Primary Open-Angle Glaucoma Diagnosis from Optic Disc Photographs Using a Siamese Network

Lin, Mingquan; Liu, Lei; Gordon, Mae O.; Kass, Michael A.; Wang, Fei; Tassel, Sarah H. Van; Peng, Yifan

doi:10.1016/j.xops.2022.100209

Cited by 7 publications

(7 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, saliency and attention maps provided by the ViT model were more focused on neuroretinal rim areas than with the CNN models which showed more diffusely scored maps [14 ▪▪ ]. Longitudinal image acquisition may also be relevant in training AI algorithms as shown by Lin et al , using a Siamese CNN model [15]. Multitask transfer learning models have also been used to optimize generalizability and increase efficiency during the learning phase [16].…”

Section: Detecting Glaucoma With Artificial Intelligence In Different...mentioning

confidence: 99%

Artificial intelligence for glaucoma: state of the art and future perspectives

Correia Barão,

Hemelings,

Abegão Pinto

et al. 2023

Current Opinion in Ophthalmology

View full text Add to dashboard Cite

Purpose of review To address the current role of artificial intelligence (AI) in the field of glaucoma. Recent findings Current deep learning (DL) models concerning glaucoma diagnosis have shown consistently improving diagnostic capabilities, primarily based on color fundus photography and optical coherence tomography, but also with multimodal strategies. Recent models have also suggested that AI may be helpful in detecting and estimating visual field progression from different input data. Moreover, with the emergence of newer DL architectures and synthetic data, challenges such as model generalizability and explainability have begun to be tackled. Summary : While some challenges remain before AI is routinely employed in clinical practice, new research has expanded the range in which it can be used in the context of glaucoma management and underlined the relevance of this research avenue.

show abstract

Section: Detecting Glaucoma With Artificial Intelligence In Different...mentioning

confidence: 99%

Artificial intelligence for glaucoma: state of the art and future perspectives

Correia Barão,

Hemelings,

Abegão Pinto

et al. 2023

Current Opinion in Ophthalmology

View full text Add to dashboard Cite

show abstract

“…Es de suma importancia realizar un correcto diagnóstico clínico para tratar este tipo de afectación, primeramente llegando a la causa principal del glaucoma y a la vez poder diferenciarlo y clasificarlo a que clase pertenece, ya que cada uno de ellos presentan una clínica muy diferente si bien no hay causa aparente como tal, esto con la finalidad de prevenir complicaciones a largo plazo, pues no existe una prueba que sea sumamente verídica que sea de fácil acceso para todas las personas, de un intérprete que no sea muy complicado y que se pueda establecer en un sitio adecuado. Por el momento no se dispone de exámenes con esas características (27) (28).…”

Section: Diagnosticounclassified

“…El glaucoma, como alteración crónica, debe ser confirmado de manera temprana para evitar mayores complicaciones y a su vez sanos irreversibles como es la perdida de la vista, es por esto que se debe ser muy eficaz y realizar las pruebas pertinentes para poder detectar a este "Ladrón silencioso" en etapas iniciales o mitigables (28).…”

Section: Diagnosticounclassified

“…La enfermedad glaucomatosa es más sencilla identificar cuando hay presencia de signos característicos que nos indique que hay un posible daño en cuanto al nervio óptico y se puedan descartar otras etiologías. Dentro de las pruebas que se realizan por rutina esta la medición de la Presión intraocular mediante la tonometría y la exploración del nervio óptico en su máxima totalidad con el oftalmoscopio y otro tipo de pruebas complementarias (6) (30) (28).…”

Section: Diagnosticounclassified

See 1 more Smart Citation

Beneficios del abordaje clínico temprano en pacientes con glaucoma de ángulo abierto

Rios

Anaya

2023

View full text Add to dashboard Cite

El glaucoma de ángulo abierto es una afectación ocular más común y más prevalente en el mundo, por lo general esta enfermedad afecta a personas mayores de 40 años en un 90% de la población. El glaucoma se da cuando la presión intraocular se ve elevada debido a que el conducto se obstruye y el líquido no se puede drenar de manera normal. Este tipo de neuropatía óptica glaucomatosa es muy silenciosa y muchas de las veces no presenta ningún signo o síntoma que alerte a los pacientes, es por esto que es de suma importancia reconocer esta patología ocular en estadios tempranos, mediante un buen diagnóstico acompañado de una buena exploración ocular, con la finalidad de evitar complicaciones, como la pérdida gradual de la vista del lado que se presente la afectación, ya que es una de las principales complicaciones de este tipo de glaucoma. Para esto vamos se va a explicar la importancia del beneficio del manejo clínico temprano del glaucoma de ángulo abierto. Para esto se realizará un estudio no experimental, de tipo narrativo de revisión bibliográfica. Es importante el correcto manejo clínico de los pacientes que padecen de esta oftalmopatía, ya que de esta manera será menos complicado tratarlo de manera temprana, sin que esta evolucione y pueda generar o traer consigo misma múltiples afectaciones, que a la final va terminar deteriorando la vista del paciente y el estilo de vida como tal. Es por esto que es de gran importancia poder diagnosticar de manera oportuna y temprana este tipo de afectación ocular.

show abstract

“…With the accumulation of medical images and the development of artificial intelligence, deep learning (DL) algorithms have achieved unprecedented performance in the automatic diagnosis and lesion localization of various ophthalmic diseases, including eyelid tumors [ 7 , 8 ], keratitis [ 9 ], cataract [ 10 , 11 ], glaucoma [ 12 , 13 ], and diabetic retinopathy (DR) [ 14 , 15 ]. Among them, automatic diagnosis of eyelid tumors has received widespread attention from scholars and medical professionals due to their life-threatening potential and increasing frequency of incidence.…”

Section: Introductionmentioning

confidence: 99%

HM_ADET: a hybrid model for automatic detection of eyelid tumors based on photographic images

Jiang,

Liu,

et al. 2024

BioMed Eng OnLine

View full text Add to dashboard Cite

Background The accurate detection of eyelid tumors is essential for effective treatment, but it can be challenging due to small and unevenly distributed lesions surrounded by irrelevant noise. Moreover, early symptoms of eyelid tumors are atypical, and some categories of eyelid tumors exhibit similar color and texture features, making it difficult to distinguish between benign and malignant eyelid tumors, particularly for ophthalmologists with limited clinical experience. Methods We propose a hybrid model, HM_ADET, for automatic detection of eyelid tumors, including YOLOv7_CNFG to locate eyelid tumors and vision transformer (ViT) to classify benign and malignant eyelid tumors. First, the ConvNeXt module with an inverted bottleneck layer in the backbone of YOLOv7_CNFG is employed to prevent information loss of small eyelid tumors. Then, the flexible rectified linear unit (FReLU) is applied to capture multi-scale features such as texture, edge, and shape, thereby improving the localization accuracy of eyelid tumors. In addition, considering the geometric center and area difference between the predicted box (PB) and the ground truth box (GT), the GIoU_loss was utilized to handle cases of eyelid tumors with varying shapes and irregular boundaries. Finally, the multi-head attention (MHA) module is applied in ViT to extract discriminative features of eyelid tumors for benign and malignant classification. Results Experimental results demonstrate that the HM_ADET model achieves excellent performance in the detection of eyelid tumors. In specific, YOLOv7_CNFG outperforms YOLOv7, with AP increasing from 0.763 to 0.893 on the internal test set and from 0.647 to 0.765 on the external test set. ViT achieves AUCs of 0.945 (95% CI 0.894-0.981) and 0.915 (95% CI 0.860-0.955) for the classification of benign and malignant tumors on the internal and external test sets, respectively. Conclusions Our study provides a promising strategy for the automatic diagnosis of eyelid tumors, which could potentially improve patient outcomes and reduce healthcare costs.

show abstract

Primary Open-Angle Glaucoma Diagnosis from Optic Disc Photographs Using a Siamese Network

Cited by 7 publications

References 37 publications

Artificial intelligence for glaucoma: state of the art and future perspectives

Artificial intelligence for glaucoma: state of the art and future perspectives

Beneficios del abordaje clínico temprano en pacientes con glaucoma de ángulo abierto

HM_ADET: a hybrid model for automatic detection of eyelid tumors based on photographic images

Contact Info

Product

Resources

About