Abstract: Children are always innocent but they are not matured. Sometimes they are feared when they saw an unknown person. In the meantime, if they asked the children, they may answer properly or sometimes answered in a lunatic manner. The pediatric vision test is an example when we assume that children don’t know all the alphabets of the ‘Snellen chart’. Hence in this test, the ‘Tumbling E chart’ is used to determine the vision of the child. The ‘Tumbling E chart’ contains characters or objects like birds, fishes, homes etc. Sometimes children didn’t see those objects earlier due to which they don’t have any idea about all the objects or characters shown in the ‘Tumbling E chart’. Hence, they may see those objects properly but can"t define/name correctly about it. So, it may wrongly assess that the child can’t see properly which may carried on with wrong treatment. So, currently determination procedure of pediatric vision is highly uncertain which may lead to wrong treatment. Hence, to overcome the above mention uncertainty in currently running pediatric vision test, I have developed a vision testing system that is fully autonomous and independent of child’s knowledge. In this system, the child has to sit in front of the machine for few minutes and the machine will detect the vision of that child likes Ucvn, Bcvn, Pgvn, PH, Near, NCT, observation of retina, machine’s opinion. The machine also gives results about the IPD. Apart from that the machine will advise for left and right eye Dsph, DCyl, Axis.
Description:Vision testing process always face a serious issue on accuracy for children. So, the unique process or algorithm is developed to get the best testing reports with maximum accuracy. At first, a screen will be placed nearest to the eyes. There will be a plano-convex lens as shown in FIG. 1, which makes the screen distance 20 feet from eyes. This plano-convex lens will be parallel plated as looking to FIG. 1 for getting the best quality image at a distance of 20 feet from eyes. This process will reduce the conventional machine sizes. The screen will show mainly a scrollable ‘Snellen Chart’, which contains the different size or shapes of the child’s mother’s picture, that may be dynamic input. All the pictures will be set up according to the ‘Snellen fractions’ or ‘Visual acuity’. This process will continue for a few minutes and for both the eyes separately. In the meantime, the baby or a mature person must tries to focus on these pictures and his/her facial expressions will be changed accordingly. When the person can focus the picture properly, all the facial expressions will be relaxed as shown in FIG. 2, FIG. 3, FIG. 4. Then there will be two possibilities of set ups as shown in FIG.5 and FIG. 6. One of them will be a horizontal set up with night vision enabled camera with lights, placed mostly perpendicular to the forehead as shown in FIG. 5. When the baby or a mature person is trying to focus on the picture ,their facial landmarks will be followed by the camera. Whenever the facial expressions get relaxed it will be detected by the camera with testing of facial landmarks and equivalent relaxed distance measurements as shown in FIG. 7. Mainly the image processing will focus on the movement of eyebrows, forehead, lip. When the camera can detect the relaxed face, at that time shown picture will give the ‘Visual acuity’ of that eye. In this way, the vision of both eyes can be detected perfectly. This process can be repeated few more times to get a more accurate result.
The relaxed facial expressions can be detected with another low cost set up using LDR and Laser Beam as shown in FIG .6. According to the facial expressions, the forehead will be shrivelled and decreases or increases the illuminance or fallen light intensity to LDR. In this way, a relaxed time can be detected. By repeating any of the processes the machine can easily tell about the vision for both eyes and can advise required correction/medication.
Some machine learning algorithms can be used here for the best prediction. After recording age of baby and detecting his vision by providing the data of babies’ vision test report, we can make a machine intelligence from previous cases with comparing the recently collected data. We can predict the best possible vision test report likes Ucvn, Bcvn, Pgvn, PH, Near, NCT, observation of retina, machine’s opinion. The machine also gives results about the IPD. Apart from that the machine will advise for left and right eye Dsph, DCyl, Axis.
Claims:Claim1. The testing process is a uniquely designed algorithm, also comparatively less space required for PEDIATRIC VISION TESTING and the machine will not be misguided by any misunderstanding because of its automation.
Claim2. Addition with Claim1, we are placing a screen in a distance very closer to eyes in between eyes and screen there will be a plano-convex lens which makes the object distance at 20 feet.
Claim3. In addition to Claim2, now the screen will show a chart of different sizes of pictures of baby’s mother according to the ratio of visual acuity.
Claim4. In addition to Claim3, the baby will try to see his mother properly within the test time which runs for few minutes. At some time, his/her facial expression will be very relaxed when he/she focuses the shown image properly.
Claim5. Addition with Claim4, his/her relaxed face will be detected by night vision enabled camera with the help of image processing. Mainly the image processing will focus on changes in eyebrow, forehead and lip expressions. The relaxed face can be determined by a low cost set up using LDR and laser beam test. When the relaxed face will be detected, we can easily identify the vision of the baby for that particular eye.
Claim6. In addition to Claim5, by repeating this process for both the eyes, the machine can easily detect about the baby’s vision accurately and can advise further treatment accordingly.
Claim7. Addition with Claim6, after entering the age of baby and detecting his/her vision by providing the data of babies’ vision test reports, we can make a machine intelligence from previous cases with comparing the recently collected data. We can predict best possible the vision test report likes Ucvn, Bcvn, Pgvn, PH, Near, NCT, observation of retina, machine’s opinion. The machine also gives results about the IPD. Apart from that the machine will advise for left and right eye Dsph, DCyl, Axis.
| # | Name | Date |
|---|---|---|
| 1 | 201931036823-FORM 1 [12-09-2019(online)].pdf | 2019-09-12 |
| 2 | 201931036823-DRAWINGS [12-09-2019(online)].pdf | 2019-09-12 |
| 3 | 201931036823-COMPLETE SPECIFICATION [12-09-2019(online)].pdf | 2019-09-12 |
| 4 | 201931036823-FORM 18 [08-12-2022(online)].pdf | 2022-12-08 |
| 5 | 201931036823-FER.pdf | 2023-07-12 |
| 1 | SearchHistoryE_07-07-2023.pdf |