June 2023
Volume 64, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2023
A-Eye: Large-scale MRI automatic biomarkers extraction
Author Affiliations & Notes
  • Jaime Barranco
    Radiology, Centre Hospitalier Universitaire Vaudois, Lausanne, Vaud, Switzerland
    Universite de Lausanne, Lausanne, Vaud, Switzerland
  • Hamza Kebiri
    Radiology, Centre Hospitalier Universitaire Vaudois, Lausanne, Vaud, Switzerland
    Universite de Lausanne, Lausanne, Vaud, Switzerland
  • Óscar Esteban
    Radiology, Centre Hospitalier Universitaire Vaudois, Lausanne, Vaud, Switzerland
  • Raphael Sznitman
    Universitat Bern, Bern, Bern, Switzerland
  • Oliver Stachs
    Universitatsmedizin Rostock, Rostock, Mecklenburg-Vorpommern, Germany
  • Philipp Stachs
    Karlsruher Institut fur Technologie, Karlsruhe, Baden-Württemberg, Germany
  • Sönke Langner
    Universitatsmedizin Rostock, Rostock, Mecklenburg-Vorpommern, Germany
    Universitatsmedizin Greifswald Klinik und Poliklinik fur Neurologie, Greifswald, Mecklenburg-Vorpommern, Germany
  • Benedetta Franceschiello
    HES-SO Valais Wallis, Sion, VS, Switzerland
  • Meritxell Bach
    Universite de Lausanne, Lausanne, Vaud, Switzerland
    Radiology, Centre Hospitalier Universitaire Vaudois, Lausanne, Vaud, Switzerland
  • Footnotes
    Commercial Relationships   Jaime Barranco None; Hamza Kebiri None; Óscar Esteban None; Raphael Sznitman None; Oliver Stachs None; Philipp Stachs None; Sönke Langner None; Benedetta Franceschiello None; Meritxell Bach None
  • Footnotes
    Support  Project number: 0806-2021
Investigative Ophthalmology & Visual Science June 2023, Vol.64, 5039. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jaime Barranco, Hamza Kebiri, Óscar Esteban, Raphael Sznitman, Oliver Stachs, Philipp Stachs, Sönke Langner, Benedetta Franceschiello, Meritxell Bach; A-Eye: Large-scale MRI automatic biomarkers extraction. Invest. Ophthalmol. Vis. Sci. 2023;64(8):5039.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : MREye is raising a lot of interest1, as it provides a comprehensive view of 3D anatomy. This work benchmarks an approach for the automated extraction of biomarkers (axial length and volumetry) from 3D MRI images where eye structures were previously segmented.

Methods : Random set of healthy volunteers (N=1210, 5613 y/o, 616 females, 594 males) underwent whole-body MRI for the population-based Study of Health in Pomerania (SHIP, Germany)2,3. T1w-images were acquired at 1.5T 1mm3 isotropic. Manual annotations (9 ROIs) were performed on other set of 35 subjects.
We extracted metadata and computed BMI (Body Mass Index) per subject. We automatically extracted axial length from automatic atlas-based segmentations4 of lens and globe (Figure 1ab). First, we computed the centroid of the lens and the orthogonal line to the globe in the 2D axial slice, to finally measure the distance between the extreme intersection points of the line with those structures. We added an extra distance to overcome the lack of cornea segmentation by detecting the number of voxels until the next big change of intensity (Figure 1cd).
Furthermore, we provide mm3 volumetry for all segmented structures, and computed correlation and linear regression between BMI and volumetry per structure (Figure 2).

Results : The resulting axial lengths were close to the literature2: 22.0±1.3 for males, and 21.2±1.3mm for females, while in previous studies were 23.4±0.8 and 22.8±0.9mm.
Benchmarked volumetry (in mm3) of eye structures for the first time. For example, optic nerve is 722±92.39 for males, and 612±82.1 for females.
Existing correlation between BMI and eye structures volumes (e.g. fats, lateral and medial rectus muscles: ρ>0.3, p<0.001).

Conclusions : The automatic extraction of eye biomarkers is possible using MREye. We automatically extracted axial length using segmentation of lens and globe; and established a baseline in the volumetric measurement of different eye structures with mm3 resolution (lens, globe, optic nerve, intraconal and extraconal fats, and lateral, medial, inferior and superior rectus muscles), while investigating the relationship between metadata (sex, BMI) and volume of such structures.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

a) Axial length automatic extraction b) Results grouped by sex c) d) a) b) with extra distance

a) Axial length automatic extraction b) Results grouped by sex c) d) a) b) with extra distance

 

BMI and volume correlation per eye structure grouped by sex. In red, Huber linear regression curve (robust to outliers) for those measurements which showed correlation.

BMI and volume correlation per eye structure grouped by sex. In red, Huber linear regression curve (robust to outliers) for those measurements which showed correlation.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×