Reliability and Accuracy of Smart Eye Camera in Determining Grading of Nuclear Cataract
Article information
Abstract
Purpose
Cataracts are the cause of more than 50% of blindness in the world. Cataract is diagnosed using a slit lamp to evaluate the opacity of the lens. Slit lamps as the primary tools cataract are not available in primary care units. Smart Eye Camera (SEC) is a medical device placed on smartphone, capable of creating slit beam similar to conventional slit lamp. Previous studies have been conducted in Japan. We wanted to further compare ocular examination between SEC and conventional slit lamp with samples taken from UPTD Eye Clinic Ambon Vlissingen in Indonesia.
Methods
Observational analytical study with a cross-sectional study design. The research sample consisted of 67 people selected by consecutive sampling. All samples were examined for cataracts using a conventional slit lamp and SEC. To evaluate the accuracy of SEC, diagnosis of cataract was carried out by one observer based on slit-lamp images. The results of examinations were then compared with the examinations by two observers based on the SEC video, where the videos were read two times in a span of 1 week.
Results
Sixty-seven eye samples were taken with average age of 61.76 ± 5.83 years, visual acuity was 0.07 ± 0.12, intraocular pressure was 14.12 ± 2.06 mmHg, and SEC video duration was 4.15 ± 1.09 seconds. SEC intraobserver reliability has a κ of 0.795/0.818, SEC interobserver reliability has a κ of 0.795/0.817. SEC assessment accuracy for observer 1 has a κ of 0.606/0.681 and for observer 2 has a κ of 0.629/0.717, with the highest accuracy for cataract grade 3 (86.67% and 91.18%, respectively).
Conclusions
This study shows SEC has good consistency and reliability in assessing the degree of nuclear cataract compared to conventional slit lamp. This tool could be used for cataract screening in remote areas.
The eyes are the five most important senses for humans because humans absorb more than 80% of information through visual stimulation [1]. Blindness is defined by the World Health Organization (WHO) as visual acuity worse thatn 3 / 60. Globally, there are currently 2.2 billion people with blindness or visual impairment worldwide. Among them, about 1 billion people experience blindness or visual impairment caused by diseases that can be prevented or cured. Among the 1 billion population there are 94 million people with blindness caused by cataracts [2].
Based on the Rapid Assessment Avoidable of Blindness (RAAB) survey, cataracts are the main cause of world blindness, especially in people aged over 50 years. The latest RAAB data for 15 provinces in Indonesia for the 2014–2016 period found the prevalence of blindness in the population over the age of 50 years in Indonesia ranged from 1.7% to 4.4% with an average prevalence of blindness in Indonesia of 3.0%. This value is much higher than the ideal blindness rate according to WHO, which is less than 0.5%. The prevalence of blindness in Bali itself is still high at around 2%. Cataracts are the first cause of blindness in Bali with a percentage of 78% [1].
Blindness directly affects the quality of life, cognitive function and depressive mental status of the patient. Countries with a high percentage of blindness will certainly have a high economic burden. Globally, Marques et al. [3] estimates that blindness has caused a decrease in world labor productivity by 30.2%. While the estimated economic loss due to blindness using the gross domestic product is estimated at US $322.1 to $518.7 billion annually.
The World Health Assembly Global Action Plan (WHA-GAP) categorizes cataracts as a visual disorder that falls into avoidable blindness. The WHA-GAP targets a 25% reduction in avoidable obesity for 2010–2019, according to the WHO’s initial vision and mission in Vision 2020: Right to Sight. This target has not been achieved mainly due to the improvement of the world health system in general which has led to an increase in the life expectancy of the community in general. This causes degenerative diseases such as cataracts to have a fairly high prevalence increase [4].
Cataracts are opacity in the lens of the eye. The main factor causing cataracts is aging or age. Age-related cataracts are the cause of 50% of blindness in the world, so that cataracts are considered a major problem in world vision health [5]. Cataracts can be treated by performing cataract surgery. The standard for diagnosing cataracts is by having an eye examination performed by an ophthalmologist using a slit lamp. The slit lamp produces a thin, high-intensity beam of bright light (light gap). For the examination of nuclear type cataracts, the slit-lamp beam is focused at an angle of 45° to 60° and is directed to illuminate the center of the eye lens. The opacity seen in the lens of the eye is then assessed by comparing it to the standard opacity images using the Lens Opacities Classification System III (LOCS III) or the WHO Simplified Cataract Grading System (NUC) [6].
Cataract grading based on the NUC grading is a cataract assessment based on its nuclear part. This classification assesses the turbidity in the nucleus of the eye lens. When carrying out a NUC grading examination, only pay attention to the nucleus which is limited to the anterior and posterior parts of the nucleus. Grading according to the NUC is as follows (Fig. 1) [6]:
Grade NUC-0: Thinner than NUC-1
Grade NUC-1: Equal to or thicker than NUC-1 but thinner than NUC-2. The anterior and posterior embryonal nuclei are cloudier while the central portion is clearer and can be easily distinguished.
Grade NUC-2: Equal to or thicker than NUC-2 but thinner than NUC-3. The nucleus is cloudy and difficult to distinguish between the anterior and posterior parts, in the posterior part a cloudier lens is visible.
Grade NUC-3: Equal to or thicker than NUC-3. The cloudy nuclear portion is thickened to the periphery of the nuclear zone. The center of the nucleus itself is difficult to distinguish.
Grade NUC-9: Cannot be graded. This classification is used in further nuclear grading. For example, in Morgagni cataracts or when the assessment is difficult because the cataract has the dominant type of cortical cataract or subcapsular cataract.
Slit lamps are generally available at eye polyclinics or eye health centers. The disadvantage of this tool is that it is quite expensive and difficult to mobilize (nonportable). This makes cataract management difficult to reach remote places [2]. Based on the Regulation of the Indonesian Minister of Health Number 75 of 2014 concerning the Public Health Center are health service facilities that carry out public health programs as well as first-level individual health programs, by prioritizing promotive and preventive efforts [7]. The Public Health Center can act as a primary service unit for cataract screening, but due to limited funding and human resources these primary care units could not act as it should be [7].
The Smart Eye Camera (SEC) developed by OUI Inc is a medical device mounted in front of a smartphone camera, capable of creating light similar to a conventional slit lamp (Fig. 2). SEC was developed by Dr. Shimizu Eisuke from Keio University (Tokyo, Japan) and approved as a medical device in Japan (No. 13B2X10198030101). The SEC device is used in combination with the SEC APP. The SEC APP will capture images and then automatically store and upload the images into the SEC cloud system. This way, an ophthalmologist can access it from different places and diagnose an eye disease such as cataract without being present at the primary care unit, saving the cost and time of travel. In addition, with a lower price, easier to use and portable, SEC is expected to be used for cataract screening in primary care units [2].
The new era of technological development opens opportunities for the development of eye examinations using technology-based tools. The high cost of eye tools is one of the reasons for the development of eye examination technology using smartphones. The high resolution of the camera offered combined with the appropriate program will result in better work efficiency and effectiveness [8]. SEC is a portable device installed on a smartphone and converts the light from the smartphone into a slit beam, which allows the light to reach the eye. To convert the light into a 0.2- to 1.0-mm slit lamp, a cylindrical lens made of acrylic resin is used which is placed in front of the light source. The SEC slit beam is not fixed at an angle of 40° where this angle is considered to be the most commonly used in direct focal examination of the eye [9].
In front of the smartphone camera lens there is a detachable convex macro lens which is useful for adjusting the focus when examining the anterior segment of the eye. This lens was produced using a three-dimensional (3D) printer (Multi Jet Fusion 3D Model 4210, Hewlett-Packard Company) using polyamide 12. For the current study, it can only be used on an iPhone 7 (Apple Inc) which is used as the camera and light source. The resulting video resolution is set to 1080p with 30 frames per second, equivalent to 2.1 megapixels for each file [2].
As a new tool, research needs to be done to test its reliability and accuracy compared to the existing gold standard tools. Research has been conducted in Japan to test the SEC in diagnosing nuclear cataracts when compared with the results of examination using a conventional slit lamp. In this study, SEC reliability and accuracy tests were carried out on populations at the UPTD Ambon Vlissingen Eye Clinic (Kota Ambon, Indonesia). If the SEC has good reliability and accuracy, it can be used as a diagnostic tool for cataract screening in primary care units.
Materials and Methods
Ethics statement
The study protocol (No. 2022.02.1.1078) was reviewed and approved by the Research Ethics Commission of the Faculty of Medicine, Udayana University (No:2374/U N14.2.2.VII.14/LT/2022). Written informed consents for the research details and clinical images were obtained from the patients. All of the data included in this study were anonymized and handled with strict confidentiality. Data are stored in a password-protected computer and will be retained for 5 years before being completely deleted from the computer. This study was conducted in compliance with the International Conference on Harmonisation guidelines for good clinical practice.
Study design and setting
This research is an analytic observational study with a cross-sectional study design. The SEC and slit-lamp assessments are conducted at the same time. The SEC video assessment will be carried out by two observers while the enforcement of the patient’s nuclear cataract degree will be enforced by slit-lamp assessment as a reference which will be carried out by one observer.
The research sample was all nuclear cataract patients who came to the eye polyclinic UPTD Eye Clinic Ambon Vlissingen for the period between February and March 2023 who met the inclusion and exclusion criteria. The sampling method is to use the consecutive sampling method. All samples that came and met the inclusion and exclusion criteria were included in the study until the number of samples was fulfilled.
The inclusion criteria for this study are as follows: (1) aged over 40 years; (2) diagnosed with predominant nuclear type cataract; (3) pupils may be dilated >6.5 mm; and (4) willing to participate in research and sign informed consent. The exclusion criteria for this study were as follows: (1) abnormalities of the cornea, ocular infection, or inflammation which may affect video visualization; and (2) video quality is not up to the standard it should be.
Data collection was carried out by interview, visual acuity examination, eye pressure examination, and cataract examination using SEC and slit lamp. A research questionnaire was used for anamnesis and cataract examination, and mydriatic agents were prepared to dilate the patient’s pupil. Video of nuclear cataract examination using SEC and slit lamp is kept for accuracy and consistency by the observer.
Patients who were willing to be the research sample signed an informed consent. Patient given cycloplegic agent and the effect will disappear after approximately 4 hours after using the drug. After conducting an eye examination to ensure abnormalities in the eye that can affect the visualization of cataract grading. In addition to ensuring the patient’s dominant type of cataract, whether he enters NUC or not. A video was taken of nuclear cataract examination using a slit lamp. Videos are marked with a code number and inspection time. For grading NUC cataracts using a slit lamp, a setting with a magnification of ×10, medium brightness, beam angle of 45°, the beam height must be higher than the height of the nuclear part of the eye lens, and using a beam thickness of 0.1 mm [6]. The video capture distance is set to produce a successful video on the eyepiece. The slit lamp and SEC videos were taken in the same room so that the room lighting conditions are the same, thereby reducing bias. Slit-lamp video uses 1080p video resolution at 30 frames per second to match the SEC video resolution. Slit-lamp grading was determined on a different day than SEC grading. Videos are saved with the help of OUI software applications. Videos are tagged with a number code and inspection time.
The grading of nuclear cataracts was assessed based on the slit lamp and the grading of nuclear cataracts was carried out based on the SEC video by two observers. Observers have never examined a patient directly. The grading determination was carried out in a separate place randomly by the researcher. The determination of cataract grading based on SEC video was repeated 1 week later by the same two observers in separate places with a different video playback order from when the cataract was first determined. Observers who carry out grading assessments based on video have never directly examined patients so that grading of nuclear cataracts is purely based on video observations. The determination of cataract grading based on SEC video was carried out by two observers in separate places randomly without the observer knowing the slit-lamp grading of cataracts. The examination was repeated 1 week later with the same two observers in separate places with a different video playback order from when the cataract was first determined. The results of the inspection will be recorded to determine the reliability of the SEC video.
Statistical data analysis
The stages that were carried out in data management were editing, coding, data entry, and cleaning. This process was carried out using IBM SPSS ver. 23 (IBM Corp). Descriptive analysis was used to describe the general characteristics of the patient including age, sex, history of eye surgery, and the results of cataract grading using a slit lamp or SEC. Categorical data are presented in the form of frequency and percentage while data on a numerical scale is presented in the form of mean and standard deviation. Visual acuity examination, eyeball pressure, and eye examination were carried out using a conventional slit lamp. Visual acuity was measured using the Snellen chart. The results are converted into the logarithm of the minimum angle of resolution (logMAR) form to facilitate analysis.
The Normality test uses the Kolmogorov-Smirnov test (sample of more than 50) to determine whether the data is normally distributed (p > 0.05) or not (p < 0.05). The homogeneity test was carried out using the Levene test to find out whether the data was homogeneous (p > 0.05) or not (p < 0.05).
Intraobserver and interobserver reliability tests aimed to assess the reliability of nuclear cataract grading using SEC. Cataract grading assessment based on the SEC video will be carried out by two observers (two eye specialists). Each observer was asked to determine the grading of cataracts using the NUC classification two times with a random order of examinations and carried out 1 week after the first examination.
The intraobserver reliability test aims to assess whether the assessment of cataract grading by the observer is consistent when the same SEC video is shown for the second time. While interobserver reliability aims to assess whether the assessment of cataract grading between observer 1 and observer 2 is consistent. This test was carried out by cross-tabulation by comparing the ratings between observers. The suitability test will be analyzed using the categorical limit of agreement (κ) suitability test to assess the consistency of cataract grading examinations using the SEC by looking at the intraobserver and interobserver reliability.
Accuracy was assessed by comparing the grading of nuclear cataracts based on the SEC compared to conventional slit lamps. This test was carried out by cross-tabulation by comparing the ratings of the observers using the SEC compared to conventional slit lamps. The conformity test will be analyzed using the categorical limit of agreement (κ) suitability test to assess the accuracy of cataract grading examinations using the SEC (two observers) compared to conventional slit lamps.
Results
Characteristics of research subjects
The research sample was all nuclear cataract patients who came to the eye polyclinic UPTD Eye Clinic Ambon Vlissingen for the period between February and March 2023 who met the inclusion and exclusion criteria. The characteristics of the study sample are shown in Table 1, including the patient’s age, sex, visual acuity, laterality, and intraocular pressure.
Characteristics of research subjects (n = 67)
Characteristics of nuclear cataracts
In this study, cataract grading used the NUC classification. As a comparison, the SEC video used a slit-lamp specialist grading assessment at two different times as the gold standard. Of the 67 eyes used as the sample for this study, most of them were assessed as entering the grade NUC-2 as much as 47.8% in week 1 and 53.7% in week 2 (Table 2).
Cataract characteristics based on week 1 and 2 slit-lamp observations (n = 67)
Consistency of SEC examinations
Intraobserver reliability of cataract grading assessment using the SEC conducted by two observers at two different times. Based on the cross-tabulation of the nuclear cataract grading assessment of the two observers, there were similar ratings for grade NUC-1 (28.4% and 26.9%), grade NUC-2 (35.8% and 35.8%), and grade NUC-3 (22.4% and 25.4%). Total agreement between the two observers (86.6% and /88.1%) with a κ coefficient of 0.795/0.818 with a p-value of <0.001. It can be concluded that SEC intraobserver reliability is good.
Interobserver reliability of cataract grading assessments using SEC conducted by two observers at two different times. Based on the cross-tabulation of the nuclear cataract grading assessment of the two observers, there were similar ratings for grade NUC-1 (29.9% and 26.9%), grade NUC-2 (34.3% and 37.3%), and grade NUC-3 (22.4% and 23.9%). Agreement between the two observers (86.6% and 88.1%) with a κ coefficient of 0.795/0.817 with a p-value of <0.001. It can be concluded that SEC interobserver reliability is good.
Slit-lamp validity compared to SEC
Accuracy of slit-lamp cataract grading compared to SEC grading by two observers at two different times. Based on cross-tabulation, it was found that in week 1 there were equal scores for grade NUC-1 (20.9% and 22.4%), grade NUC-2 (32.8% and 32.8%), and grade NUC-3 (20.9% and 23.9%). Agreement between two observers (86.6% and 88.1%) with a κ coefficient of 0.606/0.681 with a p-value of <0.001. Meanwhile, based on cross-tabulation in week 2, there was an equal score for grade NUC-1 (17.9% and 17.9%), grade NUC-2 (35.8% and 40.3%), and grade NUC-3 (22.4% and 21.9%). Agreement between the two observers (76.1% and 82.1%) with a κ coefficient of 0.629/0.717 with a p-value of <0.001. It can be concluded that the accuracy of the SEC in determining cataract grading is comparable to observations made with a slit lamp where there is an increase in the accuracy of assessing cataract grading between the SEC and the slit lamp.
Accuracy of the slit-lamp cataract grading assessment compared to the SEC grading assessment conducted by two observers at two different times with a limit of agreement analysis for each cataract grading. Based on the calculation, it was found that the highest accuracy was in grading cataract 3 (hard) with an average accuracy of 88.93% (86.67% and 91.18%) (Table 3).
Slit-lamp accuracy of the SEC observer 1 and 2 in weeks 1 and 2 based on grading (n = 67)
Discussion
Characteristics of the research sample
The study sample of 67 eyes from 45 people were all nuclear cataract patients who came to the eye polyclinic UPTD Eye Clinic Ambon Vlissingen for the period between February and March 2023 who met the inclusion and exclusion criteria. The mean age of the study subjects was 61.76 ± 5.83 years, this result is in accordance with data from the WHO using the RAAB method in 15 provinces in Indonesia in the 2014–2016 period where data blindness and visual impairment in the population that occurs at the age of 50 years and over and the most common cause is cataracts [4,10,11]. According to Agarwal et al. [12], 48% of blindness in the world is caused by cataracts where most of them occur at the age of over 50 years. Globally, eye health services have not been able to keep pace with the increasing number of degenerative vision diseases such as cataracts, glaucoma, age-related macular degeneration, and refractive errors [4].
Judging from the sex of the research sample, 59.7% were female. This result is in line with the results obtained by Lou et al. [5] where the disability-adjusted life-year (DALY) cataract score for women was 67.0, higher than men with a score of 52.3. Globally, women have higher rates of cataracts, which is due to a lack of education and issues of gender equality. The DALY score is higher for women, especially in developing countries and in people over 50 years of age. Where this is in accordance with what we got in this research sample.
SEC and slit lamp comparison
This study evaluates the grading of NUC cataracts based on slit-lamp examination (the gold standard) compared to SEC conducted by two observers. The examination was repeated 1 week after the first examination. This aims to minimize recall bias. Before assessing the accuracy of SEC video compared to the slit lamp in assessing nuclear cataract grading, intraobserver and interobserver SEC reliability assessments were carried out. SEC intraobserver reliability was 86.6% and 88.1% with a κ of 0.793 and 0.818, and SEC interobserver reliability was 86.6% and 88.1% with a κ of 0.795 and 0.817. From these results it can be concluded that the observer has fairly good consistency in assessing the accuracy of the slit lamp and SEC. In the accuracy test, it was found that the accuracy of the slit lamp was SEC (week 1) 74.6% and 79.1%, κ of 0.606 and 0.681, while the slit-lamp accuracy was SEC (week 2) 76.1% and 82.1%, and κ of 0.629 and 0.717. Good reliability and accuracy were obtained in all statistical tests using κ, where agreement was obtained with a high κ-value (>0.600) and a p-value of <0.001. It can be concluded that the video produced by the SEC is not inferior when compared to the slit lamp when used to determine the NUC grading system. The accuracy of the slit lamp and SEC assessment in this study obtained a κ-value of 0.606/0.629 for observer 1, and 0.629/0.717 for observer 2. This κ result was lower when compared to the results obtained by Yazu et al. [2], which was 0.807. According to Cohen, a κ-value of 0.60 to 0.79 has a fairly strong agreement level, while a κ-value of >0.80 is a strong agreement level. The researchers considered that the difference in kappa values was due to the absence of a standardization of the SEC video capture like that of a slit lamp. For SEC, the OUI does not yet have a feature to change the intensity of light, setting the thickness and size of the light, the patient’s eyelid opening is less, and the patient’s head is not properly fixed are the factors that determine the quality of the collected SEC videos (Fig. 3). Researchers used a blepharostat to help open the patient’s eyes when taking videos, but this caused discomfort to the patient.
Changes in the position of the patient’s eyes from (A) the starting position to (B–E) going up and down can be seen due to instability in both the examiner’s hand and the patient’s head as the video progresses for 4 seconds.
An analysis was conducted to determine the accuracy of cataract grading based on slit-lamp video and SEC examination. Grade 3 cataracts demonstrated the highest accuracy (range, 86.67%–91.18%), surpassing the accuracy of grade 1 cataracts (range, 78.125%–85.71%) and grade 2 cataracts (range, 71.43%–70.84%). Observers argue that the assessment of grade 3 cataracts is easier due their distinguish color, thickness, and texture compared to grade 1 and 2 cataracts. The eradication of cataract blindness is focused on grade 3 cataract patients because their severe visual acuity impairment significantly disturb their daily activity.
Studies in 2014 estimate that 1.75 billion people in the world have smartphones. Abnormalities in the anterior eye segment were confirmed using a slit lamp connected to a computer and the images stored on the computer. The development of smartphones equipped with additional lenses that are easy to produce using a 3D printer is expected to replace the role of conventional slit lamps [13]. Currently, there are not many mobile devices that can be used to examine the anterior eye segment with a slit light. A tool similar to SEC from OUI has also been investigated by a study by Chen and Tan [14] using a Canon EOS 10D camera lens (Canon Inc) connected to an iPhone 5 (Apple Inc). Chiong et al. [13] researching a tool that is capable of producing a slit-light beam, but its size is still too thick so that the light is difficult to reach the eyepiece. The beam produced by SEC has the advantage that the resulting slit light has a thickness that is thin enough to reach the lens [9]. Besides being able to be used to assess cataracts, the SEC has also been tested and can be used to assess the depth of the anterior chamber [15], tear quality [9], as well as diagnosing anterior segment eye disorders such as conjunctivitis [16].
The average SEC video duration in our study was 4.15 ± 1.09 minutes, while the video duration by Yazu et al. [2] was 30.38 ± 6.27 minutes. The duration of the video is quite short so that the sampling process does not hinder the flow of patient examinations. In addition, the fast duration of video capture is intended for patient comfort. Video recordings of examinations stored on the iPhone can be uploaded web-based so that this feature is useful in telemedicine [2]. The video results that have been obtained can be combined with artificial intelligence systems, such as Tri-Oft (an embedded teleophthalmology application on smartphone for cataract screening that can be used in remote areas, where there is no access to doctors, specialized or not) [17], both for image processing and for cataract analysis without an observer.
Differences between the SEC and conventional slit lamp
As a portable device, SEC has different properties when compared with conventional slit lamp (Fig. 4). The feature available on slit lamp cannot be replicate onto SEC device completely yet. The most obvious one is that SEC did not have a head to chin rest as to stabilize patient’s head. This is understandable due to its specification for portability. Due to this, examiner’s hand which held the phone would not be as stable with slit-lamp device. This will results in the video from SEC examination to be shaky and could make the image out of focus.
Detailed images of (A) the Smart Eye Camera (OUI Inc) and (B) the conventional slit lamp. Note the large difference in size between the two; portability of SEC is far more superior than the conventional slit lamp.
Moreover, the video captured by SEC has smaller resolution in comparison to video captured by the camera of conventional slit lamp (Fig. 5). As of now, the SEC application installed on the phone can only set the resolution to HD (1280 × 720). Whereas, the resolution of the slit lamp used in this study which is KSL-H3-DR (Keeler) had resolution of 2056 × 1542. This means that slit lamp had 3.44 times more smoother image due to its larger amount of pixels.
Ocular images captured from (A–D) the slit-lamp camera and (E–H) the Smart Eye Camera (OUI Inc). Higher resolution ocular image was captured from the slit-lamp camera compared to the lower resolution ocular image captured using the SEC.
Despite all the disadvantages of SEC compared to slit lamp, SEC has its own advantages. The SEC device only weight 18 g and its dimension is very compact (70 mm × 25 mm × 15 mm). Its portability is far superior even when compared to the usual portable slit lamp. Apart from that, SEC also has an application intended for capturing images, saving and uploading video to the SEC cloud system. The uploaded video can be accessed by other ophthalmologists in other places via SEC website which is quite practical compared with its counterpart.
Research weaknesses
The drawback of the SEC tool when compared to conventional slit lamps is that they do not have a setting feature to change the size and width of the slit beam and light intensity. This feature certainly affects the quality of the stored video and will certainly make it difficult to assess cataracts based on SEC videos. Another drawback is the SEC data storage system which sometimes requires a long upload time. The SEC itself is still limited to its use for the iPhone 7, where this iPhone is no longer being produced. It is hoped that in the future this SEC can also be used for other types of cellphones.
Another obstacle the researchers felt was that when taking the video, some patients had difficulty opening their eyelids and fixing their heads properly. From the researcher’s point of view, the angle and distance of video capture does not yet have uniformity between videos, which of course affects the quality of SEC videos.
In conclusion, the determination of NUC grading system based on the SEC video yielded readings that were consistent with the SEC intraobserver reliability κ-value of 0.795/0.818 and the SEC interobserver reliability κ-value of 0.795/0.817. Determination of the NUC grading based on the SEC video results in accurate readings with observer 1 κ-values of 0.606/0.681 and observer 2 κ-values of 0.629/0.717 when reading SEC eye videos compared to cataract grading readings using a conventional slit lamp. Accuracy of grade 3 cataract readings has the highest accuracy of assessment (range, 86.67%–91.18%) compared to other grading (grade 1: range, 78.125%–85.71%; grade 2: range, 71.43%–70.84%).
Recommendations put forward by the authors, including standardization of the same room light, shooting angle, tool distance, sufficient eyelid opening, and good patient head fixation can produce better SEC video quality. It is necessary to develop the SEC tool so that it can have features for adjusting the light intensity, setting the thickness and width of the slit so that video capture can be done better.
Notes
Conflicts of Interest
None.
Acknowledgements
None.
Funding
None.