Search Results

The Dream Sock analyzes photoplethysmography data to identify instances when the infant's pulse rate (PR) and/or oxygen saturation (SpO2) moves outside a preset range, and provides a notification to the caregiver, prompting them to assess the infant. The Dream Sock also displays the infant's PR and SpO2 values to the caregiver and displays trends in these measured values, and their relationship to the preset ranges, over time. These PR and SpO2 notifications and displays on the Dream Sock are intended for use in infants who are 1 to 18 months of age and between 6 to 30 lbs.

The Dream Sock is intended for over-the-counter (OTC) use only in a home environment. It is not intended to provide notification for every episode of the unexpected occurrences of elevated or depressed PR or a low SpO2 level: rather, the feature is intended to provide a notification only when sufficient data are available for analysis. The notifications and associated data can be used to supplement the decision by caregivers to seek additional guidance for medical care of the infant. The feature is not intended to replace traditional methods of monitoring, diagnosis or treatment.

The Dream Sock is not intended for use with infants previously diagnosed with cardiovascular or respiratory disease or conditions.

Device Description

The Dream Sock uses photoplethysmography (PPG) on the infant's foot to measure the continuous pulse rate (PR) and oxygen saturation (SpO2) in the home environment. The device is intended to monitor healthy infants 1 to 18 months of age and between 6 to 30 lbs. The device includes alarms to alert the caregiver that the pulse rate or SpO2 has gone outside a preset range. The alarm thresholds are pulse rate <50 or >220 bpm and the SpO2<80%. The device calculates the outputs during times of motion and non-motion. The device includes 3 components: the sock hardware, the base station and mobile application. The wearable connects to the base station via Bluetooth (BLE).

The Sock is the skin-contacting component that contains the PPG sensor and accelerometer. The accelerometer is used in signal quality algorithm. If too much motion is detected, then no measurement will be output. The PR and SpO2 algorithms are housed on the Sock. The algorithm uses a 10 second rolling window of the PPG data to calculate a measurement every second. The Sock communicates with the Base station using Bluetooth.

The Base Station receives the readings and status from the sensor. When the PR and/or SpO2 measurement goes outside the preset threshold(s), a visual and auditory alarm will sound alerting the caregiver. When the caregiver acknowledges the alarm, the top of the base station is clicked to silence the alarm. The base station communicates with the Cloud which communicates with the phone application through WIFI. Additionally, this is used to charge the sensor.

The mobile application displays the measurements and trend graphs of previous data. A new measurement is displayed every 5 seconds on the mobile application. If the base station alarms. the mobile application will display a notification.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the studies proving the device meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Test Category	Specific Test/Metric	Acceptance Criteria	Reported Device Performance
Biocompatibility/Materials	Cytotoxicity	Grade $\le$ 2	PASS at 1:8 dilution
	Sensitization	No evidence of tissue reaction	PASS
	Irritation	No evidence of tissue irritation	PASS
	Acute Systemic Toxicity	No unscheduled deaths; no weight loss >10%	PASS
	Material Mediated Pyrogenicity	Animals survive and are healthy via temperature monitoring	PASS
Electromagnetic & Electrical	Essential Performance	Conforms to IEC 60601-1-2:2014, IEC 60601-1:2006 + A1:2013, ASNI C63.27:2017, AAMI TIR69:2017, IEC 60601-1-11:2015/AMD1:2020	Conforms to all listed standards
Software	Validation (Major LOC)	Appropriate documentation provided to support validation in accordance with FDA Guidance Document, "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices," (issued May 11, 2005)	Documentation provided and accepted for Major LOC
Performance Testing - Bench	Pulse Amplitude (PA) Accuracy	$\le$(b)% ARMS AND As the simulator PA rises from (b)%, the $\Delta$PR $\le$ 1bpm & Simulator PA $\Delta$SpO2 (b)% $\le$ PA $\le$ (b)% (b)% $\le$ PA < (b)% $\le$ (b)% (Specific numerical values for (b) are redacted in the provided text.)	PASS (Specific numerical values for the reported performance are not provided, but it states "PASS")
	Pulse Rate Accuracy	$A_{RMS} \le$ 1 bpm	PASS
	Alarm Testing	Functionality works as intended (per 60601-1-8:2006+ AMD1:2012+AMD2:2020)	PASS
Clinical Performance	SpO2 Accuracy (Breathe-down)	RMS < 3% (for both non-motion and motion)	Non-motion RMS: 2.51% Motion RMS: 2.46% (Meets criteria)
	Pulse Rate Accuracy (Breathe-down)	RMS < 5 bpm (for both non-motion and motion)	Non-motion RMS: 2.47 bpm Motion RMS: 2.54 bpm (Meets criteria)
	SpO2 Accuracy (At-Home Study)	(Implicit: Demonstrate acceptable accuracy for SpO2 in intended use population)	Average Root Mean Square (Ams) of 2.16% (This is a performance result, and while not an explicit "acceptance criteria" value, it shows the device's accuracy in the intended environment, implicitly meeting the expectation for clinical performance.)
	Pulse Rate Accuracy (At-Home Study)	(Implicit: Demonstrate acceptable accuracy for PR in intended use population)	Average Root Mean Square (Ams) of 3.53 bpm (Similar to SpO2, this is a performance result indicating acceptable accuracy.)
	Alarm Specificity (Bradycardia)	(Implicit: High specificity to minimize false positives)	100% (High specificity indicates very low false positives, meeting an implied acceptance criterion for alarm reliability.)
	Alarm Specificity (Oxygen Desaturation)	(Implicit: High specificity to minimize false positives)	99.86% (High specificity indicates very low false positives, meeting an implied acceptance criterion for alarm reliability.)
	Skin Irritation	Low rate and low severity of skin reactions (Implicit criterion based on biocompatibility and safety.)	Estimated rate of reactions per person per week of 0.16 (95% CI 0.01, 0.40). All skin reactions were severity 1 (Slight indentation or mild color change). (Meets implicit criterion for low rate and severity.)
Human Factors/Usability	Caregiver Use & Interpretation	Caregiver can correctly use the device based solely on reading the labeling and correctly interpret outputs and understand next steps.	All subjects responded to critical tasks by responding to the infant upon annunciation of notification events. Correct response included assessing the infant's condition and correcting the root cause of the notification. (Meets criterion for correct use and response.)

2. Sample Size for the Test Set and Data Provenance

Breathe-down Study:
- Sample Size: 18 healthy adults
- Provenance: Prospective, conducted in a controlled environment (likely a lab/clinical setting) with induced lower SpO2. "invasive pulse oximetry and the Masimo pulse oximeter" used as reference.
At-Home Study:
- Sample Size: 35 subjects (infants between post menstrual age $> $44 weeks and $\ge$ 18 months, weighing 6-30 lbs)
- Provenance: Prospective, conducted in a home environment. Independent from the algorithm training set.
NICU Study:
- Sample Size: 66 patients (infants postmenstrual age $< $44 weeks)
- Provenance: Prospective, single-center clinical study conducted in a Neonatal Intensive Care Unit (NICU).
Skin Observation Study:
- Sample Size: 43 participants (infants with an average age of 7.0 $\pm$ 4.7 months). 30 wore the device for $\ge$ 48 hours.
- Provenance: Prospective study.
Usability Testing:
- Sample Size: 20 participants (lay caregivers)
- Provenance: Conducted in a simulated-use environment.

3. Number of Experts Used to Establish Ground Truth and Qualifications

The document does not explicitly state the number or specific qualifications of experts used to establish a "ground truth" for the test sets in the typical sense of expert annotation for machine learning.

For the Breathe-down Study, an "invasive pulse oximetry and the Masimo pulse oximeter" served as reference devices for SpO2 and Pulse Rate measurements.
For the NICU Study, "hospital grade pulse oximeters and electrocardiogram (ECG)" were used as continuous monitoring references.
For the Skin Observation Study, the skin inspections were likely performed by trained medical personnel, though their specific qualifications or number are not detailed.

4. Adjudication Method for the Test Set

The provided text does not mention any formal adjudication methods (like 2+1 or 3+1) for establishing ground truth for the test sets. The ground truth seems to be established through:

Comparison against established reference medical devices (invasive pulse oximetry, Masimo pulse oximeter, hospital-grade pulse oximeters, ECG).
Direct observation for events like skin irritation.
Observation of user actions in usability testing.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No multi-reader multi-case (MRMC) comparative effectiveness study comparing human readers with AI assistance versus without AI assistance was mentioned or conducted, as this device is a direct monitoring device and not an AI-assisted diagnostic tool for human interpretation.

6. Standalone (Algorithm Only) Performance

Yes, standalone (algorithm only) performance was evaluated for the Dream Sock.

Breathe-down Study: This study directly measured the SpO2 and Pulse Rate accuracy of the device (algorithm) against reference methods under simulated motion and non-motion conditions.
At-Home Study: This study evaluated the SpO2 and Pulse Rate accuracy of the device in a home environment.
NICU Study: This study evaluated the alarm functionality (specificity) of the device's algorithms against reference monitoring.

These studies assess the device's inherent ability to measure and alarm without direct human input beyond setting up the device and receiving notifications.

7. Type of Ground Truth Used

The ground truth used in the studies includes:

Reference Medical Devices: Invasive pulse oximetry, Masimo pulse oximeter, hospital-grade pulse oximeters, and electrocardiogram (ECG) for SpO2 and Pulse Rate measurements and alarm validation.
Direct Observation/Subjective Assessment:
- Skin irritation assessed through inspections (Skin Observation Study).
- Caregiver responses to notifications in usability testing.

8. Sample Size for the Training Set

The document explicitly states that the subjects in the At-Home Study (35 subjects) were "independent from the algorithm training set." However, it does not provide the sample size for the training set itself.

9. How the Ground Truth for the Training Set Was Established

The document does not provide details on how the ground truth for the training set was established. It only mentions that the At-Home study's subjects were independent from the training set, implying a separate training phase occurred, but offers no specifics on its data sources or annotation methods.

Ask a Question

Ask a specific question about this device

Page 1 of 1