DESC:The invention is a mobile-based artificial intelligence system designed to assist individuals experiencing cognitive decline, particularly due to dementia. The system(100) captures spontaneous spoken thoughts, processes them into structured digital content using speech-to-text and AI elaboration techniques, and provides users with a personalized interface to manage daily routines. It also includes features such as reminders, emergency alerts, and a secure caregiver access module for remote monitoring and management (106).
The system can be in the form of a –
1. Pendant or Band adopting a lightweight wearable form factor featuring:
i. Smart Pendant: of titanium housing with medical-grade silicone coating for 24/7 comfort
ii. Alternative Band: Flexible wristband option with magnetic charging dock
iii. Elderly-Friendly Design: Large tactile emergency button, haptic feedback, and voice-guided interactions
2. Always-On Architecture
Inspired by the Friend pendant ecosystem:
i. Passive Listening Mode: Continuous ambient monitoring with privacy-preserving wake-word detection ("Buddy" or customizable phrase)
ii. Contextual Activation: Automatically triggers during emotional distress or creative moments
iii. Energy Optimization: longer battery life with kinetic charging supplement
3. On-Device Small Language Model (SLM) Integration
The architecture leverages AI edge for enhanced privacy and accuracy:
i. Curated SLM: 350M parameter model optimized for dementia-specific language patterns and creative elaboration
ii. Federated Learning: Device learns from user patterns while maintaining data privacy
iii. Marketplace Ready: Modular API architecture enables third-party cognitive apps
iv. Offline Capability: Core functions operate without internet connectivity
4. Clinical-Grade Vitals Integration
Enhanced health monitoring for comprehensive medical support:
i. Continuous Monitoring: Heart rate variability, blood oxygen, body temperature, and movement patterns
ii. Medication Correlation: Links creative output quality with medication adherence and timing
iii. Doctor Dashboard: Real-time vitals streaming with cognitive performance analytics
iv. Predictive Alerts: Early warning system for cognitive decline episodes
Individuals with dementia often experience rapid memory loss, confusion, and difficulty in communication, which leads to forgotten thoughts, missed medications, and increased dependence on caregivers. This invention addresses these issues by providing a cognitive support system that captures and organizes thoughts before they are forgotten, ensures timely task (105) and medication reminders, and allows caregivers to monitor (106) and assist remotely. It offers a non-intrusive, user-friendly solution that improves patient autonomy while reducing the emotional and logistical burden on caregivers.
The objectives are achieved by designing a modular software system that runs on a mobile device and optionally synchronizes with a cloud server for data management and remote access. The system(100) captures spoken input(101) from the user through the device’s microphone, processes it using natural language processing algorithms to convert it into text(102), and applies artificial intelligence techniques to categorize or elaborate on the content(103). This processed data is stored and displayed within an intuitive user interface that allows the user or their caregiver to retrieve or modify it later (106).
Task scheduling, reminder notifications, and alert systems (105) are implemented using event-driven software components that trigger based on time, location, or user interaction. The caregiver interface (106) is accessible through a secure login, which allows monitoring and updates from remote devices. All data exchanges are encrypted, and the system is designed to function with minimal user effort, considering the cognitive limitations of the intended users.
AI-driven module that listens to patient speech and automatically categorizes entries into To-Do List, Ideas Vault, or Memory Vault.
A Priority Matrix module that applies the Eisenhower Matrix to prioritize tasks and ideas based on urgency and importance.A Memory Exercises and Activities module providing daily cognitive exercises tailored to patient needs.

The system comprises the following primary components:
• Speech Input Module (101): Captures verbal thoughts or instructions via the device microphone.
• Speech-to-Text Engine (102): Converts speech into editable and storable text format.
• AI/LLM Categorization Module (103): Automatically classifies entries into To-Do List, Ideas Vault, or Memory Vault, and generates appropriate titles.
• AI Elaboration Module (104): Uses cloud-based or on-device models to refine, complete, or rephrase thoughts into coherent outputs.
• Priority Matrix Module (105): Applies the Eisenhower Matrix to prioritize tasks and ideas based on urgency and importance.
• Memory Exercises & Activities Module (106): Provides daily cognitive exercises tailored to the patient's needs.
• Reminder and Calendar System (107): Manages medication alerts, daily tasks, and schedules.
• Memory Vault (108): A secure digital space where labeled photos, names, events, and notes are stored for recall assistance.
• Emotional Monitoring Module (109): Analyzes patient speech and behavior to detect emotional states and sends alerts to caregivers in emergencies.
• Emergency SOS Module (110): Sends alerts and GPS location to registered caregivers or emergency contacts with a single tap.
• Caregiver Dashboard (111): Allows authorized family members or healthcare providers to view, update, and monitor patient data remotely.
These modules interact within a unified software framework that is lightweight, adaptable, and optimized for ease of use by elderly individuals. The backend system optionally includes encrypted cloud storage and synchronization to ensure data availability and security across multiple devices.
Operational Logic of the System:
Upon initialization, the system (100) loads a personalized session for the patient, connecting securely to local or cloud storage for previous entries and system state. The app awaits speech input through the Speech Input Module (101), which is processed via the Speech-to-Text Engine (102).
The AI/LLM Categorization Module (103) then interprets the transcribed content and uses contextual cues to determine if the input reflects a task, a creative idea, a memory, or general narration. Based on semantic relevance, it categorizes the content into the To-Do List, Ideas Vault, or Memory Vault.
Once categorized, the Priority Matrix Module (105) applies the Eisenhower Matrix technique to automatically assign urgency and importance labels. This ensures time-sensitive tasks are highlighted for both patient and caregiver.
The app continuously analyzes user interaction patterns, including sentiment and speech characteristics. If emotional distress, frustration, or sadness is detected using the Emotional Monitoring Module (109), the system triggers a pre-configured alert to caregivers, including a summary of the emotional change and GPS location (if enabled).
Independently, the Memory Exercises & Activities Module (106) recommends daily exercises personalized to the patient’s performance and behavior trends. Engagement history is logged and optionally shared with caregivers through the Caregiver Dashboard (111) for follow-up.
All logic is embedded within an event-driven architecture, where modules asynchronously trigger functions based on user inputs, time schedules, or sensor/emotion-based events. The system prioritizes usability, requiring minimal manual input, and adapts over time based on interaction history.
User Interaction Scenarios:
The user interacts primarily through voice commands. For example, if the patient says “Remind me to take my pills at 6,” the system identifies the intent as a reminder and categorizes it under the To-Do List with a time trigger. If the user says, “I had an idea about a new water bottle design,” the system classifies it under Ideas Vault and uses the AI Elaboration Module (104) to refine and structure the idea. Caregivers may review, approve, or modify any entries via the Caregiver Dashboard (111).
Background and Passive Listening Behavior:
The system supports a passive listening mode when idle, waiting for a wake-word (e.g., “Buddy,” or a customized phrase) to begin interaction. In this state, minimal resources are consumed, and privacy protocols ensure that recording or processing occurs only upon activation. If enabled, the system can prompt the user with questions like “Do you want to record something?” if no activity is detected for a set duration.
Adaptive Personalization and Learning Over Time:
The system includes a continual learning mechanism that adapts to the user's behavior and preferences. For instance, if the patient frequently reschedules tasks or forgets exercises, the app adjusts notification frequency or suggests simpler activities. It prioritizes recurring task types and suggests improvements to caregivers. Emotional response data, exercise performance, and task compliance are also used to fine-tune the AI Elaboration Module (104) and scheduling logic over time.
ADVANTAGES OF THE INVENTION
• Enables early capture of cognitive expressions before memory fades.
• Enhances task management by prioritizing patient-generated tasks using the Eisenhower Matrix.
• Offers structured cognitive stimulation through memory exercises and activities.
• Improves safety by monitoring emotional well-being and alerting caregivers proactively
• Supports caregivers through alerts and real-time monitoring.
• Reduces emergency response time with built-in GPS and SOS features (107).
• Bridges the technology gap for the elderly through simplified design.
The embodiments herein provide a system for assisting people with dementia to create artistic work (hereinafter referred to as “system” or “system (100)”). The system comprises a plurality of modules, and application software configured to be installed and run on a smartphone carried by the patient himself or the caregiver and the GSM server connecting the sensor with the application software on the smartphone.
Throughout this application, with respect to all reasonable derivatives of such terms, and unless otherwise specified (and/or unless the particular context clearly dictates otherwise), each usage of:
“a” or “an” is meant to read as “at least one”,
“the” is meant to be read as “the at least one.”
References in the specification to “one embodiment” or “an embodiment” mean that a particular feature, structure, characteristic, or function described in connection with the embodiment is included in at least one embodiment. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.
Hereinafter, embodiments will be described in detail. For clarity of the description, known constructions and functions will be omitted.
The present invention is illustrated with reference to the accompanying drawings, throughout which reference numbers indicate corresponding parts in the various figures. These reference numbers are shown in brackets in the following description.
In an implementation of one of the exemplary embodiments of the present invention, structural and functional aspects of the system (100) are explained by referring to Figure 1. The system (100) comprises a plurality of modules and application software configured to be installed and run on a smartphone. The plurality of modules may comprise a speech input module (101), a speech-to-text conversion module (102), and an artistic work or invention elaboration module (104). The modules (101, 102, 103) may be a part of a smartphone with application software that is configured to generate the artwork by leveraging artificial intelligence techniques. The speech input module (101) is communicatively coupled to the speech-to-text module (102) and may comprise a microphone of the smartphone that is configured to receive speech input from the dementia patient. The speech input may be related to a creative idea originating from the dementia patient. The speech-to-text conversion module (102) converts the received speech input into equivalent text. The artistic work or invention elaboration module (104) is configured to search resources available on the internet to check the originality of the idea presented as text and create an artwork elaborating the idea presented by the dementia patient. One or more modules among the plurality of modules (101, 102, 103) can be a part of the application software run on the smartphone.
Features included in-app and module:
1. Memory Aids
• Photo Albums: Include labelled photos of family, friends, and important events.
• Notes and Reminders: Allow caregivers to set reminders for medications, appointments, and daily routines, ideas for authors and inventors.
2. Navigation Assistance
• GPS Tracking: Track the patient's location and provide directions back home.
• Safe Zones: Set up geofenced areas and alert caregivers if the patient leaves these zones.
3. Emergency Features
• Alerts: When the user goes into dementia mode, automatically the app observes based on historical data and sends an alert to the relatives/friends/caretakers through a call/message/vibration alert so that they go to the user and take necessary action quickly
• Emergency Button: One-tap access to contact caregivers or emergency services.
• Medical Information: Store and easily share important medical information, such as allergies and medications.
4. Brain Games: Simple puzzles and games that stimulate cognitive function.
• Memory Exercises: Activities to help with recall and recognition.
5. Daily Routines
• Schedule Management: A clear, easy-to-read daily schedule to help patients remember their tasks and routines.
7. Health Monitoring
• Medication Tracker: Reminders and logs for taking medications.
• Vital Signs Logging: Record and monitor blood pressure, glucose levels, etc.
8. User-Friendly Interface
• Simple Design: Large buttons, high contrast, and intuitive navigation to cater to cognitive impairments.
• Alerts and Notifications: Real-time notifications to caregivers about important events or changes.
• Community Features: Forums or chat rooms for connecting with others in similar situations.
9. Diet and Nutrition
• Meal Planning: Nutritional advice and meal planning tools.
• Recipe Book: Easy-to-follow recipes with step-by-step instructions and visuals.
10. Exercise and Mobility
• Exercise Programs: Guided physical activity routines tailored for dementia patients.
• Step Counter: Track daily steps and encourage mobility.
11. Speech and Language
• Speech Therapy: Tools and exercises to assist with communication and language skills.
• Phrasebook: Common phrases and responses to aid in conversation.

12. Mindfulness and Relaxation
• Meditation Guides: Simple meditation and breathing exercises to reduce anxiety.
• Relaxation sounds ambient sounds or nature recordings to help with relaxation and sleep.
,CLAIMS:We claim:
1. A system for assisting individuals with dementia in the form of a standalone device as a pendant or band, comprising:
a speech input module for capturing verbal inputs;
a speech-to-text conversion engine for transforming spoken input into editable text;
an AI/LLM categorization module for classifying said text into at least one of: To-Do list, Memory Vault, or Ideas Vault;
an AI elaboration module for refining or completing user input into structured content;
a priority matrix module using the Eisenhower Matrix for ranking inputs by urgency and importance;
a memory exercises module for delivering personalized cognitive activities;
a reminder and calendar system;
a memory vault for storing labeled multimedia;
an emotional monitoring module for detecting patient mood or stress;
an emergency SOS module for caregiver alerts;
a caregiver dashboard for real-time remote support and monitoring.
2. The system of Claim 1, wherein said AI elaboration module is configured to automatically generate creative artistic works or inventions based on original verbal ideas expressed by the user.
3. The system of Claim 1, wherein the speech input module is activated through a customizable wake-word and supports passive listening with strict privacy enforcement.
4. The system of Claim 1, further comprising adaptive learning algorithms to personalize task suggestions, cognitive exercises, and notification frequency based on historical user behavior.
5. The system of Claim 1, wherein the memory vault comprises multimedia content including labeled photos, notes, and events stored for user recall and accessible by caregivers.
6. The system of Claim 1, wherein said emotional monitoring module utilizes speech tone and semantic analysis to assess emotional changes and triggers alerts when distress is detected.
7. The system of Claim 1, wherein the emergency SOS module transmits GPS-based location data and emergency alerts to preconfigured contacts or caregivers.
8. The system of Claim 1, wherein the caregiver dashboard allows secure, authorized access to real-time and historical data for monitoring and adjusting user support.
9. The system of Claim 1, wherein the categorization and elaboration modules are configured to operate both on-device and via a cloud-based infrastructure.
10. The system of Claim 1, further comprising:
a navigation module with GPS tracking and geofencing;
a diet and nutrition module with meal planning;
a speech therapy module with phrasebook integration;
a mindfulness and relaxation module with audio content.
11. The system of Claim 1, wherein the mobile application is configured with a user-friendly interface featuring high-contrast visuals, large buttons, and voice-guided interaction.
12. A method for assisting people with dementia using the system of Claim 1, comprising the steps of:
capturing verbal inputs;
processing them into structured text;
categorizing and elaborating ideas using AI modules;
storing said outputs in categorized vaults;
and enabling caregiver support and alerts via remote interface.