Post

New research · Ophthalmology
Diagnostic and interventional imaging · 2d
AI / informaticsDiagnostic and interventional imaging · 2026

Combination of color-Doppler ultrasound and MRI to improve lacrimal gland lesion characterization: A machine learning approach.

Axelle Riehm, Lucile Senicourt, Emma O'Shaughnessy … Augustin Lecler
Read paper
OphthalmologyAI / informatics

Machine learning combining magnetic resonance imaging and ultrasound distinguished malignant from benign lacrimal gland lesions

Combination of color-Doppler ultrasound and MRI to improve lacrimal gland lesion characterization: A machine learning approach.

Axelle Riehm … Augustin Lecler
Diagnostic and interventional imaging · 2026
Purpose

The purpose of this study was to develop a machine learning-based algorithm based on a combination of magnetic resonance imaging (magnetic resonance imaging) and color-Doppler ultrasound (color-Doppler ultrasound) to characterize lacrimal gland lesions.

Methods

All patients with a lacrimal gland lesion who underwent magnetic resonance imaging examination and color-Doppler ultrasound between 2014 and 2025 were retrospectively included.

n = 130 lesions
Results

correctly told cancer from non-cancer lesions far above chance, though small study, wide confidence range

AUC 0.88 (95% CI 0.69 to 1)
null = 00.691
CI excludes the null - significant
More results

In multiclass analysis (benign non-epithelial, benign epithelial, malignant non-epithelial and malignant epithelial), the random forest yielded a macro-averaged AUC of 0.857 (95% CI: 0.722-0.972) for the all-features model.

More results

A 5-top features signature comprising apparent diffusion coefficient and resistance index values, echogenicity, age and lesion type (infiltrative vs. well-delineated mass), yielded an AUC of 0.785 (95% CI: 0.641-0.941) to distinguish between the four classes.

“
Conclusion · 1 of 2

A combination of magnetic resonance imaging and color-Doppler ultrasound features demonstrated high diagnostic performance for characterizing lacrimal gland lesions.

Conclusion · 2 of 2

A simplified 5-feature signature showed similar diagnostic performance compared to the all-features model and warrants prospective multicenter validation for clinical application.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
AI / Informatics
0·962 for OCT
MerMED-FM was very accurate at diagnosing diseases using eye scans
AI / Informatics
0.98
Artificial intelligence's eyelid-height measurements matched doctors' manual measurements almost perfectly
AI / Informatics
92.1%
combining eye scans and photos correctly told benign from cancerous lesions apart nearly every time
Cohort Study
28.6%
recurred locally in more than 1 in 4 patients over years of follow-up
Cohort Study
-0.395
excision group's post-op eyelid fullness score was lower - greater improvement
Cohort Study
86.7%
of infants probed after 12 months still had unresolved tear duct blockage
Observational
16%
eyelid tissue in rosacea patients showed less of this key repair-signaling protein inside cell nuclei
Cohort Study
25%
about 1 in 4 treated cases had symptoms return after improving