Harmonizing visual readings of tau PET tracers ‐ the HEAD study
Bibliographic record
Abstract
BACKGROUND: F]MK-6240 (MK) and provide preliminary data toward a unified visual reading approach for all tau-PET tracers using a head-to-head dataset. METHODS: The study design and development plans are illustrated in Figure 1. To evaluate previously published visual reading methods, two blinded raters conducted FTP and MK visual reads on 340 participants from the HEAD study. A unified visual reading method was developed and tested on 101 participants, using a composite of adjusted Braak regions to determine 3 stages of severity while harmonizing between tracers. For external validation, the method will be tested on four additional cohorts, each using one of the following tracers: FTP, MK, PI, or RO. RESULTS: Inter-rater agreement using previous methods (Figure 2A) showed high Cohen's kappa values (0.71-0.77) for the low and high tau burden categories in both FTP and MK visual reads. However, agreement was substantially lower in the non-AD-like category, leading to overall agreement rates of 0.55 for FTP and 0.69 for MK. In contrast, the method developed here improved inter-rater agreement across all categories (0.87-0.96) and introduced a moderate tau (0.79-0.83) burden group (Figure 2B), leading to higher overall agreement rates of 0.81 for FTP and 0.87 for MK (Figure 2C). For inter-tracer agreement (Figure 3), previous methods resulted in discordant visual reads for FTP and MK in 34.4% of cases for rater 1 and 24.7% for rater 2. The developed method reduced this disagreement to 11.9% (rater 1) and 15.8% (rater 2), demonstrating improved consistency across different tracers. CONCLUSION: This study demonstrates that previously published visual reading methods produced varying classifications depending on the tracer used. The method proposed here improved inter-rater and inter-tracer visual read agreements by developing a unified, tracer-agnostic approach easy to be used. Importantly, this clinician-friendly method has the potential for widespread adoption, offering a single harmonized visual reading technique for all tau-PET tracers.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.039 | 0.065 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.001 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".