Table 2 Summary of related techniques
From: A smart intraocular pressure risk assessment framework using frontal eye image analysis
Algorithm | Ref. | Characteristics and performance | Data type |
|---|---|---|---|
Smartphone-based system for assessing intraocular pressure | [12] | • Measuring the eye pressure. • Smartphone used. • Patient must buy an adapter connected to camera. • Technique must be operated by a specialist (not the patient). • Video recording by smartphone camera. • Physical contact with eye in order to apply pressure on the cylinder. • Accuracy: N/A. | • Video frame images |
Glaucoma detection using intraocular pressure monitoring | [14] | • Monitoring IOP. • Use of special contact lenses with sensor. • Java software used to manage data and for simulation. • Specialist must help in order to install and remove the contact lenses sensor. • Acuracy: N/A. | • Contact lens sensor |
Micro-electromechanical pressure sensor for measuring intraocular pressure | [16] | • Measuring IOP based on P++silicon. • Finite element analysis (FEA) used for simulation. • Pressure applied on the eye. • Accuracy: N/A. | • Pressure sensor |
Glaucoma detection progression | • Visual field defect. • Glaucomatous progression. • Fundus images used. • Mixture of Gaussian and generalized expectation maximization (GEM) techniques. • Specificity 96%, sensitivity 87%; accuracy: N/A. | • Fundus Images | |
Proposed IOP risk assessment |  | • Segment the iris and pupil with accuracy of 95.30%. • Extract the sclera. • Measure the mean redness level with accuracy of 96.06%. • Measure the red area percentage with accuracy of 98.80%. • Measure area and height features of the extracted sclera contour. • Frontal images used. • No need for specialist to take the image. • Controlled environment (closed room with lights on). • SVM and neural network tested (neural network adopted). • 96.0% test phase accuracy. | • Regular camera images |