EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ISPAD, bridging clinical expertise and AI for autoimmune-related pernicious anemia diagnosis

Nora Boumela, Guillaume Cadiot, Farid Chaoui

Frontiers in Immunology · 28.04.2026 · 17

DOI: 10.3389/fimmu.2026.1700751 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Immunology, online erschienen am 28.04.2026. Verfasst von Nora Boumela und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electronics, University of Batna 2, Batna, Algeria. Hauptthemen laut MEDLINE: Anemia, Pernicious, Autoimmune Diseases, Artificial Intelligence. Die Arbeit nennt 76 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Pernicious anemia (PA) is a severe clinical consequence of autoimmune gastritis. It results from immune-mediated damage to gastric parietal cells in the oxyntic mucosa. This process leads to intrinsic factor deficiency and subsequent vitamin B 12 malabsorption. This complex autoimmune response, combined with non-specific clinical manifestations, generates substantial diagnostic uncertainty. This uncertainty frequently results in misdiagnosis or delayed diagnosis and, consequently, multiple adverse outcomes, including irreversible neurological complications.

Methods

To address the intrinsic diagnostic uncertainty of PA—arising from heterogeneous, graded, and often discordant clinical, histological, immunological, and biochemical information—we developed ISPAD (Intelligent System for Pernicious Anemia Diagnosis), an explainable AI–based probabilistic framework to integrate this information, producing a probability estimate of pernicious anemia that reflects clinical reasoning rather than rigid diagnostic thresholds.

Results

ISPAD was examined using a series of published diagnostically challenging cases reflecting real-world complexity. These cases included antibody assay interference, hemolysis-masked macrocytosis, seronegative presentations, and cancer-associated atrophy. Across cases, the system generated a continuous and adaptable probabilistic assessment of pernicious anemia. This assessment relied on dynamic, context-dependent integration of available information and illustrated the potential for formalizing complex diagnostic reasoning.

Conclusion

ISPAD illustrates how explainable artificial intelligence can formalize expert reasoning in autoimmune-related pernicious anemia. By integrating heterogeneous and often discordant information into a transparent probabilistic framework, this proof-of-concept approach provides a structured approach to diagnostic reasoning, particularly in complex or atypical situations.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Anemia, PerniciousAutoimmune DiseasesArtificial IntelligenceHumansParietal Cells, GastricFemaleDiagnosis, Differential

Stichwörter der Autor:innen

AI-assisted diagnosisB12 deficiencyOLGA classificationfundic atrophypernicious anemia

Autor:innen und Institutionen

  • Department of Electronics, University of Batna 2, Batna, Algeria.
  • Departement of Electronics/LDCCP lab., Ecole Nationale Polytechnique, Algiers, Algeria.
  • Hepato-Gastroenterology and Digestive Oncology Department, Reims University Hospital Center, Reims, France.
  • Faculty of Medicine, University of Algiers, Algiers, Algeria.

Zitieren

APA

Boumela, N., Cadiot, G., & Chaoui, F. (2026). ISPAD, bridging clinical expertise and AI for autoimmune-related pernicious anemia diagnosis. Frontiers in Immunology, 17. https://doi.org/10.3389/fimmu.2026.1700751

Vancouver

Boumela N, Cadiot G, Chaoui F. ISPAD, bridging clinical expertise and AI for autoimmune-related pernicious anemia diagnosis. Frontiers in Immunology. 2026;17. doi:10.3389/fimmu.2026.1700751

BibTeX
@article{Boumela2026,
  author = {Boumela, Nora and Cadiot, Guillaume and Chaoui, Farid},
  title = {ISPAD, bridging clinical expertise and AI for autoimmune-related pernicious anemia diagnosis},
  journal = {Frontiers in Immunology},
  year = {2026},
  volume = {17},
  doi = {10.3389/fimmu.2026.1700751}
}
Kostenlos starten →