Smartphone-integrated Microscopy and Deep Learning for Early Mold Screening in Cinnamon: A Scoping Review of Transferable Evidence and Validation Requirementshttps://doi.org/10.31559/VMPH2026.7.3.5
Year: 2026
Type: article
Study design: Systematic Review
Abstract: Objectives: To map direct and transferable evidence for smartphone-integrated microscopy and deep learning as a low-cost approach to early mold screening in cinnamon and to identify requirements for end-to-end validation. Methods: This PRISMA-ScR-informed scoping review searched PubMed/MEDLINE and IEEE Xplore on 5 September 2026 and examined backward citations. Screening, charting, and narrative transferability appraisal were performed by one reviewer. The two-database, English-language search and unretrieved reports limit completeness. Results: The documented aggregate flow comprises 786 record appearances (369 PubMed, 370 IEEE Xplore, and 47 backward-citation appearances), 29 duplicates removed, 757 records screened, 168 reports sought, 62 not obtained, 106 full texts assessed, 24 exclusions, and 82 included publications. The evidence domains were portable microscopy (n = 38), fungal-image analysis (n = 24), food-mold recognition (n = 5), and spice/cinnamon contamination (n = 15). Published examples support component feasibility, but only a subset of individual-study outcomes could be independently rechecked in this revision. Conclusion: Among the retrieved and mapped studies, no report was identified that prospectively validated a complete culture-referenced, sample-level smartphone workflow for pre-visible mold growth in cinnamon. Smartphone microscopy remains a candidate triage technology, not a substitute for culture, taxonomic confirmation, or separate mycotoxin analysis.
Language: English
Subfield: Biomedical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7226012693
