Ophthalmological ultrasonic automatic screening method and system based on deep learning

Abstract

The invention discloses an ophthalmological ultrasonic automatic screening method and system based on deep learning. The method comprises a model training stage comprising the steps of collecting historical ophthalmological ultrasonic pictures and videos and training the historical ophthalmological ultrasonic pictures and videos to obtain a segmentation model and a classification model, and a detection stage comprising the following steps of: S1, acquiring a to-be-detected ophthalmologic ultrasonic image; S2, performing artificial intelligence network segmentation on the to-be-detected ophthalmologic ultrasonic image by using the trained segmentation model, identifying the segmented ophthalmologic ultrasonic image by using the classification model to obtain a lesion category, and promptinga focus position; S3, evaluating a corresponding risk level according to the identified lesion; and S4, performing next diagnosis and treatment prompt according to the risk level. According to the invention, the workload of doctors can be reduced, the timely diagnosis rate of diseases can be improved, and meanwhile, specific lesion positions can be displayed, so that auxiliary training can be carried out on novice doctors.

Description

technical field [0001] The invention relates to the field of medical technology assistance, in particular to a method and system for automatic ophthalmic ultrasound screening based on deep learning. Background technique [0002] Ultrasonic diagnosis is an effective means of using the physical characteristics of ultrasound, that is, the reflection characteristics of sound waves, to apply to pathological changes in human organ tissues. It evaluates the detection site by imaging human organ tissues. , Intuitive advantages. Ultrasound examination in ophthalmology has almost become the only diagnosis and treatment method that can show intraocular diseases after refractive media opacity, and plays an irreplaceable important role in ophthalmology clinical auxiliary examination. Early diagnosis of blinding conditions such as cataracts can help reduce preventable vision loss. There is still a huge screening burden internationally and in places where medical resources are scarce, su...

AI Technical Summary

Problems solved by technology

However, ultrasound training standards are not uniform across countries

In low-income developing countries, even standard training is not possible

Images