Development of a web application for the cognitive stimulation of older adults with mild cognitive impairments, using advanced web scraping techniques and generative artificial intelligence

This study describes the design, development, and evaluation of VitalMindAI, a web application that provides personalized cognitive stimulation for older adults with mild cognitive impairment, focusing on memory, attention, and language. This innovative tool integrates web scraping techniques and generative artificial intelligence to provide cognitive exercises tailored to each older adult's unique needs, facilitating caregivers' work and improving the quality of geriatric care. Requirements for developing VitalMindAI were identified through interviews with caregivers at a gerontological center. Based on this information, we created a functional prototype that enables the generation, validation, and management of cognitive activities tailored to the individual needs of older adults. The results showed a significant reduction in the time required to generate cognitive exercises, from several minutes to less than a minute (with VitalMindAI), which evidences the efficiency of VitalMindAI and its high acceptance among caregiver users. Previously, exercises were generated manually by caregivers, taking an average of 15 to 20 min per activity. This demonstrates the improvement achieved with VitalMindAI. In addition, VitalMindAI has demonstrated its ability to dynamically generate content that stimulates cognitive functions in memory, attention, and language, without causing technical interruptions. The evaluation of VitalMindAI was conducted through tests with professional caregivers from a gerontological center, utilizing the Think-Aloud technique and the System Usability Scale (SUS) questionnaire. The Think-Aloud sessions revealed that users found VitalMindAI to be intuitive, agile, and appropriate to the real needs of the gerontological context. However, as a result of the evaluation, users suggested improvements in the visual hierarchy and terminology used. After applying the SUS questionnaire, VitalMindAI obtained an average score of 87.5, placing it in the “Excellent” category for usability. Overall, VitalMindAI proves to be a viable, scalable, and valuable solution for geriatric care settings, automating key tasks and enhancing cognitive stimulation of memory, attention, and language through accessible and intelligent technologies.