This paper is marked retracted in the scholarly record (OpenAlex). Interpret its findings with caution.
Retraction: Machine Learning-Based Prediction of COVID-19 Prognosis Using Clinical and Hematologic Data
Fatemah O Kamel, Rania Magadmi, Sulafah Qutub, Maha Badawi, Mazen Badawi, Tariq A Madani, Areej Alhothali, Ehab A Abozinadah, Duaa M Bakhshwin, Maha H Jamal, Abdulhadi S Burzangi, Mohammed Bazuhair, Hussamaldin Alqutub, Abdulaziz Alqutub, Sameera M Felemban, Fatin Al-Sayes
Abstract
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsCOVID-19 diagnosis using AI · AI in cancer detection · Digital Imaging for Blood Diseases
This article has been retracted by the Editor-in-Chief due to major flaws which undermine the credibility of the article. The article reports biologically impossible hematology values used in modeling, including hemoglobin up to 116 g/dL and neutrophil percentages exceeding 100. Multiple models also report mathematically incompatible performance metrics, such as near-perfect sensitivity with zero or very low F1 scores, or AUC values below 0.5 alongside high accuracy and sensitivity. The ROC results themselves are internally inconsistent, showing near no-skill performance while strong classification metrics are simultaneously reported. Finally, the strikingly uniform, near-identical AUCs across fundamentally different model families further undermine the credibility of the reported model performance. The authors disagree with the retraction.