소아에서 플러스옵틱스 에스09로 측정한 굴절력의 신뢰성과 유용성

임종훈,배지현,신승주,Jong Hoon Lim,Gi Hyun Bae,Seong Joo Shin 대한안과학회 2014 대한안과학회지 Vol.55 No.7

<b>Purpose:</b> To compare the refractive measurements obtained using a photorefractor (PlusoptiX S09, PlusoptiX GmbH, Germany) with those obtained using cycloplegic refraction in children. <b>Methods:</b> We assessed the refractive status of 268 eyes in 134 children. The values acquired via photorefraction with a PlusoptiX S09 device were compared with those obtained by cycloplegic retinoscopy. Hyperopia (≥+3.5 D), myopia (≥-3.0 D), with the rule or against the rule astigmatism (≥-1.5 D), and oblique astigmatism (≥-1.0 D) were set as diagnostic criteria for refractive amblyopia risk factors (RARFs). The difference in the detection of RARFs by the two methods was the main outcome measure. <b>Results:</b> The average spherical refractive power was -0.81 ± 1.68 D for PlusoptiX S09 versus -0.26 ± 2.00 D for cycloplegic retinoscopy (average difference -0.54 ± 0.61 D; <em>p</em> < 0.001). The average spherical equivalent was -1.20 ± 1.62 D for PlusoptiX S09 versus -0.64 ± 1.94 D for cycloplegic retinoscopy (average difference -0.56 ± 0.62 D; <em>p</em> < 0.001). The average cylinder power was -0.79 ± 0.93 D for PlusoptiX S09 versus -0.76 ± 0.94 D for cycloplegic retinoscopy (average difference -0.03 ± 0.33 D; <em>p</em> = 0.135). Even though cycloplegic retinoscopy is considered the gold standard, the sensitivity and specificity for detecting RARFs with the PlusoptiX S09 were 88.0% and 96.3%, respectively. <b>Conclusions:</b> PlusoptiX S09 is a relatively useful method for detecting RARFs, but the device tends toward myopic shift compared to cycloplegic refraction, and hyperopia is underestimated. J Korean Ophthalmol Soc 2014;55(7):1071-1076

빛간섭단층촬영에 따른 당뇨황반부종의 형태학적 분류와 치료 후 임상경과

임종훈,김인혁,배지현,김하경,배소현,Jong Hoon Lim,In Hyuk Kim,Gi Hyun Bae,Ha Kyoung Kim,So Hyun Bae 대한안과학회 2014 대한안과학회지 Vol.55 No.2

Purpose: To analyze the optical coherence tomographic patterns and clinical courses of patients with diabetic macular edema (DME) after treatment. Methods: The charts of 65 patients with DME were retrospectively reviewed. Baseline optical coherence tomographic patterns of DME were categorized into the 4 groups: group 1 (9 eyes, 13.8%) showed diffuse retinal thickening, group 2 (21 eyes, 32.3%) had cystoid macular edema (CME), group 3 (13 eyes, 20.0%) demonstrated serous retinal detachment (SRD) and group 4 (22 eyes, 33.9%) had combined CME and SRD. Treatments for DME included intravitreal bevacizumab/ triamcinolone injection, focal laser photocoagulation and vitrectomy. During 12 months of follow-up, changes in the patterns of DME were assessed. Additionally, the central retinal thickness (CRT) and best-corrected visual acuity (BCVA) were measured at baseline, 6 and 12 months. Results: During 12 months of follow-up, 21 eyes (32.3%) showed changes in the DME pattern: 2 eyes (22.2%) in group 1, 3 (14.3%) in group 2, 4 (30.8%) in group 3 and 12 (54.5%) in group 4. A significantly greater proportion of eyes with changes in DME pattern underwent vitrectomy compared with those without changes in DME pattern (p = 0.012). There was a significant difference in CRT among the 4 groups; group 4 demonstrated the largest CRT at baseline, 6 and 12 months (p < 0.001, 0.002 and 0.029, respectively). However, there were no significant differences in BCVA among the 4 groups at baseline, 6 or 12 months (p = 0.879, 0.375 and 0.246, respectively). Conclusions: Clinical courses varied according to the tomographic patterns of DME after treatment, and the poorest anatomic outcome was found in group 4. Change in tomographic pattern of DME was correlated with the treatment of DME, which might suggest a poorer outcome in those patients than in the patients who maintained their DME patterns. J Korean Ophthalmol Soc 2014;55(2):222-229

소아에서 저간섭성반사계, 자동각막곡률측정계, 각막지형도로 측정한 각막곡률측정치의 비교

정재균(Jae Gyun Jeung),배지현(Gi Hyun Bae) 대한안과학회 2016 대한안과학회지 Vol.57 No.10

Purpose: To compare the keratometry obtained from optical low-coherence reflectometry (Lenstar LS900®), autokeratometer (KR-8100®), and topography (Medmont E300®) in children. Methods: A retrospective study was performed in 316 eyes of 160 subjects. Subjects were divided into 3 groups according to age. Group 1 was younger than 10 years, group 2 was older than 10 years and younger than 18 years, and group 3 was older than 18 years. Subjects were tested using the Lenstar LS900®, KR-8100®, and Medmont E300®. Comparisons were made for steep K, flat K, mean K, and astigmatism among three groups. Agreement among the 3 devices was examined using the Bland-Altman method. Results: The keratometry measured by Medmont E300® was highest, followed by that of Lenstar LS900® and KR-8100® in all 3 groups. Almost all keratometry was significantly different among the 3 devices except for the flat K measured by LS900® and KR-8100® in all 3 groups and flat K measured by KR-8100® and Medmont E300® in Group 3 (p 〈 0.05). With regard to mean K, the agreement between Lenstar LS900® and KR-8100® was better than that between the other two pairs in both Groups 1 and 2, while agreement between Lenstar LS900® and Medmont E300® was better in Group 3. The agreement between KR-8100® and Medmont E300® was worse than that between the other two pairs in Groups 1 and 3, while the agreement between Lenstar LS900® and Medmont E300® was worse in Group 2. Conclusions: There were significant differences in keratometry among the 3 devices in all 3 groups. In children, Medmont E300® showed relatively less agreement compared with the other two devices. In adults, however, the agreement between 2 devices showed variable results. Consideration of the keratometry measurement from Lenstar LS900®, KR-8100®, and Medmont E300® might be helpful to estimate accurate corneal keratometry in children. J Korean Ophthalmol Soc 2016;57(10):1535-1541 목적: 소아에서 저간섭성반사계(Lenstar LS900®), 자동각막곡률측정계(KR-8100®), 각막지형도(Medmont E300®)로 측정한 각막곡률측정치를 비교하고자 한다. 대상과 방법: 160명 316안을 대상으로 Lenstar LS900®,KR-8100®,Medmont E300?瑛? 이용하여 flat K, steep K, mean K, astigmatism을 측정하였다. 10세 미만을 1군, 10세 이상 18세 미만을 2군, 18세 이상을 3군으로 나누어 측정치를 비교하였고, Bland-Altman plot을 이용하여 각 기기 간의 일치도를 분석하였다. 결과: 세 군 모두에서 대부분의 각막곡률측정치가 기기 간의 의미있는 차이를 보였다(p〈0.05).1군과 2군에서는 Lenstar LS900?玲? KR-8100?瑛? 일치도가 가장 높았고, Medmont E300?玲? 다른 두 기기 간의 일치도는 비교적 낮았다. 3군에서 flat K, astigmatism에서는 Lenstar LS900?玲? KR-8100?瑛? 일치도가 높고, Medmont E300?瑛? 일치도가 상대적으로 낮았다. 그러나 steep K와 mean K에서는 Medmont E300?瑛? 일치도가 높고, Lenstar LS900?玲? KR-8100?瑛? 일치도가 낮았다. 결론: 소아에서는 Medmont E300?瑛? Lenstar LS900®,KR-8100?怜? 비교하여 낮은 일치도를 보인 반면, 성인에서는 기기 간의 일치도가 일관되지 않고 각막곡률측정치 지표에 따라 다양한 결과를 보였다. 〈대한안과학회지 2016;57(10):1535-1541〉

시각 장애인을 위한 화장품 종류 인식

정신아(San Ah Jeong),배지현(Ji Hyun Bae),정호기(Ho Gi Jung) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6

There are many visually impaired people who use cosmetics, but it is said that they feel uncomfortable in distinguishing products with similar shapes. Therefore, if there is a system recognizing the kind and brand of cosmetics with similar shape (such as lip, mascara, eyebrow, and eyeliner) and informs them by voice, it will be able to relieve the inconvenience of the visually impaired. This paper proposes a system that automatically recognizes the type and brand of cosmetics from images taken with a mobile phone. For four types of cosmetics, we constructed a data set by directly taking 4400 cosmetics photos of two brands. As a result of training Yolov3 with this dataset, about 89.7% accuracy was confirmed.

접촉식 초음파와 세 가지 광학간섭계를 이용한 생체계측과 백내장수술 후 굴절력의 비교

문지선(Ji Sun Moon),신정아(Jeong Ah Shin),배지현(Gi Hyun Bae),정성근(Sung Kun Chung) 대한안과학회 2015 대한안과학회지 Vol.56 No.11

목적: 접촉식 초음파와 세 가지 광학간섭계를 이용하여 얻은 생체계측치를 비교하고, 백내장수술 후 굴절력 예측의 정확성을 비교하 고자 하였다. 대상과 방법: 백내장수술을 시행 받은 31안을 대상으로 접촉식 초음파, Lenstar LS900??, AL-Scan??, OA-2000?瑛? 이용하여 안축장, 전방깊이, 각막굴절력을 측정하였다. SRK/T 공식으로 인공수정체 도수를 결정했으며 예상 굴절력과 백내장수술 2개월 후의 실제 굴 절력 간의 오차를 비교하였다. 결과: 접촉식 초음파, Lenstar LS900??, AL-Scan??, OA-2000?瑛막? 측정된 안축장의 평균은 각각 23.39 ± 0.95 mm, 23.42 ± 0.98mm, 23.43 ± 0.98 mm, 23.44 ± 0.98 mm였으며 통계적으로 유의한 차이가 없었다( p =0.996). Lenstar LS900?? , AL-Scan??, OA-2000?? 의 전방깊이는 각각 3.14 ± 0.41 mm, 3.10 ± 0.38 mm, 3.13 ± 0.39 mm ( p =0.936), 평균 각막굴절력은 각각 44.41 ± 1.52D, 44.54 ± 1.57D, 44.44 ± 1.52D ( p =0.937)로 모두 유의한 차이가 없었다. 네 가지 계측 방법에서 평균 절대오차는 통계적으 로 유의한 차이가 없었다( p =0.722). 결론: 접촉식 초음파, Lenstar LS900??, AL-Scan??, OA-2000?? 을 이용하여 측정한 안축장, 전방깊이 및 각막굴절력은 높은 일치도를 보였으며, 백내장수술 후 굴절력 예측의 정확도는 네 가지 측정 방법 간 차이가 없었다. Purpose: To compare ocular biometry and refractive results measured using conventional applanation ultrasonography and 3 different optical interferometries, Lenstar LS900 ?? , AL-Scan ?? and OA-2000 ?? . Methods: The biometries of 31 cataractous eyes were measured using ultrasonography, Lenstar LS900 ?? , AL-Scan ?? or OA-2000 ?? . The axial length, anterior chamber depth and keratometry were measured. The SRK/T formula was used to calculate intraocular lens power. Two months after cataract surgery, the refractive outcome was determined and results from the 4 differ-ent biometry methods were compared. Results: Axial lengths were 23.39 ± 0.95 mm, 23.42 ± 0.98 mm, 23.43 ± 0.98 mm and 23.44 ± 0.98 mm measured using ultra-sonography, Lenstar LS900 ?? , AL-Scan ?? and OA-2000 ?? , respectively with no statistically significant differences observed (p = 0.996). The anterior chamber depth and keratometry were 3.14 ± 0.41 mm, 3.10 ± 0.38 mm and 3.13 ± 0.39 mm (p = 0.936) and 44.41 ± 1.52 D, 44.54 ± 1.57 D and 44.44 ± 1.52 D (p = 0.937) for Lenstar LS900 ?? , AL-Scan ?? and OA-2000 ?? respectively. There were no statistically significant differences between the 3 optical devices. The mean absolute error of the 4 different devices were not statistically significant (p = 0.722). Conclusions: The ocular biometric measurements and prediction of postoperative refraction using ultrasonography, Lenstar LS900 ?? , AL-Scan ?? or OA-2000 ?? showed no significant differences.

여러 가지 기기를 이용하여 측정한 전방깊이와 중심각막두께의 비교

정재균(Jae Gyun Jeung),길태영(Tae Young Gil),배지현(Gi Hyun Bae),신승주(Seong Joo Shin),정성근(Sung Kun Chung) 대한안과학회 2016 대한안과학회지 Vol.57 No.10

목적: 세 기기들을 이용하여 앞방깊이를 측정하고, 네 기기들을 이용하여 중심각막두께를 측정하여 각 기기 간의 정확성 및 임상적 유용성에 대해 알아보고자 하였다. 대상과 방법: 건강한 성인 90명의 180안을 대상으로 A-scan, Lenstar LS900Ⓡ, PentacamⓇ을 이용하여 앞방깊이를 측정하였고, 초음파 각막두께측정기, Lenstar LS900Ⓡ, PentacamⓇ, 전안부 빛간섭단층촬영계를 이용하여 중심각막두께를 측정하고 분석하였다. 결과: Lenstar LS900Ⓡ A-scan, PentacamⓇ을 이용하여 측정한 앞방깊이는 각각 3.27 ± 0.36 mm, 3.26 ± 0.36 mm, 3.25 ± 0.36 mm였고, 측정치 간에 유의하게 높은 양의 상관관계를 보였으며(p<0.001), 통계적으로 유의한 차이는 없었다(p=0.017). Bland Altman plot으로 분석한 일치도는 높지 않았다. 중심각막두께는 PentacamⓇ, 초음파 각막두께 측정계, Lenstar LS900Ⓡ, 전안부 빛간섭단층촬영계에서 각각 553.31 ± 25.23 μm, 547.26 ± 23.83 μm, 541.38 ± 24.49 μm, 531.40 ±22.33 μm로 측정치 간에 유의하게 높은 양의 상관관계를 보였으나(p<0.001) 통계적으로 유의한 차이를 보였다(p<0.05). Bland Altman plot으로 분석한 일치도는 높지 않았다. 결론: 여러 가지 기기로 측정한 앞방깊이 및 중심각막두께는 모두 상관성은 높으며, 앞방깊이는 유의한 차이가 없었으나, 중심각막두께는 유의한 차이를 보였다. 일치도는 두 군에서 모두 높지 않았다. <대한안과학회지 2016;57(10):1570-1576> Purpose: To compare the accuracy and clinical usefulness of different devices by measuring anterior chamber depth (ACD) with three devices and central corneal thickness (CCT) with four devices. Methods: In 180 eyes of 90 healthy subjects, ACD was measured using A-scan, Lenstar LS900??, Pentacam??, and CCT was measured using ultrasound pachymetry (USP), Lenstar LS900??, Pentacam??, and anterior segment optical coherence tomography (OCT). Results: The average ACT measurements using Lenstar LS900??, A-scan, and Pentacam?? were 3.27 ± 0.35 mm, 3.26 ± 0.36 mm, and 3.25 ± 0.36 mm, respectively. The measurements were significantly correlated (p < 0.001) but without statistically significant difference (p = 0.017). The Bland-Altman plots showed a low degree of agreement. The average CCT measurements using Pentacam??, USP, Lenstar LS900??, and OCT were 553.31 ± 25.23 μm, 547.26 ± 23.83 μm, 541.38 ± 24.49 μm, and 531.40 ± 22.33 μm, respectively. The measurements were significantly correlated (p < 0.001) and statistically significantly different (p < 0.05). The Bland-Altman plots showed a low degree of agreement. Conclusions: ACD and CCT measured using different devices were highly correlated, but the ACD measurements were not statistically different; however, the CCT measurements were statistically different, and agreement was low between both measurements. J Korean Ophthalmol Soc 2016;57(10):1570-1576