitude 0.572 is supported by the Pearson correlation coefficient (r)
Attitude -> Behavioral intention 0.344 value of 0.343 for AI health, with related variables falling
Abbreviation: AI: Artificial intelligence. within the range of 0.3, showing a good model and data

Volume 2 Issue 2 (2025) 121 doi: 10.36922/aih.3384