Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SpecstatOR: Speckle statistics-based iOCT Segmentation Network for Ophthalmic Surgery

Kristina Mach, Hessam Roodaki, Michael Sommersperger, Nassir Navab

Abstract

This paper presents an innovative approach to intraoperative Optical Coherence Tomography (iOCT) image segmentation in ophthalmic surgery, leveraging statistical analysis of speckle patterns to incorporate statistical pathology-specific prior knowledge. Our findings indicate statistically different speckle patterns within the retina and between retinal layers and surgical tools, facilitating the segmentation of previously unseen data without the necessity for manual labeling. The research involves fitting various statistical distributions to iOCT data, enabling the differentiation of different ocular structures and surgical tools. The proposed segmentation model aims to refine the statistical findings based on prior tissue understanding to leverage statistical and biological knowledge. Incorporating statistical parameters, physical analysis of light-tissue interaction, and deep learning informed by biological structures enhance segmentation accuracy, offering potential benefits to real-time applications in ophthalmic surgical procedures. The study demonstrates the adaptability and precision of using Gamma distribution parameters and the derived binary maps as sole inputs for segmentation, notably enhancing the model's inference performance on unseen data.

SpecstatOR: Speckle statistics-based iOCT Segmentation Network for Ophthalmic Surgery

Abstract

This paper presents an innovative approach to intraoperative Optical Coherence Tomography (iOCT) image segmentation in ophthalmic surgery, leveraging statistical analysis of speckle patterns to incorporate statistical pathology-specific prior knowledge. Our findings indicate statistically different speckle patterns within the retina and between retinal layers and surgical tools, facilitating the segmentation of previously unseen data without the necessity for manual labeling. The research involves fitting various statistical distributions to iOCT data, enabling the differentiation of different ocular structures and surgical tools. The proposed segmentation model aims to refine the statistical findings based on prior tissue understanding to leverage statistical and biological knowledge. Incorporating statistical parameters, physical analysis of light-tissue interaction, and deep learning informed by biological structures enhance segmentation accuracy, offering potential benefits to real-time applications in ophthalmic surgical procedures. The study demonstrates the adaptability and precision of using Gamma distribution parameters and the derived binary maps as sole inputs for segmentation, notably enhancing the model's inference performance on unseen data.
Paper Structure (23 sections, 7 equations, 8 figures, 8 tables)

This paper contains 23 sections, 7 equations, 8 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Heterogeneity assessment
  4. Homogeneity Assessment
  5. Distribution Fitting
  6. Goodness-of-Fit Analysis
  7. Classification
  8. Data Acquisition and Preprocessing
  9. Results
  10. Levene's test
  11. ANOVA
  12. Distribution fit for Group One
  13. ILM
  14. RPE
  15. Tool
  16. ...and 8 more sections

Figures (8)

  • Figure 1: Retinal ex-vivo porcine iOCT B-scan with ILM (green), RPE (light blue), and Tool (Purple) classes marked.
  • Figure 3: The pipeline of the proposed method. 1) Distribution fitting. 2) Parameter isolation to classes. 3) Refinement with deep neural network.
  • Figure 4: Network architecture and training configurations A, B, C, and D as different inputs. Labels: weak from the parameter isolation and segmented ground truth.
  • Figure 5: Scatterplots of statistical distribution parameters, fitted for Group Two
  • Figure 6: Calibrated Gamma distribution parameters, Group One iOCT, alpha parameters between 10 and 30 shown with green, below 10 black, and the rest blue.
  • ...and 3 more figures