Multi Center Clinical Trial Confirms FMTVDM © ℗ MPI in Seven Modern Clinical Laboratories in the USA and Asia. Artificial Intelligence (AI) with True Quantification

,


Background
The foundational work of nuclear cardiology began with Blumgart's 1925 study of circulation time. Blumgart focused on quantitatively measuring isotope change over time. Since then Nuclear Cardiology has focused on qualitative image interpretation resulting in sensitivity and specificity issues. A demand for reducing patient radiation exposure has additionally resulted in a call for "stress-first/stress-only/ stress-stress" imaging.
A preliminary VA study demonstrated FMTVDM ©℗ "stress-stress" imaging statistically out performed conventional two-injection MPI "stress-rest" approach, while lowering patient mSv radiation exposure [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15] with FMTVDM ©℗ resulting in statistically improved clinical outcomes, providing the only True quantitative AI method [16][17][18] removing the qualitative errors made using "stress-rest" MPI methods [19,20], and reducing the amount of patient exposure during MPI as seen in Figure 1. This study examined the ability to implement FMTVDM ©℗ in seven different centers in the USA and Asia. .75 mSv per study (less than Coronary CT or CT Angiography) while conventional studies using 2-injections of isotope yield between 8 and 25 mSv patient exposure, depending upon the isotope and dose used. As seen in the top panel, the 2 injection approaches and Coronary CT/Angiography provide more radiation exposure than the annual background radiation. 5-mSv is frequently considered the standard accepted limit, below which the risk of inducing cancer is considered minimal.

FMTVDM ©℗
Following the protocol detailed in FMTVDM ©℗ 401 men and women were studied as detailed previously [15] at seven different Centers-of-Excellence (COE).

Coronary arteriograms
Following standard procedures previously detailed [10][11][12][13][14] coronary angiography was performed, %DS and QCA results were obtained. Initial visual reporting of percent diameter stenosis (%DS) was determined only after clinicians had undergone extensive training necessary to reduce the qualitative errors [21].

Calculation of %DS from isotope redistribution
The derived isotope redistribution was then calculated using pixelby-pixel isotope measurement to determine redistribution wash-in (initial delay in isotope uptake) as described previously [4][5][6][7] and redistribution washout (failure to retain isotope) or expected (normal) redistribution as shown in Figure 2. Calculation of Quadratic/Fleming Coronary Flow Reserve (QCFR © , FCFR © ) from Isotope Redistribution determined %DS: The AI proprietary equations and process derived from measured human coronary artery analysis [10][11][12][13][14] followed the FMTVDM ©℗ protocol (p=3.8 x 10 -8 ) directly introducing those measurements into proprietary equations protected by patent #9566037 automatically deriving CFR. The measurements and subsequent derivation of CFR (QCFR/FCFR) were done at multiple times and are NOT dependent on isotope, camera type (SPECT, PET), camera company or operator availability.

Statistical analysis
Comparison of FMTVDM ©℗ measured/quantified isotope redistribution and subsequent AI calculated Quadratic/Fleming Coronary Flow Reserve (QCFR © , aka. FCFR © ) Using the FMTVDM ©℗ and proprietary equations (QCFR © , aka. FCFR © ), 401 individuals were studied and the AI quantified results from the FMTVDM ©℗ Proprietary equations derived (QCFR © /FCFR © ) were compared with QCA derived outcomes. Residual plots were analyzed for data bias. The comparison of QCFR © (aka FCFR © ) derived from the FMTVDM ©℗ with QCA measured CFR was performed using regression analysis and best fit linear regression modeling.

Results
Following assessment of each clinical facility to determine adequately trained personnel, operators at 7-Centers of Excellence (COE) were recruited to conduct "stress-first/stress only/stress-stress" FMTVDM Imaging. Outcomes were independent of isotope, camera type, Camera Company or "stressor" employed. Outcome analysis was obtained using AI measured and quantitatively derived CFR (QCFR © / FCFR © ) results using FMTVDM ©℗ measurements and proprietary equations, compared with that obtained from direct QCA. The results from the 7COE yielded regression analysis of y=(0.8758•x)+0.4291, where y=the QCFR and x=the QCA measured CFR. The R2 value 0.87582 demonstrated the 7-COE accurately, consistently and reproducibly clinically implemented FMTVDM ©℗ to derive by AI, QCFR © /FCFR © in all 401 patients.

Discussion
The first step in the demonstration of a new method for diagnostically studying a health problem is to compare the new test with the currently accepted approach [4][5][6][7]. If the results are promising investigators conduct a single center clinical trial [15]. If the newer method provides better clinical and diagnostic outcomes than the currently accepted method [4][5][6][7]15] and provides a significant advancement for the field [16][17][18][19][20], a multicenter trial is initiated to demonstrate that clinicians with various backgrounds, in various clinical settings are able to accurately, consistently and reproducibly utilize the new diagnostic test and obtain the expected outcomes.
In this study, seven COE from the USA and Asia independently conducted clinical MPI studies using FMTVDM ©℗ using the first true quantification method for nuclear imaging studies and AI. The outcomes demonstrated the ability of each COE to accurately, consistently and reproducibly use FMTVDM ©℗ to diagnose the level of clinical disease without error, while saving time, reducing healthcare costs and markedly decreasing patient radiation exposure with potential associated risks.

Conclusions
FMTVDM ©℗ utilizing AI accurately, consistently and reproducibly provided true quantification of CAD in 7-COE in the USA and Asia. The utilization of AI FMTVDM ©℗ provided accurate, consistent and reproducible quantitative CFR beginning with the true quantification of isotope redistribution using the patented proprietary equations independent of isotope, camera type (SPECT, PET), camera company, operator variables or site location. FMTVDM ©℗ provides the first truly quantitative, truly AI Nuclear Imaging of CAD and CFR introducing a new era of nuclear cardiology. The use of FMTVDM ©℗ provides the first ever "Quantitative" AI method able to differentiate tissue vascularity and metabolism and as such provides an evolutionary quantum leap forward for the fields of nuclear cardiology and nuclear medicine.