Accepted for/Published in: JMIR Medical Informatics
Date Submitted: May 6, 2026
Date Accepted: Oct 21, 2026
Deep-learning analysis of peripheral blood smear for prediction of acute inflammatory states and chronic comorbidities: Development and Internal Validation Study
ABSTRACT
Background:
Peripheral blood smear (PBS) examination provides morphology of immune cells reflecting systemic pathophysiology, yet whether PBS examination can be utilized to predict systemic inflammatory states has not been systematically established.
Objective:
To develop and validate an integrated deep learning framework that automates PBS analysis through hierarchical cell-level and slide-level assessment to predict patient inflammatory states across acute and chronic conditions.
Methods:
An integrated deep learning framework was developed for PBS analysis, in which cell-level morphological features were extracted and aggregated into slide-level representations to predict inflammatory states. A total of 51,369 PBS examinations were used for self-supervised cell-level pretraining and 10,329 PBS examinations for slide-level downstream tasks. Outcomes comprised acute and chronic inflammatory states. Acute states included white blood cell (WBC) count categorized as low, normal, or high, elevated C-reactive protein (CRP), elevated erythrocyte sedimentation rate (ESR), and sepsis. Chronic states included the Charlson Comorbidity Index (CCI), cardiovascular disease (CVD), chronic kidney disease (CKD), and diabetes mellitus (DM). Discriminative performance was assessed using the area under the receiver operating characteristic curve (AUROC).
Results:
The framework achieved robust discrimination of acute inflammatory states across both datasets (AUROCs, 0.83–0.86 for WBC count, 0.86–0.89 for CRP, 0.87–0.90 for ESR, and 0.82–0.84 for sepsis; holdout and development, respectively). Discrimination of chronic states was more modest but similarly consistent (AUROCs, 0.72–0.73 for CCI, 0.66–0.67 for CVD, 0.75–0.76 for CKD, and 0.65–0.66 for DM).
Conclusions:
Integrated deep learning framework for automated PBS analysis can discriminate systemic inflammatory states. These findings support the potential of deep learning–based PBS analysis as a novel approach to assessing systemic inflammatory states in clinical practice.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.