Human Versus Machine: Comparing a Deep Learning Algorithm to Human Gradings for Detecting Glaucoma on Fundus Photographs

To compare the diagnostic performance of human gradings vs predictions provided by a machine-to-machine (M2M) deep learning (DL) algorithm trained to quantify retinal nerve fiber layer (RNFL) damage on fundus photographs. Evaluation of a machine learning algorithm. An M2M DL algorithm trained with R...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:American journal of ophthalmology Ročník 211; s. 123 - 131
Hlavní autoři: Jammal, Alessandro A., Thompson, Atalie C., Mariottoni, Eduardo B., Berchuck, Samuel I., Urata, Carla N., Estrela, Tais, Wakil, Susan M., Costa, Vital P., Medeiros, Felipe A.
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States Elsevier Inc 01.03.2020
Elsevier Limited
Témata:
Aged
Algorithms
Area Under Curve
Cross-Sectional Studies
Datasets
Deep Learning
Defects
Diabetic retinopathy
Female
Fundus Oculi
Glaucoma
Glaucoma, Open-Angle - diagnosis
Glaucoma, Open-Angle - diagnostic imaging
Gonioscopy
Humans
Intraocular Pressure - physiology
Male
Middle Aged
Nerve Fibers - pathology
Neural networks
Optic nerve
Optic Nerve Diseases - diagnosis
Optic Nerve Diseases - diagnostic imaging
Optics
Photography
Physical Examination
Retinal Ganglion Cells - pathology
Retrospective Studies
ROC Curve
Software
Tomography
Tomography, Optical Coherence
Vision Disorders - diagnosis
Visual Field Tests - methods
Visual Fields - physiology
ISSN:0002-9394, 1879-1891, 1879-1891
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:To compare the diagnostic performance of human gradings vs predictions provided by a machine-to-machine (M2M) deep learning (DL) algorithm trained to quantify retinal nerve fiber layer (RNFL) damage on fundus photographs. Evaluation of a machine learning algorithm. An M2M DL algorithm trained with RNFL thickness parameters from spectral-domain optical coherence tomography was applied to a subset of 490 fundus photos of 490 eyes of 370 subjects graded by 2 glaucoma specialists for the probability of glaucomatous optical neuropathy (GON), and estimates of cup-to-disc (C/D) ratios. Spearman correlations with standard automated perimetry (SAP) global indices were compared between the human gradings vs the M2M DL–predicted RNFL thickness values. The area under the receiver operating characteristic curves (AUC) and partial AUC for the region of clinically meaningful specificity (85%-100%) were used to compare the ability of each output to discriminate eyes with repeatable glaucomatous SAP defects vs eyes with normal fields. The M2M DL–predicted RNFL thickness had a significantly stronger absolute correlation with SAP mean deviation (rho=0.54) than the probability of GON given by human graders (rho=0.48; P < .001). The partial AUC for the M2M DL algorithm was significantly higher than that for the probability of GON by human graders (partial AUC = 0.529 vs 0.411, respectively; P = .016). An M2M DL algorithm performed as well as, if not better than, human graders at detecting eyes with repeatable glaucomatous visual field loss. This DL algorithm could potentially replace human graders in population screening efforts for glaucoma.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Article-2
ObjectType-Feature-1
content type line 23
ObjectType-Undefined-3
ISSN:0002-9394
1879-1891
1879-1891
DOI:10.1016/j.ajo.2019.11.006