EUVIMEDEUVIMED entdecken
Fallbericht

The Role of Artificial Intelligence Algorithms in Challenging Diagnostic Cases – Between Potential and Real Clinical Support

Agata Doligalska-Dolina, Klaudia Ziółkowska, Maciej J Wróbel

Otolaryngologia Polska · 07.05.2026 · 80 (2) · 39-45

DOI: 10.5604/01.3001.0055.7287 ↗

Kurz gefasst

Fallbericht in der Zeitschrift Otolaryngologia Polska, erschienen am 07.05.2026. Verfasst von Agata Doligalska-Dolina und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Otolaryngology and Laryngological Oncology, Collegium Medicum, Nicolaus Copernicus University, Bydgoszcz, Poland. Hauptthemen laut MEDLINE: Artificial Intelligence, Lymphoma, Extranodal NK-T-Cell, Algorithms, Nose Neoplasms. Die Arbeit nennt 12 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Extranodal natural killer/T-cell lymphoma, nasal type (ENKTL), is a rare EBV-associated malignancy characterized by destructive tumors in the nasal cavity, often causing delayed diagnosis due to tissue necrosis. Artificial intelligence (AI) has the potential to assist in complex diagnostic challenges.

Aim

This study aimed to evaluate the diagnostic performance and clinical relevance of publicly accessible AI-driven platforms in identifying ENKTL based on a real, challenging clinical case.

Materials and methods: Clinical data from a 26-year-old male patient with protracted and complicated symptoms initially suggesting infectious conditions were used. Eight AI symptom assessment platforms were tested for diagnostic outputs using identical anonymized case inputs at early and advanced disease stages. Diagnostic suggestions, alignment with the final histopathological diagnosis, and quality of recommendations were analyzed.

Results

Six of eight platforms provided valid differential diagnoses consistent with clinical data. Early-stage data mainly generated infectious diagnoses such as peritonsillar abscess. Advanced-stage data shifted diagnoses toward invasive fungal, autoimmune, and malignant diseases, with platforms such as Doctronic and Symptomate explicitly including lymphoma. Diagnostic pathways varied, with recommendations for comprehensive laboratory tests, imaging, biopsies, and immunological assays. AI systems demonstrated oncologic vigilance when EBV viral data were included, emphasizing clinician oversight for accurate diagnosis.

Conclusions

AI-based diagnostic platforms show promise in supporting preliminary assessment and differential diagnosis in rare, complex cases such as ENKTL. While AI can guide diagnostic reasoning and propose investigations, it does not replace clinical judgment. Further development and validation of AI tools are needed to enhance their reliability and integration into medical practice.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceLymphoma, Extranodal NK-T-CellAlgorithmsNose NeoplasmsHumansMaleAdultDiagnosis, Differential

Stichwörter der Autor:innen

clinical decision support; diagnostic algorithms; diagnostic delay; EBV-associa…

Autor:innen und Institutionen

  • Department of Otolaryngology and Laryngological Oncology, Collegium Medicum, Nicolaus Copernicus University, Bydgoszcz, Poland.

Zitieren

APA

Doligalska-Dolina, A., Ziółkowska, K., & Wróbel, M. J. (2026). The Role of Artificial Intelligence Algorithms in Challenging Diagnostic Cases – Between Potential and Real Clinical Support. Otolaryngologia Polska, 80(2), 39-45. https://doi.org/10.5604/01.3001.0055.7287

Vancouver

Doligalska-Dolina A, Ziółkowska K, Wróbel MJ. The Role of Artificial Intelligence Algorithms in Challenging Diagnostic Cases – Between Potential and Real Clinical Support. Otolaryngologia Polska. 2026;80(2):39-45. doi:10.5604/01.3001.0055.7287

BibTeX
@article{DoligalskaDolina2026,
  author = {Doligalska-Dolina, Agata and Ziółkowska, Klaudia and Wróbel, Maciej J},
  title = {The Role of Artificial Intelligence Algorithms in Challenging Diagnostic Cases – Between Potential and Real Clinical Support},
  journal = {Otolaryngologia Polska},
  year = {2026},
  volume = {80},
  number = {2},
  pages = {39--45},
  doi = {10.5604/01.3001.0055.7287}
}
Kostenlos starten →