Clinical modeling / 2026

DeepSelective.

Interpretable prognostic feature selection and compression for electronic health records.

Pattern Recognition · 174, 112970

01 / The problem

Electronic-health-record features need useful, interpretable prognostic selection and compression.

02 / The core idea

Select and compress prognostic features with an interpretable modeling approach.

FIGURE 01 / ORIGINAL CONCEPT SKETCH

Original concept sketch; not a reproduced paper figure.

03 / What the public evidence shows

Published in Pattern Recognition 174 (2026), 112970. The final paper, rather than an older preprint label, anchors the citation.

04 / Where the claim stops

Clinical modeling is a separate research thread. No claim of prospective clinical benefit or deployment is made.

An editorial scope note, not an unverified quotation of author limitations.