Keratoviz-A multistage keratoconus severity analysis and visualization using deep learning and class activated maps DOI Open Access

D. Mohana Priya,

Mamatha Gowdra Shivanandappa, Srijan Devnath

et al.

International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, Journal Year: 2022, Volume and Issue: 13(1), P. 920 - 920

Published: Oct. 25, 2022

<span lang="EN-US">The detection of keratoconus has been a difficult and arduous process over the years for ophthalmologists who have devised traditional approaches diagnosis including slit-lamp examination observation thinning corneal. The main contribution this paper is using deep learning models namely Resnet50 EfficientNet to not just detect whether an eye infected with or but also accurately stages infection mild, moderate, advanced. dataset used consists corneal topographic maps pentacam images. Individually achieved 97% 94% accuracy on dataset. We employed class activated (CAM) observe help visualize which areas images are utilized when making classifications different keratoconus. Using predict severity can drastically speed up provide accurate results at same time.</span>

Language: Английский

Management of keratoconus: an updated review DOI Creative Commons
Rashmi Deshmukh, Zun Zheng Ong,

Radhika Rampat

et al.

Frontiers in Medicine, Journal Year: 2023, Volume and Issue: 10

Published: June 20, 2023

Keratoconus is the most common corneal ectatic disorder. It characterized by progressive thinning with resultant irregular astigmatism and myopia. Its prevalence has been estimated at 1:375 to 1:2,000 people globally, a considerably higher rate in younger populations. Over past two decades, there was paradigm shift management of keratoconus. The treatment expanded significantly from conservative (e.g., spectacles contact lenses wear) penetrating keratoplasty many other therapeutic refractive modalities, including cross-linking (with various protocols/techniques), combined CXL-keratorefractive surgeries, intracorneal ring segments, anterior lamellar keratoplasty, more recently, Bowman's layer transplantation, stromal keratophakia, regeneration. Several recent large genome-wide association studies (GWAS) have identified important genetic mutations relevant keratoconus, facilitating development potential gene therapy targeting keratoconus halting disease progression. In addition, attempts made leverage power artificial intelligence-assisted algorithms enabling earlier detection progression prediction this review, we provide comprehensive overview current emerging propose algorithm for systematically guiding clinical entity.

Language: Английский

Citations

71
Optimized Artificial Intelligence for Enhanced Ectasia Detection Using Scheimpflug-Based Corneal Tomography and Biomechanical Data DOI Creative Commons
Renato Ambrósio, Aydano Pamponet Machado,

Edileuza Virginio Leão

et al.

American Journal of Ophthalmology, Journal Year: 2022, Volume and Issue: 251, P. 126 - 142

Published: Dec. 19, 2022

To optimize artificial intelligence (AI) algorithms to integrate Scheimpflug-based corneal tomography and biomechanics enhance ectasia detection.Multicenter cross-sectional case-control retrospective study.A total of 3886 unoperated eyes from 3412 patients had Pentacam Corvis ST (Oculus Optikgeräte GmbH) examinations. The database included 1 eye randomly selected 1680 normal (N) 1181 "bilateral" keratoconus (KC) patients, along with 551 topography very asymmetric (VAE-NT), their 474 ectatic (VAE-E) eyes. current TBIv1 (tomographic-biomechanical index) was tested, an optimized AI algorithm developed for augmenting accuracy.The area under the receiver operating characteristic curve (AUC) discriminating clinical (KC VAE-E) 0.999 (98.5% sensitivity; 98.6% specificity [cutoff: 0.5]), VAE-NT, 0.899 (76% 89.1% 0.29]). A novel random forest (TBIv2), 18 features in 156 trees using 10-fold cross-validation, a significantly higher AUC (0.945; DeLong, P < .0001) detecting VAE-NT (84.4% sensitivity 90.1% specificity; cutoff: 0.43; similar (0.999; = .818; 98.7% 99.2% 0.8]). Considering all cases, TBIv2 (0.985) than (0.974; .0001).AI optimization biomechanical assessments augments accuracy detection, characterizing susceptibility diverse group. Some VAE may have true unilateral ectasia. Machine learning considering additional data, including epithelial thickness or other parameters multimodal refractive imaging, will continuously accuracy. NOTE: Publication this article is sponsored by American Ophthalmological Society.

Language: Английский

Citations

54
A Deep Feature Fusion of Improved Suspected Keratoconus Detection with Deep Learning DOI Creative Commons
Ali H. Al‐Timemy, Laith Alzubaidi,

Zahraa M. Mosa

et al.

Diagnostics, Journal Year: 2023, Volume and Issue: 13(10), P. 1689 - 1689

Published: May 10, 2023

Detection of early clinical keratoconus (KCN) is a challenging task, even for expert clinicians. In this study, we propose deep learning (DL) model to address challenge. We first used Xception and InceptionResNetV2 DL architectures extract features from three different corneal maps collected 1371 eyes examined in an eye clinic Egypt. then fused using detect subclinical forms KCN more accurately robustly. obtained area under the receiver operating characteristic curves (AUC) 0.99 accuracy range 97-100% distinguish normal with established KCN. further validated based on independent dataset 213 Iraq AUCs 0.91-0.92 88-92%. The proposed step toward improving detection

Language: Английский

Citations

17
Systematic reviews of machine learning in healthcare: a literature review DOI Creative Commons
Katarzyna Kolasa,

Bisrat Yeshewas Admassu,

Malwina Hołownia-Voloskova

et al.

Expert Review of Pharmacoeconomics & Outcomes Research, Journal Year: 2023, Volume and Issue: 24(1), P. 63 - 115

Published: Nov. 13, 2023

The increasing availability of data and computing power has made machine learning (ML) a viable approach to faster, more efficient healthcare delivery.

Language: Английский

Citations

14
Artificial intelligence and corneal diseases DOI

Linda Kang,

Dena Ballouz, Maria A. Woodward

et al.

Current Opinion in Ophthalmology, Journal Year: 2022, Volume and Issue: 33(5), P. 407 - 417

Published: July 12, 2022

Purpose of review Artificial intelligence has advanced rapidly in recent years and provided powerful tools to aid with the diagnosis, management, treatment ophthalmic diseases. This article aims most current clinical artificial applications anterior segment diseases, an emphasis on microbial keratitis, keratoconus, dry eye syndrome, Fuchs endothelial dystrophy. Recent findings Most approaches have focused developing deep learning algorithms based various imaging modalities. Algorithms been developed detect differentiate keratitis classes quantify features. may early detection staging keratoconus. Many advances made detect, segment, features syndrome Fuchs. There is significant variability reporting methodology, patient population, outcome metrics. Summary shows great promise detecting, diagnosing, grading, measuring a need for standardization improve transparency, validity, comparability algorithms.

Language: Английский

Citations

20
Keratoconus: imaging modalities and management DOI
Noor Alqudah

Medical Hypothesis Discovery & Innovation in Ophthalmology, Journal Year: 2024, Volume and Issue: 13(1), P. 44 - 54

Published: July 1, 2024

Keratoconus (KCN) is characterized by gradual thinning and steepening of the cornea, which can lead to significant vision problems owing high astigmatism, corneal scarring, or even perforation. The detection KCN in its early stages crucial for effective treatment. In this review, we describe current advances diagnosis treatment KCN.

Language: Английский

Citations

4
Strategies for Early Keratoconus Diagnosis: A Narrative Review of Evaluating Affordable and Effective Detection Techniques DOI Open Access

Arige Gideon Abou Said,

Joan Gispets,

Einat Shneor

et al.

Journal of Clinical Medicine, Journal Year: 2025, Volume and Issue: 14(2), P. 460 - 460

Published: Jan. 13, 2025

Keratoconus is a progressive corneal disorder that can lead to irreversible visual impairment if not detected early. Despite its high prevalence, early diagnosis often delayed, especially in low-to-middle-income countries due limited awareness and restricted access advanced diagnostic tools such as topography, tomography, optical coherence biomechanical assessments. These technologies are essential for identifying early-stage keratoconus, yet their cost limits accessibility resource-limited settings. While portability important accessibility, the sensitivity specificity of must be considered primary metrics ensure accurate effective detection keratoconus. This review examines both traditional techniques, including use machine learning artificial intelligence, enhance diagnosis. Artificial intelligence-based approaches show significant potential transforming keratoconus by improving accuracy diagnosis, when combined with imaging devices. Notable innovations include SmartKC, smartphone-based machine-learning application, mobile topography through null-screen test, Smartphone-based Keratograph, providing affordable portable solutions. Additionally, contrast testing demonstrates detection, although precise platform routine clinical has established. The emphasizes need increased among clinicians, particularly underserved regions, advocates development accessible, low-cost tools. Further research needed validate effectiveness these emerging detecting

Language: Английский

Citations

0
Ethnicity optimized indices enhance the diagnostic efficiency of early Keratoconus: A multicenter validation study DOI
Yan Huo, Ruisi Xie, Jing Li

et al.

Contact Lens and Anterior Eye, Journal Year: 2025, Volume and Issue: unknown, P. 102382 - 102382

Published: Feb. 1, 2025

Language: Английский

Citations

0
A new morphological classification of keratoconus using few-shot learning in candidates for intrastromal corneal ring implants DOI
Zhila Agharezaei, Mohammad Shirshekar,

Reza Firouzi

et al.

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 106, P. 107664 - 107664

Published: Feb. 17, 2025

Language: Английский

Citations

0
Artificial intelligence as diagnostic modality for keratoconus: A systematic review and meta-analysis DOI Creative Commons
Azzahra Afifah, Fara Syafira, Putri Mahirah Afladhanti

et al.

Journal of Taibah University Medical Sciences, Journal Year: 2024, Volume and Issue: 19(2), P. 296 - 303

Published: Jan. 1, 2024

The challenges in diagnosing keratoconus (KC) have led researchers to explore the use of artificial intelligence (AI) as a diagnostic tool. AI has emerged new way improve efficiency KC diagnosis. This study analyzed modality for KC.

Language: Английский

Citations

3