Expert-level detection of M-proteins in serum protein electrophoresis using machine learning

• Verfasst von: Elfert, Eike [VerfasserIn]
• Verfasst von: Kaminski, Wolfgang E. [VerfasserIn]
• Verfasst von: Matek, Christian [VerfasserIn]
• Verfasst von: Hoermann, Gregor [VerfasserIn]
• Verfasst von: Axelsen, Eyvind W. [VerfasserIn]
• Verfasst von: Marr, Carsten [VerfasserIn]
• Verfasst von: Piehler, Armin [VerfasserIn]
• Titel: Expert-level detection of M-proteins in serum protein electrophoresis using machine learning
• Verf.angabe: Eike Elfert, Wolfgang E. Kaminski, Christian Matek, Gregor Hoermann, Eyvind W. Axelsen, Carsten Marr, Armin P. Piehler
• E-Jahr: 2024
• Jahr: 17. Juni 2024
• Umfang: 9 S.
• Fussnoten: Gesehen am 25.11.2024
• Titel Quelle: Enthalten in: Clinical chemistry and laboratory medicine
• Ort Quelle: Berlin [u.a.] : De Gruyter, 1998
• Jahr Quelle: 2024
• Band/Heft Quelle: 62(2024), 12, Seite 2498-2506
• ISSN Quelle: 1437-4331
• DOI: doi:10.1515/cclm-2024-0222
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: artificial intelligence
• Sach-SW: electrophoresis
• Sach-SW: machine learning
• Sach-SW: monoclonal gammopathy
• Sach-SW: myeloma
• K10plus-PPN: 190946385X

Abstract

Objectives Serum protein electrophoresis (SPE) in combination with immunotyping (IMT) is the diagnostic standard for detecting monoclonal proteins (M-proteins). However, interpretation of SPE and IMT is weakly standardized, time consuming and investigator dependent. Here, we present five machine learning (ML) approaches for automated detection of M-proteins on SPE on an unprecedented large and well-curated data set and compare the performance with that of laboratory experts. Methods SPE and IMT were performed in serum samples from 69,722 individuals from Norway. IMT results were used to label the samples as M-protein present (positive, n=4,273) or absent (negative n=65,449). Four feature-based ML algorithms and one convolutional neural network (CNN) were trained on 68,722 randomly selected SPE patterns to detect M-proteins. Algorithm performance was compared to that of an expert group of clinical pathologists and laboratory technicians (n=10) on a test set of 1,000 samples. Results The random forest classifier showed the best performance (F1-Score 93.2%, accuracy 99.1%, sensitivity 89.9%, specificity 99.8%, positive predictive value 96.9%, negative predictive value 99.3%) and outperformed the experts (F1-Score 61.2 ± 16.0%, accuracy 89.2 ± 10.2%, sensitivity 94.3 ± 2.8%, specificity 88.9 ± 10.9%, positive predictive value 47.3 ± 16.2%, negative predictive value 99.5 ± 0.2%) on the test set. Interestingly the performance of the RFC saturated, the CNN performance increased steadily within our training set (n=68,722). Conclusions Feature-based ML systems are capable of automated detection of M-proteins on SPE beyond expert-level and show potential for use in the clinical laboratory.