U.S. flag An official website of the United States government

U.S. Food and Drug Administration

The Linear Fit Tool (LiFT) for Orthopedic Device Testing

Catalog of Regulatory Science Tools to Help Assess New Medical Devices 

This regulatory science tool (RST) is a MATLAB script that automates the determination of stiffness from the slope of a linear region from mechanical test data using an algorithm that is in compliance with ASTM E3076-18 [1]. Specifically, it analyzes test data (i.e., force-displacement curve or torque-angle curve) and then generates output parameters including bending and torsional stiffness typically requested in the preclinical mechanical performance test standards, ASTM F3574 [2] and ASTM F2267 [3].

Technical Description

This RST is a MATLAB live script that enables automated identification and visualization of the linear portion of test data for calculation of the stiffness (i.e., bending stiffness and torsion stiffness) from a force-displacement curve or torque-angle curve typically found in the preclinical mechanical performance test data of medical devices. Specifically, this RST is intended for use with the test methods outlined in ASTM F3574 “Standard Test Methods for Sacroiliac Joint Fusion Devices” and ASTM F2267 “Standard Test Method for Measuring Load-Induced Subsidence of Intervertebral Body Fusion Device Under Static Axial Compression”. The fundamental algorithm for determining slope of a linear region is directly replicated from the ASTM E3076 standard [1] titled “Standard Practice for Determination of the Slope in the Linear Region of a Test Record”. 

The tool analyzes the test data and generates output parameters as required by the above-mentioned standards along with associated linear fit metrics and plots. The output from the tool includes a tabular data summary of all processed samples, a tabular quality metrics that consistent with ASTM E3076, and a report file which includes the linear fit plot for each sample and an integrated plot including the raw data from all processed samples. This tool provides an alternative way for medical device manufacturers and test houses to calculate the preclinical performance data of transverse sacroiliac joint fusion devices and spinal intervertebral body cage subsidence.

Intended Purpose 

The intended purpose of this RST is to automate the standardized method in ASTM E3076, and to determine the stiffness from the linear region of the mechanical test data collected from tests described in ASTM F3574 or ASTM F2267 standards. Such tests are commonly used during the development phase of transverse sacroiliac joint fusion devices (eg product code: OUR) and intervertebral body fusion devices (eg product codes: MAX, ODP. OVD, OVE, PHM). Test curves stemming from the mechanical bench tests described in both standards require determination of stiffness from the slope of the linear region, which has traditionally been calculated manually or with custom processes specific to each manufacturer. This RST will give users an automated approach that is intended to provide increased consistency. 

Testing

The LiFT RST was tested using the benchmark data set (E3076-DS1(2018) File 01) provided by ASTM E3076 to verify computer implementation of the algorithm in E3076. The results were compared to those provided in E3076 Appendix X1 in tabular format. The results generated using LiFT RST exactly matched those provided in E3076 Appendix X1. 

Further, outputs from this RST were compared to outputs from two similar publicly-available tools, pySDAR and NIST’s Excel sheet macro SDAR linear fitting.xlsm [4, 5]. Using the validation dataset as provided by the pySDAR[4], this comparison showed that the “Slope Determination Results” per E3076 were within 1.20% across these three tools, thereby showing consistency of the LiFT RST.

Limitations

This RST is a MATLAB live script and therefore requires MATLAB (2016b or later version) installed to run. Additionally, this tool is dependent on the test data quality and the tool may not perform correctly or ideally if the data have no apparent linear region or are excessively noisy. Criteria for evaluation of quality metric parameters and the associated acceptance criteria are provided in E3076 Appendix X1. User should refer to E3076 and the listed acceptance criteria in Appendix X1 to evaluate their results. Additional information related to the limitations of this tool can be found in the user manual and ASTM E3076 [1].

Supporting Documentation

  1. E3076 - 18: Standard Practice for Determination of the Slope in the Linear Region of a Test Record. American Society for Testing and Materials
  2. F3574 - 22: Standard Test Methods for Sacroiliac Joint Fusion Devices. American Society for Testing and Materials
  3. F2267 - 22: Standard Test Method for Measuring Load-Induced Subsidence of Intervertebral Body Fusion Device Under Static Axial Compression. American Society for Testing and Materials
  4. Cajon Gonzales, M., Kimberly E, Nathan Madey, Peter McKeighan, Robert W Kozar, Stephen Graham. pySDAR: An implmenetation of the ASTM E3076-18 Standard. 2024  [cited 2024 07/16].
  5. Lucon, E., Use and validation of the slope determination by the analysis of residuals (SDAR) algorithm. 2019: US Department of Commerce, National Institute of Standards and Technology.

Contact

Tool Reference