What question did this study set out to answer?

To develop a non-invasive, automated framework that accurately predicts perioperative hemoglobin concentration from palpebral conjunctiva images using deep learning guided by prior causal knowledge.

November 30, 2022Frontiers in Public HealthOpen Access

Two-stage hemoglobin prediction based on prior causality

View Full Paper

Ask AI

Bookmark

To develop a non-invasive, automated framework that accurately predicts perioperative hemoglobin concentration from palpebral conjunctiva images using deep learning guided by prior causal knowledge.
Designed a two-stage pipeline utilizing neural semantic segmentation to localize the palpebral conjunctiva region based on causal knowledge, followed by a convolutional neural network for hemoglobin regression.
Trained and evaluated the prediction framework on clinical image datasets collected from perioperative patients.
The proposed deep learning model achieved a predictive score (R²) of 0.512 and an explained variance score of 0.535.
Hemoglobin concentration regression resulted in a mean absolute error of 1.521 across the clinical dataset.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Predictive modeling study demonstrates non-invasive hemoglobin estimation from palpebral conjunctiva images in surgical patients, highlighting an alternative to blood draws.

Cite This Study

Chen et al. (2022) studied this question.

synapsesocial.com/papers/6a87d6793f31f6b699bb659e https://doi.org/10.3389/fpubh.2022.1079389

View Full Paper

Ask AI

Bookmark

Implication

Predictive modeling study demonstrates non-invasive hemoglobin estimation from palpebral conjunctiva images in surgical patients, highlighting an alternative to blood draws.

Two-stage hemoglobin prediction based on prior causality

Key Points

Discussion

Implication

Cite This Study

Two-stage hemoglobin prediction based on prior causality

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider