Publication
Análise de sinais ultrassónicos na caraterização de catarata ocular
| Summary: | Cataracts result from an aggregation of proteins present in the lens of the humaneye, which over time can cause partial or total loss of vision. Currently, about65.2 million people suffer from cataracts, making it the leading cause of visionloss worldwide. The current treatment for cataracts is surgery.The most used surgical treatment to remove the cataract and recover visual acuityis called phacoemulsification. This procedure requires the use of high-intensityultrasound or laser beams to fragment the cataractous lens followed by aspirationand implant of an artificial intraocular lens.The success of this technique heavily depends on the application of an appropriateenergy level for lens fragmentation. An incorrect estimation of this energylevel could result in the rupture of the lens’s posterior capsule and the loss of cornealendothelial cells. Currently, the energy level of the ultrasound is estimatedbased on a qualitative assessment of cataract hardness.For this purpose, images of the lens are observed through a slit lamp attached toa microscope to assess its opacity.In this study, using ultrasonic signals (A-scans), the aim is to locate the maininterfaces of the human eye (anterior and posterior lens) and detect the stiffnessassociated with each type of cataract (incipient cataract or mature cataract), if thisexists.The propagation of ultrasonic signals will interact with the different media of theeye’s structures, returning an echo from each interface through which the signalpasses through. By this echo, it will be possible to analyze the position of eyestructures, as well as the location and hardness of the cataract.In the initial phase, using a rule-based algorithm, the goal is to detect the eye’sinterfaces. Subsequently, employing machine learning methods, the aim is toautomatically detect and classify the existing type and severity of cataract. |
|---|---|
| Main Authors: | Neto, Marco António Oliveira |
| Subject: | Cataract Phacoemulsification Ultrasound A-scans Machine Learning Catarata Facoemulsificação Ultrassom A-scans Machine Learning |
| Year: | 2023 |
| Country: | Portugal |
| Document type: | master thesis |
| Access type: | open access |
| Associated institution: | Universidade de Coimbra |
| Language: | Portuguese |
| Origin: | Estudo Geral - Universidade de Coimbra |
| Summary: | Cataracts result from an aggregation of proteins present in the lens of the humaneye, which over time can cause partial or total loss of vision. Currently, about65.2 million people suffer from cataracts, making it the leading cause of visionloss worldwide. The current treatment for cataracts is surgery.The most used surgical treatment to remove the cataract and recover visual acuityis called phacoemulsification. This procedure requires the use of high-intensityultrasound or laser beams to fragment the cataractous lens followed by aspirationand implant of an artificial intraocular lens.The success of this technique heavily depends on the application of an appropriateenergy level for lens fragmentation. An incorrect estimation of this energylevel could result in the rupture of the lens’s posterior capsule and the loss of cornealendothelial cells. Currently, the energy level of the ultrasound is estimatedbased on a qualitative assessment of cataract hardness.For this purpose, images of the lens are observed through a slit lamp attached toa microscope to assess its opacity.In this study, using ultrasonic signals (A-scans), the aim is to locate the maininterfaces of the human eye (anterior and posterior lens) and detect the stiffnessassociated with each type of cataract (incipient cataract or mature cataract), if thisexists.The propagation of ultrasonic signals will interact with the different media of theeye’s structures, returning an echo from each interface through which the signalpasses through. By this echo, it will be possible to analyze the position of eyestructures, as well as the location and hardness of the cataract.In the initial phase, using a rule-based algorithm, the goal is to detect the eye’sinterfaces. Subsequently, employing machine learning methods, the aim is toautomatically detect and classify the existing type and severity of cataract. |
|---|
Funded activities
No funded projects found.