- UAA Congress 2025

Back

Final Presentation Format

Non-Moderated Poster Abstract

Eposter Presentation

Eposter in PDF Format

https://storage.unitedwebnetwork.com/files/1237/a6e4db4bc3f44004b06e33bdee5a2aab.pptx

Accept format: PDF. The file size should not be more than 5MB

Eposter in Image Format

https://storage.unitedwebnetwork.com/files/1237/3b95b72533cbad91516714b2e3da022e.pptx

Accept format: PNG/JPG/WEBP. The file size should not be more than 2MB

Abstract

Abstract Title

Using an AI Predictive Model in Detecting High-Risk Erectile Dysfunction Patients to Prevent Disease Deterioration

Presentation Type

Podium Abstract

Manuscript Type

Clinical Research

Abstract Category *

AI in Urology

Author's Information

Number of Authors (including submitting/presenting author) *

No more than 10 authors can be listed (as per the Good Publication Practice (GPP) Guidelines).
Please ensure the authors are listed in the right order.

Co-author 1

Jhih-Cheng Wang tratadowang@gmail.com Chi Mei Medical Center Urology Taianan Taiwan *

Co-author 2

Yung-Fu Chen tratadowang@gmail.com Central Taiwan University of Science & Technology Department of Dental Technology & Materials Science, Taichung Taiwan -

Co-author 3

Co-author 4

Co-author 5

Co-author 6

Co-author 7

Co-author 8

Co-author 9

Co-author 10

Co-author 11

Co-author 12

Co-author 13

Co-author 14

Co-author 15

Co-author 16

Co-author 17

Co-author 18

Co-author 19

Co-author 20

Abstract Content

Introduction

Erectile dysfunction (ED) is characterized by difficulty in attaining or maintaining an adequate erection during sexual activity and is associated with related comorbidities. Hence, designing predictive systems based on features related to acquired comorbidities is effective in predicting the occurrence of ED for high-risk patients. Annual visits of ED patients were significantly less than those without ED according to the data from the National Health Insurance Research Database. We applied the developed ED predictive model to find high-risk patients for early intervention by designing an m-Health system for managing comorbidities and avoiding disease deterioration so that ED occurrences can be reduced.

Materials and Methods

An m-Health system integrating health educational materials and regular reminding of visiting physicians was designed for managing comorbidities associated with ED. The developed ED predictiv.e model was applied to the urological clinic for detecting patients with higher risk of developing ED. To verify the effectiveness of the m-Health intervention, a total of 31 high-risk patients were recruited for the m-Health intervention. The participants were asked to install an m-Health APP on their smartphones before intervention as well as to fill SHIM questionnaire at the baseline and after 6 months of m-Health intervention. The TAM questionnaires were also obtained after the intervention.

Results

The paired t-test was conducted to compare SHIM scores between baseline and post-intervention, while the one-sample t-test (neutral value=3) was applied to analyze the TAM surveys. The SHIM score after m-Health intervention exhibited significantly greater than the baseline (14.78±8.96 vs 12.70±8.45, p<0.05), indicating m-Health intervention is useful in improving ED symptoms. The perceived usefulness was validated with the result of the TAM survey. The participants strongly (p<0.001) agreed that the m-Health system was useful in managing comorbidities (4.13±0.54) and alleviating ED symptoms (3.65±0.76) with a QoL of 4.00±0.59.

Conclusions

Such results show that the ED m-Health system is useful for consolidating appropriate lifestyles for patients who may develop ED.

Keywords

predictive model, erectile dysfunction (ED), comorbidity, m-Health, disease management.

Figure 1

Figure 1 Caption

Figure 2

Figure 2 Caption

Figure 3

Figure 3 Caption

Figure 4

Figure 4 Caption

Figure 5

Figure 5 Caption

Presentation Order