Search Results

VerTouch is indicated to aid in the localization of a lumbar interspinous space, and the marking of an identified insertion site, for diagnostic and therapeutic neuraxial procedures.

VerTouch is indicated for use in hospital facilities and clinics by right- and left-handed emergency medicine, neurology, anesthesiology, and pain medicine professionals for assistance with spinal punctures.

The VerTouch handle is intended to be held with the left hand. Imaging and marking are intended to be performed with the right hand.

VerTouch is indicated for use on patients at least 18 years of age with BMI

Device Description

IntuiTap's VerTouch™ Spinal Imaging Device ("VerTouch" or "Device") is intended to provide a two-dimensional (2D) pressure map of posterior spinal anatomy to help the physician plan and approach his/her needle placement during neuraxial procedures. The Device is intended to aid in the localization of an interspinous space, for diagnostic and therapeutic neuraxial procedures, including lumbar punctures, neuraxial anesthesia (spinals, and combined spinalepidurals), epidural steroid injections, and epidural blood patches. The Device includes functionality to guide a marking tool.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

Acceptance Criteria (Special Controls)	Reported Device Performance and Evidence
(1) Clinical performance testing must demonstrate that the device performs as intended under anticipated conditions of use and must evaluate:	STU-001 (Multicenter RCT, N=90): - Primary Endpoint (Number of insertion attempts): Did not reach statistical significance in the Intent-to-Treat (ITT) population. In the Per-Protocol (PP) population, mean insertions were 1.2 for VerTouch vs. 2.1 for manual palpation (p=0.022), demonstrating a significant reduction in insertion attempts for the marking workflow. - Secondary Outcomes: None reached statistical significance for the device group. - Procedure Success Rate: In the marking workflow (intended use), STU-001 showed an 87.5% success rate for VerTouch, comparable to palpation (89.1%). - Adverse Events: Two adverse events in the VerTouch arm (syncope, hemorrhage) were determined to be not related or unlikely related to the device. No significant difference in other safety metrics (unintended dural puncture, paresthesia, traumatic taps, referral to radiology, conversion from spinal to epidural).
STU-002 (Single-center RCT, N=81): - Primary Endpoint (Combined total number of insertions or redirections): Not directly reported as a combined endpoint in the results table, but redirections and reinsertions are reported separately. - Redirections: VerTouch 0.72 ± 0.9, Palpation 2.35 ± 3.9, Ultrasound 2.81 ± 3.7 (p=0.041), suggesting VerTouch may reduce needle redirections. - Reinsertion: VerTouch 0.11 ± 0.3, Palpation 0.50 ± 1.0, Ultrasound 0.96 ± 1.5 (p=0.017), suggesting VerTouch significantly reduces reinsertions. - Procedure Success: 100% procedural success reported for all subjects across all groups. - Adverse Events: No complications or adverse events observed.
(i) Rate of neuraxial procedure success across the intended use population;	STU-001: Marking workflow: 87.5% success. Overall palpable: 89.1% success. STU-002: 100% procedural success reported for all subjects.
(ii) Procedure time;	STU-001: Tertiary and exploratory analyses showed total procedural time was not significantly different between device and palpation methods. STU-002: Total Procedure Time (min): VerTouch 8.66 ± 2.6, Palpation 7.88 ± 3.1, Ultrasound 9.81 ± 4.1 (p=0.121). Not significantly different.
(iii) Adverse events, including pain, bleeding, unintended dural puncture, and injury to other spinal structures;	STU-001: Two adverse events in VerTouch arm (syncope, hemorrhage) deemed unrelated/unlikely related. No significant difference in unintended dural puncture, paresthesia, traumatic taps. STU-002: No complications or adverse events observed.
(iv) Comparison of performance to manual palpation or alternative method of determination of the target insertion site.	STU-001: Compared to manual palpation. VerTouch (marking workflow) showed comparable success rates. STU-002: Compared to manual palpation and ultrasound. VerTouch significantly reduced redirections and reinsertions compared to both palpation and ultrasound.
(2) Non-clinical performance testing must demonstrate that the device performs as intended under anticipated conditions of use and must evaluate the accuracy, repeatability, and resolution of imaging of the underlying anatomy representative of intended use populations.	Benchtop performance testing on custom-built lumbar phantoms (average and high-BMI) validated accuracy and repeatability of the imaging technology (piezoelectric sensor array, software, image resolution). Testing addressed resolution characterization, contrast sensitivity, and target detectability in challenging high-BMI scenarios.
(3) Human factors testing must demonstrate that the user can correctly use the device based solely on the directions for use.	The VerTouch Summative Usability Validation 2 study (N=68 participants, including at least 15 from each primary user group) was conducted in simulated hospital rooms. Participants performed simulated use tasks with VerTouch on custom phantom models after training. Identified potential harms (sterile sleeve caught, marker not reaching skin) were mitigated by updating IFU and a minor device form factor change, respectively. Additional usability validation was not required after the form factor change.
(4) Software verification, validation, and hazard analysis must be performed. Validation testing must verify and validate proper image construction.	Software documentation was reviewed per FDA Guidance, found to have Moderate Level of Concern. Included Software Description, Hazard Analysis, Requirements Specification, Architecture Design Chart, Design Specification, Traceability Analysis, Development Environment Description, Verification and Validation Documentation, Revision Level History, and Unresolved Anomalies. The software documentation was found to be acceptable.
(5) Electromagnetic compatibility and electrical, mechanical, and thermal safety testing must be performed.	Electrical safety and electromagnetic compatibility testing performed per IEC 60601-1-12 and IEC 60601-1-2. Test results support electrical safety and electromagnetic compatibility.
(6) The patient-contacting components of the device must be demonstrated to be biocompatible.	Biocompatibility evaluation was performed. (Specific details not provided in this extract, but it is listed as a mitigation measure).
(7) Performance testing must demonstrate the sterility of any device components intended to be labeled sterile.	Sterilization validation was performed. (Specific details not provided in this extract, but it is listed as a mitigation measure).
(8) Labeling must include a warning that risks inherent to neuraxial procedures under standard of care conditions are still present with use of the device.	Labeling includes a warning about over-reliance on device output and that risks inherent to neuraxial procedures under standard of care conditions are still present. Instructions for Use include appropriate warnings and cautions.

Study Details

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

STU-001:
- Sample Size (Test Set): 90 subjects randomized (42 VerTouch, 48 Palpation for ITT).
- Data Provenance: Multicenter, prospective, randomized controlled trial. Sites included Rhode Island Hospital, University of Texas at Houston, and Medical University of South Carolina (USA).
STU-002:
- Sample Size (Test Set): 81 subjects (26 Palpation, 26 Ultrasound, 29 VerTouch).
- Data Provenance: Single-center, prospective, randomized clinical trial. Conducted at Prentice Women's Hospital at Northwestern Memorial (NMH) in Chicago, IL (USA).

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience)

STU-001: The ground truth for successful spinal canal access was determined clinically by the practicing providers performing the procedures, based on procedure-specific methods (e.g., CSF return for LP, sensory blockade for spinal anesthesia). There isn't a separate panel of experts establishing ground truth for the test set data after the procedure for assessment. Rather, the outcome was determined by the treating provider.
STU-002: For redirections and reinsertions, three blinded observers independently reviewed all insertion videos to count these events. Their specific qualifications (e.g., years of experience, medical specialty) are not detailed in the provided text, but it states they were "clinically trained device expert" for trainer, implying clinical expertise.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set

STU-001: Not explicitly stated as a formal adjudication by an independent panel for the primary endpoints. The outcomes were recorded by a "designated study observer" within each setting.
STU-002: For redirections and reinsertions, "three blinded observers who independently counted redirections and reinsertions of each procedure" were used. The interclass correlation coefficients (ICC) were reported (0.87 for redirections, 0.97 for reinsertion), indicating agreement, but a formal adjudication process (e.g., 2+1 if they disagreed) is not described.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

A formal MRMC study as typically defined for image interpretation (i.e., multiple readers interpreting multiple cases with and without AI) was not explicitly described for assessing the VerTouch device's primary function in identifying insertion sites. The clinical studies (STU-001 and STU-002) compared the procedure outcomes when using VerTouch assistance versus standard palpation/ultrasound, directly involving the clinicians performing the procedure.
While STU-002 used three blinded observers to count redirections/reinsertions from videos, this was for outcome measurement reliability, not a traditional MRMC study on diagnostic interpretation.
The device's effect size can be inferred from the differences in mean numbers of insertion attempts, redirections, and reinsertions observed in the clinical studies when comparing VerTouch to control methods.
- STU-001 (PP population): VerTouch mean #insertions = 1.2, Palpation mean #insertions = 2.1. (Difference: 0.9 attempts reduction)
- STU-002: VerTouch mean #redirections = 0.72, Palpation = 2.35, Ultrasound = 2.81. (VerTouch reduced redirections by ~1.63 compared to palpation and ~2.09 compared to ultrasound). VerTouch mean #reinsertions = 0.11, Palpation = 0.50, Ultrasound = 0.96. (VerTouch reduced reinsertions by ~0.39 compared to palpation and ~0.85 compared to ultrasound).

6. If a standalone (i.e., algorithm only without human-in-the loop performance) was done

Not explicitly described. The VerTouch device is designed as an assistant for human users (clinicians) to aid in localization and marking. It provides a tactile map and guidance, rather than making an autonomous diagnostic decision. Therefore, a standalone performance of "algorithm only" would be less relevant as it operates with the human user's interpretation of the pressure map to identify the interspinous space. Benchtop testing did validate the imaging technology's accuracy and resolution, but this is a component test, not a standalone performance of the "AI" identifying a TIS without human input.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

Clinical Outcomes Data: For STU-001, the primary ground truth was confirmation of successful spinal canal access (e.g., CSF return, sensory blockade). This is a direct measure of procedure success. For STU-002, the ground truth for primary/secondary endpoints like redirections and reinsertions was established by blinded expert review of recorded videos. The ultimate goal was also procedural success, which was 100% in STU-002, suggesting effective TIS identification.
Benchtop Testing: Used custom-built lumbar phantoms to provide ground truth for evaluating the accuracy, repeatability, and resolution of the imaging technology against known anatomical features.

8. The sample size for the training set

The provided text does not specify a training set size for the AI component of the VerTouch device. It describes the software verification and validation, and benchtop phantom testing which would contribute to the device's development and internal validation, but not a distinct "training set" like in typical machine learning studies. The device uses a "calibrated, piezoresistive sensor array" and "software" to acquire, process, and display a TactoMap, but the process of "identifying a blue region along the midline corresponding to an interspinous space" seems to be user-driven interpretation aided by the device's visual output, rather than an AI automatically identifying and suggesting the space.

9. How the ground truth for the training set was established

Given that a specific "training set" for AI is not detailed, the establishment of ground truth for such a set is also not described. If we interpret "training" in a broader sense for the device's internal calibration and development (e.g., for the sensor array and image construction), then the benchtop phantom testing served to establish ground truth for this purpose. The phantoms were "custom-built" with controlled variations in tissue geometries, spinous process size, and depth, allowing for objective evaluation of the device's imaging performance against known physical properties.

Ask a Question

Ask a specific question about this device

Page 1 of 1