Site-related variance is generally small (typically <10%) for MRI measures in the AMP SCZ multi-site neuroimaging protocol, while participant variance dominates for structural and fMRI SNR/CNR (40-76%), but platform-related variance is substantial for diffusion scan SNR/CNR (>55%) and spatial smoothness.
In the AMP SCZ multi-site neuroimaging study of individuals at clinical high risk for psychosis, site-related variance in MRI measures was small (<10%). Participant variance was the largest component for structural and fMRI signal-to-noise and contrast-to-noise ratios (40-76%), but platform (scanner model) variance dominated for diffusion imaging SNR/CNR (>55%) and spatial smoothness due to hardware and vendor differences.
This article is automatically generated from the structured evidence profile behind the claim above. Scores reflect the quality and quantity of available research, not clinical advice.
Site-related variance is generally small (typically <10%) for MRI measures in the AMP SCZ multi-site neuroimaging protocol, while participant variance dominates for structural and fMRI SNR/CNR (40-76%), but platform-related variance is substantial for diffusion scan SNR/CNR (>55%) and spatial smoothness. The current body of evidence comprises 1 study. EvidenceHub rates the overall confidence at 32/100 (low).
The Claim
Site-related variance is generally small (typically <10%) for MRI measures in the AMP SCZ multi-site neuroimaging protocol, while participant variance dominates for structural and fMRI SNR/CNR (40-76%), but platform-related variance is substantial for diffusion scan SNR/CNR (>55%) and spatial smoothness.
This conclusion is most relevant to: Individuals at clinical high risk (CHR) for developing psychosis (part of the AMP SCZ cohort).
What the Research Shows
The conclusion draws on 1 linked study. Highlights from the cited literature:
- ▸The MR neuroimaging protocol for the Accelerating Medicines Partnership® Schizophrenia Program. (Schizophrenia (Heidelberg, Germany), 2026) —
How It Works
The proposed biological pathway:
- ▸Multi-site MRI data collected across different scanners
- ▸Variance decomposition performed for SNR/CNR and smoothness
- ▸Participant variance largest for structural/fMRI (40-76%)
- ▸Platform variance largest for diffusion (>55%) and smoothness
Who Might Benefit
Evidence fit by population:
- ▸Individuals at clinical high risk (CHR) for developing psychosis (part of the AMP SCZ cohort)
Recommended Dose
N/A
Limitations & Caveats
Important context when interpreting this evidence:
- ▸Preliminary variance component analyses only; full cohort results not yet reported
- ▸Platform-related variance in diffusion scans may limit cross-vendor generalizability
- ▸Spatial smoothness differences between vendors are difficult to control
Frequently Asked Questions
What is the main finding about site-related variance in this study?▼
Site-related variance was generally small, typically less than 10%.
Which MRI measure showed the largest platform-related variance?▼
Diffusion scan SNR/CNR showed substantial platform-related variance (>55%), and spatial smoothness also had large platform-related variance.
What population was studied in the AMP SCZ neuroimaging protocol?▼
Individuals at clinical high risk (CHR) for developing psychosis.
Why is platform variance important for multi-site MRI studies?▼
High platform variance can reduce the ability to combine data across different scanner models and may confound results if not properly accounted for.
References
- 1.Harms MP, Cho KK, Anticevic A, Bolo NR, Bouix S, Campbell D, Cannon TD, Cecchi G, Goncalves M, Haidar A, Hughes DE, Izyurov I, John O, Kapur T, Kim N, Kotler E, Kubicki M, Kuperman JM, Laulette K, Lindberg U, Markiewicz C, Ning L, Poldrack RA, Rathi Y, Romo PA, Tamayo Z, Wannan C, Wickham A, Yassin W, Zhou JH, Addington J, Alameda L, Arango C, Breitborde NJK, Broome MR, Cadenhead KS, Calkins ME, Chen EYH, Choi J, Conus P, Corcoran CM, Cornblatt BA, Diaz-Caneja CM, Ellman LM, Fusar-Poli P, Gaspar PA, Gerber C, Glenthøj LB, Horton LE, Hui CLM, Kambeitz J, Kambeitz-Ilankovic L, Keshavan MS, Kim SW, Koutsouleris N, Kwon JS, Langbein K, Mamah D, Mathalon DH, Mittal VA, Nordentoft M, Pearlson GD, Perez J, Perkins DO, Powers AR, Rogers J, Sabb FW, Schiffman J, Shah JL, Silverstein SM, Smesny S, Stone WS, Strauss GP, Thompson JL, Upthegrove R, Verma SK, Wang J, Wolf DH, Kahn RS, Kane JM, McGorry PD, Nelson B, Woods SW, Shenton ME, Wood SJ, Bearden CE, Pasternak O. “The MR neuroimaging protocol for the Accelerating Medicines Partnership® Schizophrenia Program..” Schizophrenia (Heidelberg, Germany), 2026. PMID: 40175382 DOI: 10.1038/s41537-025-00581-6