AI-sCare+ Care Monitoring System
AI-sCare+ is an AI-powered care monitoring system that combines AI image analysis, IoT sensors, and WebRTC live video to support nursing care facilities with resident monitoring, reduced nighttime patrol burden, anomaly detection, and care record integration.
Care facilities face numerous challenges: nighttime patrols, early detection of falls and sudden health changes, documentation workloads, and privacy-conscious video utilization. AI-sCare+ integrates cameras, sensors, AI, video streaming, recording, and notifications to support staff verification tasks and enable more stable monitoring operations.
Currently, AI-sCare+ related systems are deployed across 8 facilities and 700 rooms.
What You’ll Find in This Document
This page covers the basic architecture, use scenarios, technical components, and key verification points for deploying the AI-sCare+ care monitoring system.
- Overall architecture of the care monitoring system
- Roles of AI image analysis and IoT sensors
- WebRTC live video and recording capabilities
- Approach to reducing nighttime patrol burden
- Care record integration approach
- Privacy-conscious video utilization
- Key points to verify before PoC and deployment
Key Challenges in Care Facilities
The following challenges commonly arise in nursing care facilities:
- Heavy burden of nighttime patrols
- Need for early detection of falls and sudden health changes
- Inconsistent verification quality due to varying staff experience levels
- Excessive false positives that increase on-site workload
- Time-consuming care record entry and verification tasks
- Need for video utilization while respecting privacy
- Tendency for cameras, sensors, and recording systems to be fragmented
AI-sCare+ addresses these challenges by combining AI image analysis, IoT sensors, video verification, recording, notifications, and care record integration.
System Overview
AI-sCare+ is a system that monitors resident status based on in-room camera footage and various IoT sensor data, and notifies staff or administrators as needed.
The main components are as follows:
| Component | Role |
|---|---|
| AI camera / camera module | Captures video of rooms and target areas |
| AI image analysis | Recognizes posture, movement, and abnormal conditions |
| IoT sensors | Acquires supplementary data such as temperature/humidity, toilet occupancy detection, and biometric information |
| Biometric sensors | Monitors bed presence/absence, respiration, heart rate, and other conditions |
| WebRTC live video | Low-latency video viewing from iPad / PC |
| Local recording | Used for situation review, retrospective analysis, and AI improvement data |
| Cloud / server | Infrastructure for status assessment, notifications, and record integration |
| Client interface | Verification screens for staff and administrators |
Monitoring Flow
The basic monitoring flow of AI-sCare+ is as follows:
- Cameras and sensors capture the in-room status
- AI image analysis and sensor data determine the resident’s condition
- Results are sent to the cloud or server side as needed
- Notifications are delivered to client devices such as iPad / PC
- Staff review video, recordings, and record information
- Necessary care and responses are carried out
This flow enables a supportive monitoring framework that leverages AI and sensors rather than relying solely on manual patrols for all verifications.
Key Features
AI Auto-Patrol
AI image analysis automatically checks the status within rooms. While it does not completely replace regular nighttime patrols, it helps reduce staff burden by making it easier to identify situations requiring attention.
Anomaly Detection
Combines AI recognition processes to detect conditions requiring attention in care settings, such as falls, prolonged unnatural postures, and changes in movement.
Detection targets and judgment criteria can be adjusted according to facility operations and camera installation environments.
False Positive Reduction
With variations in rooms, residents, lighting conditions, and bed positions, simple image recognition may produce increased false positives. AI-sCare+ works to improve recognition accuracy using on-site data, aiming for operations that do not add unnecessary verification work.
WebRTC Live Video
Using WebRTC enables low-latency video viewing from iPads and PCs. This is useful for assessing situations before going to the site or for verification after anomaly notifications.
Local Recording and Video Search
Video is recorded locally and can be searched by time period for review. This can be used for situation verification during incidents, sharing among staff, and data review for AI recognition improvements.
Care Record Integration
By linking detection events and verification results with care record systems, manual documentation burden is reduced. Record items, notification content, and integration methods can be designed according to operational needs.
Privacy Considerations
When handling video in care settings, privacy measures are critical. AI-sCare+ supports design options aligned with facility policies, including blurring, local recording, access control, and minimal necessary video review.
Technical Architecture
AI Image Analysis
Combines AI recognition algorithms such as person detection, pose estimation, motion recognition, and anomaly detection.
Depending on the use case, the following recognition processes can be considered:
- Person detection
- Posture and condition recognition
- Fall detection
- Bed-exit / in-bed status monitoring
- Detection of prolonged stillness or abnormal posture
- Custom recognition tailored to facility operations
IoT Sensor Integration
Combining sensor data with video enables more stable condition assessment.
Integration examples:
- Temperature/humidity sensors
- Toilet occupancy sensors
- Biometric sensors
- Bed presence / absence sensors
- Respiration / heart rate data
- Room environment data
WebRTC / Recording SDK
Combines WebRTC low-latency live video, local recording, video download, and PC/iPad interfaces.
Primary use cases:
- Remote video verification
- Video verification after anomaly notifications
- Recording search
- Verification screens for staff and administrators
- Video integration with cloud / server
Edge AI and Cloud Integration
AI processing can be distributed across the camera itself, edge AI processing boards, AI engines, and cloud-side processing, depending on the system configuration.
For on-site processing requirements, edge AI boards can be utilized. For multi-facility or large-scale deployments, cloud-side management, notification, and analysis capabilities are also incorporated.
Expected Benefits
The following benefits can be expected from deploying AI-sCare+:
- Reduced burden of nighttime patrols
- Earlier anomaly detection
- More efficient staff verification tasks
- Labor savings in documentation work
- Situational awareness combining video and sensor data
- Privacy-conscious video utilization
- Construction of a monitoring system tailored to facility operations
PoC and Pre-Deployment Checklist
Before deployment, it is recommended to verify the following items:
Installation Environment
- Room size
- Bed position
- Camera installation position
- Lighting conditions
- Nighttime brightness
- Wi-Fi / wired LAN environment
- Power outlet locations
Conditions to Detect
- Falls
- Bed exits
- Prolonged stillness
- Unnatural postures
- Toilet use
- Biometric data such as respiration and heart rate
- Conditions requiring staff notification
Operational Rules
- Who receives notifications
- Which devices are used for verification
- How long recordings are retained
- How to integrate with care records
- How to manage video access permissions
- How to define privacy policies
Related Technologies
AI-sCare+ is built by combining the following technologies:
- Camera modules
- Edge AI processing boards
- AI image recognition algorithms
- IoT sensors
- WebRTC live video
- Recording SDK
- Cloud notifications
- PC / iPad client interfaces
- Care record system integration
Related Products and Documentation
Frequently Asked Questions
Can it be deployed in existing facilities?
Yes. Retrofit deployment to existing facilities can be considered. However, it is necessary to confirm camera installation positions, network, power supply, privacy policies, and integration conditions with existing record systems in advance.
Can I consult on a PoC-only basis?
Yes. Consultations are welcome starting from the PoC stage. We recommend first organizing the target rooms, conditions to detect, cameras and sensors to use, notification destinations, and recording policies, then beginning with small-scale verification.
Can privacy measures be implemented for camera footage?
Yes. Designs can be tailored to facility operational policies, including blurring, recording scope, retention periods, access permissions, and local recording.
Can AI recognition algorithms be customized?
Yes. Customization of detection targets and judgment criteria can be considered according to the facility and use case. After reviewing camera footage, sensor information, and operational conditions, the necessary AI recognition processes will be designed.
Contact
For inquiries about AI-sCare+ care monitoring system deployment, PoC, customization, and technical documentation, please contact us.
- Deployment consultation for care facility monitoring systems
- PoC consultation for AI image analysis
- Consultation on WebRTC video verification / recording SDK
- Consultation on care record system integration
- Consultation on camera and sensor configuration