Automated Imaging of Cataract Surgery Using Artificial Intelligence
<b>Objectives:</b> This study proposes a state-of-the-art technology to estimate a set of parameters to automatically display an optimized image on a screen during cataract surgery. <b>Methods:</b> We constructed an architecture comprising two stages to estimate the parameter...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-02-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/4/445 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849718553687621632 |
|---|---|
| author | Young Jae Kim Sung Ha Hwang Kwang Gi Kim Dong Heun Nam |
| author_facet | Young Jae Kim Sung Ha Hwang Kwang Gi Kim Dong Heun Nam |
| author_sort | Young Jae Kim |
| collection | DOAJ |
| description | <b>Objectives:</b> This study proposes a state-of-the-art technology to estimate a set of parameters to automatically display an optimized image on a screen during cataract surgery. <b>Methods:</b> We constructed an architecture comprising two stages to estimate the parameters for realizing the optimized image. The Pix2Pix approach was first introduced to generate fake images that mimic the optimal image. This part can be considered a preliminary step; it uses training datasets comprising both an original microscopy image as the input data and an optimally tuned image by ophthalmologists as the label data. The second part of the architecture was inspired by ensemble learning, in which two ResNet-50 models were trained in parallel using fake images obtained in the previous step and unprocessed images. Each set of features extracted by the ensemble-like scheme was exploited for the regression of the optimal parameters. <b>Results:</b> The fidelity of our method was confirmed through relevant quantitative assessments (NMSE 121.052 ± 181.227, PSNR 29.887 ± 4.682, SSIM 0.965 ± 0.047). <b>Conclusions:</b> Subsequently, surgeons reassured that the objects to be highlighted on the screen for cataract surgery were faithfully visualized by the automatically estimated parameters. |
| format | Article |
| id | doaj-art-85fa2d99e5884bdbbfad0583b99076a5 |
| institution | DOAJ |
| issn | 2075-4418 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-85fa2d99e5884bdbbfad0583b99076a52025-08-20T03:12:20ZengMDPI AGDiagnostics2075-44182025-02-0115444510.3390/diagnostics15040445Automated Imaging of Cataract Surgery Using Artificial IntelligenceYoung Jae Kim0Sung Ha Hwang1Kwang Gi Kim2Dong Heun Nam3Gachon Biomedical & Convergence Institute, Gil Medical Center, Gachon University, Incheon 21565, Republic of KoreaDepartment of Ophthalmology, Gil Medical Center, College of Medicine, Gachon University, Incheon 21565, Republic of KoreaDepartment of Biomedical Engineering, Gil Medical Center, College of Medicine, Gachon University, Incheon 21565, Republic of KoreaDepartment of Ophthalmology, Gil Medical Center, College of Medicine, Gachon University, Incheon 21565, Republic of Korea<b>Objectives:</b> This study proposes a state-of-the-art technology to estimate a set of parameters to automatically display an optimized image on a screen during cataract surgery. <b>Methods:</b> We constructed an architecture comprising two stages to estimate the parameters for realizing the optimized image. The Pix2Pix approach was first introduced to generate fake images that mimic the optimal image. This part can be considered a preliminary step; it uses training datasets comprising both an original microscopy image as the input data and an optimally tuned image by ophthalmologists as the label data. The second part of the architecture was inspired by ensemble learning, in which two ResNet-50 models were trained in parallel using fake images obtained in the previous step and unprocessed images. Each set of features extracted by the ensemble-like scheme was exploited for the regression of the optimal parameters. <b>Results:</b> The fidelity of our method was confirmed through relevant quantitative assessments (NMSE 121.052 ± 181.227, PSNR 29.887 ± 4.682, SSIM 0.965 ± 0.047). <b>Conclusions:</b> Subsequently, surgeons reassured that the objects to be highlighted on the screen for cataract surgery were faithfully visualized by the automatically estimated parameters.https://www.mdpi.com/2075-4418/15/4/445cataract surgeryparameter estimationPix2Pixensemble learningResNet-50regression |
| spellingShingle | Young Jae Kim Sung Ha Hwang Kwang Gi Kim Dong Heun Nam Automated Imaging of Cataract Surgery Using Artificial Intelligence Diagnostics cataract surgery parameter estimation Pix2Pix ensemble learning ResNet-50 regression |
| title | Automated Imaging of Cataract Surgery Using Artificial Intelligence |
| title_full | Automated Imaging of Cataract Surgery Using Artificial Intelligence |
| title_fullStr | Automated Imaging of Cataract Surgery Using Artificial Intelligence |
| title_full_unstemmed | Automated Imaging of Cataract Surgery Using Artificial Intelligence |
| title_short | Automated Imaging of Cataract Surgery Using Artificial Intelligence |
| title_sort | automated imaging of cataract surgery using artificial intelligence |
| topic | cataract surgery parameter estimation Pix2Pix ensemble learning ResNet-50 regression |
| url | https://www.mdpi.com/2075-4418/15/4/445 |
| work_keys_str_mv | AT youngjaekim automatedimagingofcataractsurgeryusingartificialintelligence AT sunghahwang automatedimagingofcataractsurgeryusingartificialintelligence AT kwanggikim automatedimagingofcataractsurgeryusingartificialintelligence AT dongheunnam automatedimagingofcataractsurgeryusingartificialintelligence |