EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning Method Applied to Autonomous Image Diagnosis for Prick Test

Ramon Hernany Martins Gomes, Edson Luiz Pontes Perger iD, Lucas Hecker Vasques iD, Elaine Gagete, Rafael Plana Simões iD

Life · 02.10.2024 · 14 (10) · 1256

DOI: 10.3390/life14101256 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Life, online erschienen am 02.10.2024. Verfasst von Ramon Hernany Martins Gomes und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Bioprocess and Biotechnology, School of Agriculture, São Paulo State University (UNESP), Avenue Universitária, 3780, Botucatu 18610-034…. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Sao Paulo State University (UNESP).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The skin prick test (SPT) is used to diagnose sensitization to antigens. This study proposes a deep learning approach to infer wheal dimensions, aiming to reduce dependence on human interpretation.

Methods

A dataset of SPT images (n = 5844) was used to infer a convolutional neural network for wheal segmentation (ML model). Three methods for inferring wheal dimensions were evaluated: the ML model; the standard protocol (MA1); and approximation of the area as an ellipse using diameters measured by an allergist (MA2). The results were compared with assisted image segmentation (AIS), the most accurate method. Bland–Altman analysis, distribution analyses, and correlation tests were applied to compare the methods. This study also compared the percentage deviation among these methods in determining the area of wheals with regular geometric shapes (n = 150) and with irregular shapes (n = 150).

Results

The Bland–Altman analysis showed that the difference between methods was not correlated with the absolute area. The ML model achieved a segmentation accuracy of 85.88% and a strong correlation with the AIS method (ρ = 0.88), outperforming all other methods. Additionally, MA1 showed significant error (13.44 ± 13.95%) for pseudopods.

Conclusions

The ML protocol can potentially automate the reading of SPT, offering greater accuracy than the standard protocol.

Schlagwörter

Stichwörter der Autor:innen

IgE responsedeep learning applied to diagnosismeasurement of wheal areaprick testsensitization to antigens

Autor:innen und Institutionen

  • Department of Bioprocess and Biotechnology, School of Agriculture, São Paulo State University (UNESP), Avenue Universitária, 3780, Botucatu 18610-034, SP, Brazil.
  • Medical School, São Paulo State University (UNESP), Avenue Prof. Mário Rubens Guimarães Montenegro, s/n, Botucatu 18618-687, SP, Brazil.
  • Dr. Elaine's Clinic (Clínica Dra. Elaine), 398 Doutor Rodrigues do Lago, Botucatu 18602-091, SP, Brazil.

ORCID-Profile: Edson Luiz Pontes Perger, Lucas Hecker Vasques, Rafael Plana Simões

Förderung

  • Sao Paulo State University (UNESP)

Zitieren

APA

Gomes, R. H. M., Perger, E. L. P., Vasques, L. H., Gagete, E., & Simões, R. P. (2024). Deep Learning Method Applied to Autonomous Image Diagnosis for Prick Test. Life, 14(10), 1256. https://doi.org/10.3390/life14101256

Vancouver

Gomes RHM, Perger ELP, Vasques LH, Gagete E, Simões RP. Deep Learning Method Applied to Autonomous Image Diagnosis for Prick Test. Life. 2024;14(10):1256. doi:10.3390/life14101256

BibTeX
@article{Gomes2024,
  author = {Gomes, Ramon Hernany Martins and Perger, Edson Luiz Pontes and Vasques, Lucas Hecker and Gagete, Elaine and Simões, Rafael Plana},
  title = {Deep Learning Method Applied to Autonomous Image Diagnosis for Prick Test},
  journal = {Life},
  year = {2024},
  volume = {14},
  number = {10},
  pages = {1256},
  doi = {10.3390/life14101256}
}
Kostenlos starten →