Work

When 99.98% is too Good to be True: Preventing Rule-Induced Overfitting in Embodied Clinical AI for Surgical Readmission Prediction

Year: 2026

Type: conference-paper

Source: ITU Kaleidoscope Academic Conference (ITU K)

DOI: 10.23919/ituk70038.2026.11632583

Authors: Antoine Bagula, Landry Mbale, Olasupo Ajayi, Patrick Sello, Jennifer Chipps, Sean Broomhead, Petra Brysiewicz, Damian Clarke

Institutions: University of the Western Cape, University of KwaZulu-Natal

Language: English

Topic: Machine Learning in Healthcare

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: closed

OpenAlex ID: W7202377083

FWCI: 1.322

Cites: 18

Cited by: 1

Related to: 10