nabu.science
MethodologyAboutContact

 

  • Snapshot
  • Key Claims
  • Strengths & Limitations
  • Commentary
  • From the Authors
  • Author Note / Field Feedback
  • Dimension breakdown
  • Readiness
Report #NBU-F967162A·Framework v5.9·Pipeline v0.4·Evaluated 08 Oct 2026·OECD: 3.2 Clinical medicine·Subfield: Psychiatry and Mental health·Basis: Full text·Type: Diagnostic accuracy study

Investigating AD and non-AD [18F]flortaucipir distribution patterns in a memory clinic cohort.

Piva et al.Piva, Boccalini, Ashton et al.·2026·European journal of nuclear medicine and molecular imaging·10.1007/s00259-026-07995-z ↗

Primary claim

[18F]flortaucipir-PET DPs derived from SSM/PCA provide information on both A and T status

Tentative · the design mainly distinguishes A+T+ from A−T± and does not independently identify amyloid positivity
In brief

A useful memory-clinic replication shows tau-PET patterns can separate A+T+ cases, but the design does not establish independent amyloid-status prediction.

ReliabilityCaveats4 of 4 checks completeRead the 3 entries ↓
Fragile — the claim's weakest requirement is that the effect exists at all. Experimental. Every point below is traceable to a quoted passage, but this forecast has not yet been validated against known replication outcomes.
Abstract

PURPOSE: Amyloid (A) deposition represents a specific pathological hallmark of Alzheimer's disease (AD). Clinical diagnostic protocols frequently rely on the combined use of amyloid (A)- and tau (T)- positron emission tomography (PET) imaging to distinguish AD from non-amyloid-associated neurodegenerative conditions. Many neurodegenerative disorders are characterized by distinct tauopathies, which result in different [1⁸F]flortaucipir T-PET topographic patterns. The aim of this study was to investigate, in a memory clinic cohort, whether [18F]flortaucipir distribution patterns derived from scaled subprofile modelling using principal component analysis (SSM/PCA) could differentiate between A + and A - individuals, potentially eliminating the need for an A-PET.

METHODS: A total of 81 subjects were included in this study: 40 with AD characterized by both A and T accumulation (A + T +), 13 with suspected non-Alzheimer's pathology (A - T +), and 28 without A and T abnormality (A - T -). SSM/PCA was applied on [18F]flortaucipir images to identify two disease patterns (DPs), representing amyloid-positive (A + DP) and amyloid-negative (A- DP) accumulation. Each subject's imaging data was then compared to these DPs and the resulting similarity scores were used as inputs for three machine learning models - Support Vector Machine, Random Forest, and Multi-Layer Perceptron - to predict amyloid status. Finally, Spearman correlation analyses were conducted between SSM/PCA-derived scores and plasma biomarkers levels.

RESULTS: The two identified disease patterns exhibited distinct PET signatures consistent with current disease knowledge. Their representativeness was further supported by significant correlations with plasma biomarkers in nearly all cases. All machine learning models confirmed the capacity of the A + DP to predict AD pathology, correctly classifying at least 36 out of 40 cases. In contrast, the A - DP showed limited discriminative power for non-AD conditions, as indicated by its higher misclassification rate (best performance: 7 correct out of 13).

CONCLUSION: SSM/PCA applied on [18F]flortaucipir-PET imaging provide information on both A and T status, supporting its potential use as first and possibly sole examination in the investigation of suspected AD. While for non-AD cases, A-PET scans are still recommended.

Key Claims

Primary finding · Tentative

[18F]flortaucipir-PET DPs derived from SSM/PCA provide information on both A and T status

Fragile — the claim's weakest requirement is that the effect exists at all. Experimental. Every point below is traceable to a quoted passage, but this forecast has not yet been validated against known replication outcomes.

the design mainly distinguishes A+T+ from A−T± and does not independently identify amyloid positivity

Conclusion
Quote verified
Secondary · Tentative

all machine learning models confirmed the capacity of the A+DP to predict AD pathology, correctly classifying at least 36 out of 40 cases

the reported count is unreconciled with the multiclass metrics and arises from label-informed same-cohort evaluation

Abstract Results
Quote verified
Secondary · Supported

the A−DP showed limited discriminative power for non-AD conditions

the direction is consistent across reported overlap and classifier results despite the small heterogeneous 13-person subgroup

Abstract Results; Discussion
Quote verified
Secondary · Tentative

Their representativeness was further supported by significant correlations with plasma biomarkers in nearly all cases

correlations use only 43 participants and include two explicitly non-significant A−DP comparisons

Abstract Results; Results §Correlation Between SSM/PCA Scores and Plasma Biomarkers
Quote verified
Secondary · Tentative

The MLP achieved the highest overall performance (balanced accuracy = 0.81, weighted-F1 = 0.87, macro-F1 = 0.81).

the point estimates are reported, but the analytical denominator and same-dataset validation limit their reliability

Results §Amyloid Status Prediction
Quote verified

Derived from the full evaluation — not a separate score.

Strengths · Limitations

Strengths

#contribution

Clinically relevant heterogeneous replication cohort

The analysis tests earlier SSM/PCA findings in a memory-clinic cohort spanning cognitive stages and includes a heterogeneous A−T+ group, exposing limitations that a narrowly defined non-AD cohort may miss.

↳ Methods §Study Cohort; Discussion, comparison with Hammes et al.

#methodological rigour

Detailed multistage imaging analysis pipeline

The Methods describe image processing, SSM/PCA construction, LOOCV subject scoring, three tuned classifiers and imbalance-robust metrics. A repeated undersampling sensitivity analysis further examines class imbalance.

↳ Methods §Multivariate Analysis and §Amyloid Status Prediction; Results §Amyloid Status Prediction

#positioning

Candid negative non-AD finding

The Results and Discussion report substantial overlap and limited discrimination for the A− pattern, then relate this to the etiological heterogeneity of the A−T+ group.

↳ Results §Multivariate Analysis and §Amyloid Status Prediction; Discussion, A−T+ heterogeneity paragraph

Limitations

#methodological rigour

Cohort denominator and target remain inconsistent

The Abstract states n=81, while AT Profiles Computation and Table 1 include 17 additional A+T− participants. The classifier defines A+T+ rather than all amyloid-positive participants as positive, obscuring which population and target the reported performance represents.

↳ Abstract; Methods §AT Profiles Computation and §Amyloid Status Prediction; Table 1

#reporting

Headline performance counts lack reconciliation

The Abstract's “36 out of 40” and “7 out of 13” counts are not presented in the Results text or reconciled with the multiclass balanced accuracy and F1 metrics. This is a material internal-coherence concern for the central performance claims.

↳ Abstract Results; Results §Amyloid Status Prediction; Figure 4

#positioning

Clinical conclusion exceeds demonstrated inference

The Discussion states that apparent A+ identification is driven by the cohort's association between T+ and A+, yet the Conclusion proposes tau PET as a possibly sole examination for suspected AD. A+T− scores resembling A−T− further limit the amyloid-specific interpretation.

↳ Results §Multivariate Analysis; Discussion, classifier interpretation paragraph; Conclusion

Does this match your read? Add your assessment ↓

Nabu Commentary

The paper contributes a useful replication because it tests SSM/PCA in a heterogeneous memory-clinic cohort and reports the limited performance of the A− pattern rather than suppressing that result. Methodological credit comes from the detailed imaging pipeline, LOOCV scoring, multiple classifiers and class-imbalance sensitivity analysis. The evaluation is nevertheless constrained by the mismatch between the claimed amyloid target and the A+T+ versus A−T± labels, alongside unclear inclusion of the 17 A+T− participants. The Abstract's unreconciled per-class counts and the Conclusion's sole-examination language further exceed what the reported and acknowledged limitations support.

Check another paper

Free · About 5 minutes · Research preview

Author Note / Field Feedback

Nabu’s assessment, alongside the field’s view.

Are you an author of this paper?

Dimension breakdown

Study Quality

Limited2.5

Confidence medium
Contribution
3.0

The study provides a modest replication and extension of prior SSM/PCA work in a heterogeneous memory-clinic cohort, including a clinically informative negative result for the A− pattern. It does not establish the stronger proposition that tau PET independently identifies amyloid status.

Introduction; Discussion, paragraphs comparing the findings with Hammes et al.; Conclusion

“the A−DP demonstrated limited discriminative power for non-AD condition”

Methodological Rigour
2.0

The imaging pipeline, LOOCV scoring, multiple classifiers and imbalance sensitivity analysis are described, but the diagnostic target excludes A+T− cases and lacks simple tau-SUVR or visual-read comparators. Selected sampling and same-dataset tuning further constrain the performance estimates.

Methods §Study Cohort, §Multivariate Analysis, and §Amyloid Status Prediction; Discussion §Limitations

“some degree of optimistic bias cannot be excluded”

Reporting
2.7

The paper's overall progression is followable, but the Abstract's per-class counts are not reconciled with the body metrics. The sole-examination language also exceeds the Discussion's explanation that apparent A+ identification reflects T+/A+ co-occurrence.

Abstract Results and Conclusion; Results §Amyloid Status Prediction; Discussion, paragraph following classifier performance

“not directly driven by amyloid deposition itself”

Positioning
2.8

The Discussion directly compares the cohort and findings with the closest prior SSM/PCA study and explains the heterogeneity of A−T+ disease. Contextual credit is limited because acknowledged dependence on cohort co-occurrence and external validation is not carried through consistently to the final clinical recommendation.

Discussion, Hammes comparison, Limitations, and Conclusion

“external validation in an independent cohort will be necessary”

Lower confidence on Methodological Rigour — domain match limited.

Reliability

Caveats4 of 4 checks


Results checkNoted
  • Their representativeness was further supported by significant correlations with plasma biomarkers in nearly all cases: "Significant correlations were observed across all comparisons (p < 0.01)." (Results, Correlation between SSM/PCA scores and plasma biomarkers) vs "No significant correlations were found between the A − DP scores and either Aβ42/Aβ40 (rs = 0.02, p = 0.92) or p-tau231 (rs = 0.30, p = 0.05)." (Results, Correlation between SSM/PCA scores and plasma biomarkers). The five biomarkers paired with two DP scores give ten comparisons, of which the text identifies two as nonsignificant, so the universal statement must exclude those two while the key claim of significance in nearly all cases remains consistent. → Conclusion holds
Reporting and transparencyNoted

The 2D classifier's stated amyloid-status target is not explicitly reconciled with its multiclass output. The Table 1 caption omits the reported age difference, and acquisition details are supplied for florbetapir but not flutemetamol. These are reporting issues, not demonstrated changes to the performance results.

Ethics approval: declared · Data availability: not stated · Code availability: not stated

Reference IntegrityNoted

34 references in manuscript 34 of 34 checkable references found in an index the reference list may be incompletely parsed; counts are indicative

References not verified automatically are resolved manually within 24 hours.

Post-Publication RecordClean

No retraction or expression of concern found in Crossref (includes Retraction Watch).

Last checked: 17 Oct 2026

Sources: Crossref (incl. Retraction Watch) ✓

Readiness

Where this paper’s evidence sits on the path from initial observation to real-world use.

Mechanistic

The study is an early single-centre validation rather than a deployment-ready diagnostic evaluation. The authors explicitly require independent-cohort testing before clinical generalization.

Discussion §Limitations and Future Research

“independent cohort testing and evaluation in new patients are needed”

Report an issue →

AI-generated, human-governed. Something look off? Contact us to request a review.

Keep reading

Related evaluations

Other papers Nabu has evaluated blind, starting with the same field.

  1. Decreased awareness of cognitive decline is associated with multimodal Alzheimer's disease biomarkers in cognitively unimpaired individuals

    2026 · Clinical medicine

    Primary claim

    decreased ACD was associated with greater AD pathological burden

    Supported
    Study quality
    Sound3.2
    Reliability
    Caveats
  2. The potential of low-field MRI for global dementia care

    2026 · Clinical medicine

    Primary claim

    Studies demonstrate reliable detection of core Alzheimer disease markers, including hippocampal atrophy and white matter hyperintensities, on portable low-field systems, with promising correspondence to conventional high-field MRI findings.

    Tentative
    Study quality
    Sound3.7
    Reliability
    Caveats
  3. UK Biobank at 20 - a growing, global resource for dementia research

    2026 · Clinical medicine

    Study quality
    Sound3.2
    Reliability
    Caveats

How the rubric works →