Integrity Screening Report: Engineering Dual Active Sites and Defect Structure in Nanozymes to Reprogram Jawbone Microenvironment for Osteoradionecrosis Therapy (Adv. Sci., 2024)

Academic fraud report · Geng Detector

Summary

This screening report evaluates the article published in Advanced Science (DOI: 10.1002/advs.202413215) for potential image-related irregularities. The overall verdict is HIGHLY SUSPICIOUS. Four findings were flagged: (1) apparent reuse of Live/Dead fluorescence images across three morphologically distinct cell types (macrophages, endothelial cells, HBMSCs) and multiple time points in Figure 6a–c, with shared background texture that contradicts biological expectation; (2) near-identical SEM micrographs for the Control and NIR-only groups of S. aureus in Figure 5g, despite these being independent treatment conditions; (3) striking morphological overlap between Figure 2a (TEM) and Figure 2d (HAADF-STEM), raising questions about whether different imaging modes of the same particle were genuinely captured; (4) unusually uniform, perfectly aligned β-actin loading-control bands in Figure 7d/f, suggestive of digital editing. All findings rely on visual inspection of the published PDF; no raw TIFFs or metadata were available. Confidence is moderate-to-high for findings 1–2 and lower for 3–4. Final determination requires author-provided originals and institutional investigation.

Verdict

🟠 Highly Suspicious — Multiple image irregularities identified across four figures; visual evidence is consistent with possible image reuse and/or digital manipulation, but cannot be conclusively confirmed without raw image files and metadata.

Key findings

Finding 1 (Critical): Reuse of Live/Dead staining images across distinct cell types (macrophages, ECs, HBMSCs) and time points (0, 24, 48 h) in Figure 6a–c; shared background noise pattern is biologically implausible.
Finding 2 (High): Near-identical SEM images for the Control and NIR (1.5 W cm⁻²) groups of *S. aureus* in Figure 5g, despite the authors' own statement that CFU counts did not differ between these groups and that NIR alone has no antibacterial activity.
Finding 3 (Moderate): Marked morphological overlap between Figure 2a (TEM) and Figure 2d (HAADF-STEM); different imaging modes should typically convey complementary rather than identical morphological information.
Finding 4 (High): Suspiciously uniform, evenly aligned β-actin bands in Figure 7d/f, lacking natural width/intensity variation and typical transfer artifacts.

Evidence highlights

Figure 6a–c: Identical background textures and cell distribution patterns across three cell types that should exhibit fundamentally different morphologies (suspension/round macrophages vs. adherent spindle-shaped ECs/HBMSCs).
Figure 5g: Pixel-level similarity in bacterial arrangement and substrate texture between Control and NIR-only groups; manuscript text confirms these groups were expected to behave similarly biologically, but visual identity remains anomalous.
Figure 2a vs. 2d: Overlapping nanosheet outlines and edge features in TEM and HAADF-STEM, which would normally be paired with complementary EDS mapping data to justify reuse.
Figure 7d/f: β-actin bands exhibit implausibly perfect alignment and uniform edge sharpness, inconsistent with typical Western blot variability.

Notes

All assessments are based solely on the published PDF (ADVS-12-2413215.pdf); no original TIFF files, RAW camera outputs, or acquisition metadata were examined.
DOI: 10.1002/advs.202413215 is preserved for indexing and PubPeer reference.
The report's confidence is highest for Findings 1–2 (visual inconsistency with established biological norms) and lower for Findings 3–4 (more subjective criteria).
Recommended follow-up: request unprocessed original images with timestamps from the corresponding author; consider a PubPeer post and notification to the journal editorial office and the authors' institution.
This is an AI-assisted screening report; it is not a formal finding of misconduct. False positives remain possible.