CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial Cells
The quantification of corneal endothelial cell (CEC) morphology using manual and semi-automatic software enables an objective assessment of corneal endothelial pathology. However, the procedure is tedious, subjective, and not widely applied in clinical practice. We have developed the CellsDeepNet sy...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-01-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/10/3/320 |
_version_ | 1797486441921511424 |
---|---|
author | Alaa S. Al-Waisy Abdulrahman Alruban Shumoos Al-Fahdawi Rami Qahwaji Georgios Ponirakis Rayaz A. Malik Mazin Abed Mohammed Seifedine Kadry |
author_facet | Alaa S. Al-Waisy Abdulrahman Alruban Shumoos Al-Fahdawi Rami Qahwaji Georgios Ponirakis Rayaz A. Malik Mazin Abed Mohammed Seifedine Kadry |
author_sort | Alaa S. Al-Waisy |
collection | DOAJ |
description | The quantification of corneal endothelial cell (CEC) morphology using manual and semi-automatic software enables an objective assessment of corneal endothelial pathology. However, the procedure is tedious, subjective, and not widely applied in clinical practice. We have developed the CellsDeepNet system to automatically segment and analyse the CEC morphology. The CellsDeepNet system uses Contrast-Limited Adaptive Histogram Equalization (CLAHE) to improve the contrast of the CEC images and reduce the effects of non-uniform image illumination, 2D Double-Density Dual-Tree Complex Wavelet Transform (2DDD-TCWT) to reduce noise, Butterworth Bandpass filter to enhance the CEC edges, and moving average filter to adjust for brightness level. An improved version of U-Net was used to detect the boundaries of the CECs, regardless of the CEC size. CEC morphology was measured as mean cell density (MCD, cell/mm<sup>2</sup>), mean cell area (MCA, μm<sup>2</sup>), mean cell perimeter (MCP, μm), polymegathism (coefficient of CEC size variation), and pleomorphism (percentage of hexagonality coefficient). The CellsDeepNet system correlated highly significantly with the manual estimations for MCD (r = 0.94), MCA (r = 0.99), MCP (r = 0.99), polymegathism (r = 0.92), and pleomorphism (r = 0.86), with <i>p</i> < 0.0001 for all the extracted clinical features. The Bland–Altman plots showed excellent agreement. The percentage difference between the manual and automated estimations was superior for the CellsDeepNet system compared to the CEAS system and other state-of-the-art CEC segmentation systems on three large and challenging corneal endothelium image datasets captured using two different ophthalmic devices. |
first_indexed | 2024-03-09T23:33:18Z |
format | Article |
id | doaj.art-8ddd62f5fffb4dcda42aae25ccceedd9 |
institution | Directory Open Access Journal |
issn | 2227-7390 |
language | English |
last_indexed | 2024-03-09T23:33:18Z |
publishDate | 2022-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Mathematics |
spelling | doaj.art-8ddd62f5fffb4dcda42aae25ccceedd92023-11-23T17:05:36ZengMDPI AGMathematics2227-73902022-01-0110332010.3390/math10030320CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial CellsAlaa S. Al-Waisy0Abdulrahman Alruban1Shumoos Al-Fahdawi2Rami Qahwaji3Georgios Ponirakis4Rayaz A. Malik5Mazin Abed Mohammed6Seifedine Kadry7Computer Technologies Engineering Department, Information Technology Collage, Imam Ja’afar Al-Sadiq University, Baghdad 10064, IraqDepartment of Information Technology, College of Computer and Information Sciences, Majmaah University, Al Majmaah 11952, Saudi ArabiaComputer Science Department, Al-Ma’aref University College, Ramadi, Anbar 31001, IraqSchool of Electrical Engineering and Computer Science, University of Bradford, Bradford BD7 1DP, UKDivision of Medicine, Weill Cornell Medicine-Qatar, Doha 24144, QatarDivision of Medicine, Weill Cornell Medicine-Qatar, Doha 24144, QatarInformation Systems Department, College of Computer Science and Information Technology, University of Anbar, Ramadi 31001, IraqFaculty of Applied Computing and Technology, Noroff University College, 4612 Kristiansand, NorwayThe quantification of corneal endothelial cell (CEC) morphology using manual and semi-automatic software enables an objective assessment of corneal endothelial pathology. However, the procedure is tedious, subjective, and not widely applied in clinical practice. We have developed the CellsDeepNet system to automatically segment and analyse the CEC morphology. The CellsDeepNet system uses Contrast-Limited Adaptive Histogram Equalization (CLAHE) to improve the contrast of the CEC images and reduce the effects of non-uniform image illumination, 2D Double-Density Dual-Tree Complex Wavelet Transform (2DDD-TCWT) to reduce noise, Butterworth Bandpass filter to enhance the CEC edges, and moving average filter to adjust for brightness level. An improved version of U-Net was used to detect the boundaries of the CECs, regardless of the CEC size. CEC morphology was measured as mean cell density (MCD, cell/mm<sup>2</sup>), mean cell area (MCA, μm<sup>2</sup>), mean cell perimeter (MCP, μm), polymegathism (coefficient of CEC size variation), and pleomorphism (percentage of hexagonality coefficient). The CellsDeepNet system correlated highly significantly with the manual estimations for MCD (r = 0.94), MCA (r = 0.99), MCP (r = 0.99), polymegathism (r = 0.92), and pleomorphism (r = 0.86), with <i>p</i> < 0.0001 for all the extracted clinical features. The Bland–Altman plots showed excellent agreement. The percentage difference between the manual and automated estimations was superior for the CellsDeepNet system compared to the CEAS system and other state-of-the-art CEC segmentation systems on three large and challenging corneal endothelium image datasets captured using two different ophthalmic devices.https://www.mdpi.com/2227-7390/10/3/320corneal confocal microscopycorneal endothelial cellsComplex Wavelet Transformdeep learningConvolutional Neural NetworkU-Net architecture |
spellingShingle | Alaa S. Al-Waisy Abdulrahman Alruban Shumoos Al-Fahdawi Rami Qahwaji Georgios Ponirakis Rayaz A. Malik Mazin Abed Mohammed Seifedine Kadry CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial Cells Mathematics corneal confocal microscopy corneal endothelial cells Complex Wavelet Transform deep learning Convolutional Neural Network U-Net architecture |
title | CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial Cells |
title_full | CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial Cells |
title_fullStr | CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial Cells |
title_full_unstemmed | CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial Cells |
title_short | CellsDeepNet: A Novel Deep Learning-Based Web Application for the Automated Morphometric Analysis of Corneal Endothelial Cells |
title_sort | cellsdeepnet a novel deep learning based web application for the automated morphometric analysis of corneal endothelial cells |
topic | corneal confocal microscopy corneal endothelial cells Complex Wavelet Transform deep learning Convolutional Neural Network U-Net architecture |
url | https://www.mdpi.com/2227-7390/10/3/320 |
work_keys_str_mv | AT alaasalwaisy cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells AT abdulrahmanalruban cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells AT shumoosalfahdawi cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells AT ramiqahwaji cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells AT georgiosponirakis cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells AT rayazamalik cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells AT mazinabedmohammed cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells AT seifedinekadry cellsdeepnetanoveldeeplearningbasedwebapplicationfortheautomatedmorphometricanalysisofcornealendothelialcells |