Personal Memory Assistant for Alzheimer’s Patients using AI and Real-Time Recognition System

Abstract

Alzheimer\\\'s disease, characterized by progressive memory loss and cognitive decline, presents significant challenges for both patients and caregivers. This design proposes the development of a particular Memory Assistant (PMA) designed to enhance the quality of life for Alzheimer\\\'s cases by integrating advanced artificial intelligence technologies. The PMA utilizes real-time facial and speech recognition to identify individuals and prisoner critical information during relations. When a new person is introduced, the system automatically records their name, snap, and applicable details, creating a substantiated database. Upon posterior hassles, the PMA recognizes the individual and provides the case with contextual information, such as the person\\\'s name and relationship, thereby reducing confusion and supporting memory retention. The system features include real-time announcements, drug monitors, and exigency cautions, all accessible intuitively. It\\\'s erected on a robust tackle foundation, exercising platforms like Raspberry Pi, combined with high-quality cameras and microphones. Crucial considerations for the design include icing data sequestration, carrying stoner concurrence, and addressing ethical considerations related to sensitive particular information.

Introduction

Problem:
Alzheimer’s disease impairs memory and cognitive skills, causing patients to forget important people, events, and daily tasks. This leads to confusion, fear, and loss of independence, creating challenges for both patients and caregivers.

Solution - Personal Memory Assistant (PMA):
A smart AI-powered system designed to support Alzheimer’s patients by assisting with memory recall and daily activities using real-time recognition technologies:

• Face Recognition: Identifies family, friends, and caregivers, announcing their names and relationships to the patient to aid recognition.

• Speech and Voice Recognition: Understands conversations, responding with helpful reminders about names, dates, or tasks.

• Interactive Reminders: Provides gentle audio prompts to reduce confusion and anxiety, encouraging independence.

• Learning Capability: Automatically learns and stores new faces, names, or voices, updating its database for future interactions.

• User-Friendly Interface: Designed for easy use on smartphones, tablets, or other devices, suitable for home, care centers, and hospitals.

Objectives:

1. Help patients recognize and remember important people.

2. Provide context-aware reminders for conversations and daily activities.

3. Promote patient independence and reduce caregiver burden.

4. Continuously learn new information to improve support.

5. Ensure privacy and security of stored data.

6. Design a non-intrusive and intuitive user experience.

Technology Overview:

• Face Detection and Recognition: Continuously monitors camera feeds to detect and identify known individuals.

• Speech-to-Text Conversion: Converts conversations to text for understanding and interaction.

• Data Storage: Organizes and stores facial and conversational data locally with secure access.

• User Interface: Allows caregivers to manage the system, add new contacts, and review past interactions.

System Requirements:

• Hardware: Computer with Intel i5+ CPU, 8–16 GB RAM, dedicated GPU, high-quality camera and microphone, optional display, and stable power supply.

• Software: Python-based system using libraries like OpenCV, TensorFlow/PyTorch, speech APIs, and databases for local storage.

• Network: Internet connection for cloud services and backups; local Wi-Fi or Ethernet for multi-device access.

Background Research:
Studies confirm AI techniques like CNNs for face recognition and speech-to-text technologies enhance memory aids effectively. Emphasis is also placed on data privacy and user-centric, easy-to-use designs for higher adoption.

Impact:
The PMA aims to reduce Alzheimer’s patients’ stress and confusion, support caregivers, and enhance patients’ quality of life by providing reliable, adaptive memory assistance through advanced AI and recognition technologies.

Conclusion

The Alzheimer\\\'s patient surveillance system effectively integrates face and voice recognition to ensure patient safety and nurse support. High reliability was achieved with a total accuracy of 96% for patient identification and emergency scenario detection. Actual warnings with response times of less than 3 seconds improve the practicality of the system in critical situations. The facial recognition module showed 95% accuracy under optimal conditions, but pretreatment improved performance in reduced lighting conditions. Nursing staff consistently rated the user\\\'s user-friendly and functional system with a satisfaction rate of 4.7/5. Continuous updates to patient databases allow for changes in appearance and adaptability to facial audio patterns. The system has the potential scalability of a wider range of health care treatments, including monitoring other patient groups. Feedback mechanisms allow systems to be developed based on user experience and real challenges. Overall, the project illustrates the important potential of AI technologies to improve health outcomes, particularly for populations in need of protection, such as those with Alzheimer\\\'s disease.

References

[1] Ranjithkumar S, Charankumar K S, Kaniskar K, Kiruthika V R, “Smart Assistance System for Alzheimer’s person,” IJARIIE-ISSN, vol.10 Issue-2, pp.1359-1364,2024. [2] Sarita, Tanupriya Choudhury, Saurabh Mukherjee, Chiranjit Dutta, Aviral Sharma, Ayan Sar, “A Wearable Device for Assistance of Alzheimer’s disease with Computer Aided Diagnosis,” EAI Endorsed Transactions on Pervasive Health and Technology, vol.10, no.1, pp.1-10,2023. [3] Reena, R., R. Srishti, M. Harini, Aezeden Mohamed, and Ergasheva Surayyo. “IoT-Based Assistant for Alzheimer\\\'s Patient with Reminder System and Tracking Using GPS” In E3S Web of Conferences, vol. 399, p.04053. EDP Sciences, 2023 [4] Arunachalam, Rajesh & Sunitha, Gurram & Shukla, Surendra & Pandey, Surya & Urooj, Shabana & Rawat, Seema. (2023). “A smart Alzheimer’s patient monitoring system with IoT- assisted technology through enhanced deep learning approach. Knowledge and Information Systems” 65. 1-39. 10.1007/s10115-023-01890-x. [5] R. Chokri, W. Hanini,,W. B. Daoud, S. A. Chelloug and A. M. Makhlouf, in “Secure IoT Assistant-Based System for Alzheimer’s Disease” in IEEE Access, vol.10,pp.4430544314,2022,doi:10.1109/ACCESS.2022.3168311. [6] B. Swasthik, H. N. Srihari, M. K. Vinay Kumar & R. Shashidhar, “Smart Assist for Alzheimer’s Patients andElderly People” journal of New Trends in Computational Vision and Bio-inspired Computing 2022 [7] Morgado-Dias, F.;Bhuyan, B.P.; Tomar, R. “Human Activity Recognition for Elderly People Using Machine and Deep Learning Approaches”. Information 2022, 13,275. https://doi.org/10.3390/info1306025 [8] X. Hong et al., “Predicting Alzheimer’s Disease Using LSTM” in IEEE Access, vol. 7, pp. 80893-80901, 2019, doi: 10.1109/ACCESS.2019.2919385[6] Hayat

Copyright

Copyright © 2025 Rajiyaalam Sheikh, Astha Katre, Sakshi Naik, Shruti Thakre, K. K. Ingole. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.