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

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.

Persisting symptoms three to eight months after non-hospitalized COVID-19, a prospective cohort study.

Long-COVID-19 is a proposed syndrome negatively affecting the health of COVID-19 patients. We present data on self-rated health three to eight months after laboratory confirmed COVID-19 disease compared to a control group of SARS-CoV-2 negative patients. We followed a cohort of 8786 non-hospitalized patients who were invited after SARS-CoV-2 testing between February 1 and April 15, 2020 (794 positive, 7229 negative). Participants answered online surveys at baseline and follow-up including questions on demographics, symptoms, risk factors for SARS-CoV-2, and self-rated health compared to one year ago. Determinants for a worsening of self-rated health as compared to one year ago among the SARS-CoV-2 positive group were analyzed using multivariate logistic regression and also compared to the population norm. The follow-up questionnaire was completed by 85% of the SARS-CoV-2 positive and 75% of the SARS-CoV-2 negative participants on average 132 days after the SARS-CoV-2 test. At follow-up, 36% of the SARS-CoV-2 positive participants rated their health "somewhat" or "much" worse than one year ago. In contrast, 18% of the SARS-CoV-2 negative participants reported a similar deterioration of health while the population norm is 12%. Sore throat and cough were more frequently reported by the control group at follow-up. Neither gender nor follow-up time was associated with the multivariate odds of worsening of self-reported health compared to one year ago. Age had an inverted-U formed association with a worsening of health while being fit and being a health professional were associated with lower multivariate odds. A significant proportion of non-hospitalized COVID-19 patients, regardless of age, have not returned to their usual health three to eight months after infection.