Search Results
Found 1 results
510(k) Data Aggregation
(90 days)
SoftSpot™ is intended to be used for cranial measurement for infants from 1 month up to 2 years of age in conjunction with other clinical methods.
SoftSpot™ is not intended to be used as a tool for preparing a helmet.
SoftSpot™ is a software that runs on a smartphone, tablet, PC, etc. and uses top view pictures of infant's head to measure cranial index (CI) and cranial vault asymmetry index (CVAI) similar to a mechanical caliper used by physicians which is also known as craniometer.
After the photo is taken, the head contour is identified and then, based on the current standards of measuring infants' head shape, characterizing parameters including the ratio of width over length (known as Cranial Index or CI) and the ratio of difference in diagonals over the larger one (known as Cephalic Vault Asymmetry Index or CVAI) are calculated (see Test Principle below).
The SoftSpot platform allows different users to upload or capture top view videos and photos of the infant's head. The user can be a physician/nurse (during well visits) or parents/guardians (between the visits). Once data are collected, they are sent to PediaMetrix Amazon Web Services (AWS) cloud. At PediaMetrix, the data analyst will use a windows-based software to download and analyze the data. The analysis includes conversion of the video into photos, standardization of the images (3x4, all nose up), selection of 5 good images, segmentation of the head contour and nose tip. Then, the software will calculate Cl and CVAI. Finally, a pdf report is generated with the results and sample top view photos and sent to the physician's account. Parents will also be notified when the results are ready. Then, the parent visits the physician (either in person or online) to discuss the results and next steps.
Here's a breakdown of the acceptance criteria and the study details for the SoftSpot™ device, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance (Prospective Study) | Reported Device Performance (Retrospective Study) |
|---|---|---|
| Sufficient correlation between SoftSpot measurements and standard caliper measurements. | CI: Spearman correlation coefficient = 0.96 | CI: Spearman correlation coefficient = 0.93 |
| CVAI: Spearman correlation coefficient = 0.94 | CVAI: Spearman correlation coefficient = 0.94 | |
| Acceptable Mean Absolute Error (MAE) for CI. | CI: MAE = 1.48 units | CI: MAE = 2.17 units |
| Acceptable Standard Deviation (STD) for CI. | CI: STD = 0.92 units | CI: STD = 1.53 units |
| Acceptable Mean Absolute Error (MAE) for CVAI. | CVAI: MAE = 1.44 units | CVAI: MAE = 1.62 units |
| Acceptable Standard Deviation (STD) for CVAI. | CVAI: STD = 0.85 units | CVAI: STD = 1.26 units |
2. Sample Size Used for the Test Set and Data Provenance
- Prospective Study:
- Sample Size: 25 infants
- Data Provenance: Prospective, collected by a physical therapist in a plagiocephaly clinic at the University of Pittsburgh Children Hospital (USA).
- Retrospective Study:
- Sample Size: 100 infants' head measurements
- Data Provenance: Retrospective, collected from patients with deformational plagiocephaly and brachycephaly (DPB) in a helmet clinic at Singapore.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
- Prospective Study: The ground truth was established by "clinicians using a standard caliper" who were "a physical therapist" at the University of Pittsburgh Children Hospital. The specific number of clinicians is not explicitly stated beyond "a physical therapist." Their specific years of experience are not mentioned.
- Retrospective Study: The ground truth was based on "direct measurements from 3D scans of the head." It's not explicitly stated that human experts established this ground truth, but rather that the 3D scans served as the reference.
4. Adjudication Method for the Test Set
The document does not explicitly describe an adjudication method (like 2+1, 3+1). For both studies, the comparison is made against a single reference measurement (standard caliper or 3D scan).
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. The studies compared the device's measurements to a "standard caliper" (prospective) or "3D scans" (retrospective), not to multiple human readers' interpretations with and without AI assistance.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) Was Done
The description of SoftSpot™ indicates a human-in-the-loop process where a data analyst at PediaMetrix manually processes the collected data (conversion of video to photos, image standardization, selection of good images, segmentation). While the calculation of CI and CVAI is automated by the software after these manual steps, the initial processing involves human intervention. Therefore, a purely standalone algorithm performance study (without any human input in the processing workflow) was not explicitly performed or described.
7. The Type of Ground Truth Used
- Prospective Study: Ground truth was established by physical measurements taken with a standard caliper by clinicians.
- Retrospective Study: Ground truth was established by direct measurements from 3D head scans.
8. The Sample Size for the Training Set
The document does not provide any information about the training set size for the SoftSpot™ algorithm. The studies described are performance evaluations (test sets).
9. How the Ground Truth for the Training Set Was Established
Since information about the training set is not provided, the method for establishing its ground truth is also not mentioned.
Ask a specific question about this device
Page 1 of 1