Deep phenotyping of blood cell data reveals novel clinical biomarkers

The complete blood count with differential (CBD) is one of the most common blood tests worldwide. Although modern CBD analyzers generate flow-cytometry based single-cell measurements, the resultant CBD markers are limited to coarse summary features, such as total cell counts and average cell sizes. This means, they cannot detect subtle cell population shifts that may signal early-stage pathogenesis. To test this, we evaluate whether AI-based analysis of the single-cell CBD data can generate novel and clinically prognostic biomarkers.

We developed two complementary methods for biomarker discovery using CBD tests from an academic medical center. To create interpretable biomarkers, we clustered cells into known sub-populations and performed statistical summarization. In tandem, self-supervised autoencoders were developed to extract novel non-linear markers. We evaluated the utility of the generated markers for patient prognostication across a range of outcomes (mortality, inpatient admission, and future disease development).

Our study included 242,584 CBD samples from 127,543 patients. Both clustering and autoencoder biomarkers showed strong prognostic associations for all-cause mortality, inpatient admission, and development of anemia, cancer, or cardiovascular disease, with associations remaining significant after adjustment for demographics and clinical CBD markers. A large subset of these markers also showed high novelty, having low correlations to existing CBD markers, while also exhibiting significant correlations with broader physiologic signals, such as inflammatory, infectious, and coagulopathic markers.

Collectively, these results demonstrate how AI techniques can allow for deeper phenotyping of routine clinical blood counts, uncovering clinically informative biomarkers that capture subtle physiological variations.