Development and Validation of a Protein Electrophoresis Classification Algorithm: Tabular Data-Based Alternative
Abstract
Serum protein electrophoresis (SPE) is routinely interpreted through visual assessment of electropherogram images by medical laboratory scientists. We introduce an efficient tabular data-based machine learning approach that directly leverages numerical SPE profiles, offering a robust and interpretable alternative to image-based deep learning methods.
Fields of Study & Subjects
Artificial Intelligence in Healthcare
Clinical Laboratory Practices and Quality Control
Machine Learning in Healthcare
Publication details
- Journal / source
- JMIR Formative Research
- Volume
- 10
- Pages
- e83124–e83124
- Publication date
- 13 January 2026
- Language
- EN