Hasil Pencarian  ::  Simpan CSV :: Kembali

"Latar Belakang: Meibomian Gland Dysfunction (MGD) adalah kelainan kelenjar Meibom yang bersifat kronik dan difus, ditandai oleh obstruksi duktus terminalis dan atau perubahan kualitatif serta kuantitatif sekresi kelenjar. Kelainan ini sering ditemui di praktik klinis dan merupakan penyebab utama Dry Eye Disease (DED). Meibografi adalah studi pencitraan yang khusus menilai morfologi kelenjar Meibom secara in vivo. Meibografi penting dilakukan rutin untuk diagnosis, evaluasi terapi dan alat edukasi pasien. Meibografi selama ini menggunakan teknik inframerah tetapi alat mahal dan tidak selalu tersedia. Penelitian ini ingin mengetahui validitas dan reliabilitas teknik meibografi filter merah oleh smartphone dibandingkan dengan meibografi inframerah dalam menilai MG (Meibomian Gland) dropout. Metode: Penelitian ini merupakan penelitian observasional analitik dengan desain potong lintang. Sebanyak 35 orang (68 mata) dengan kecurigaan MGD berdasarkan keluhan dan kelainan morfologi kelopak dilibatkan dalam penelitian ini. Kelenjar Meibom subjek penelitian difoto menggunakan dua jenis smartphone (Samsung S9 dan iPhone XR) pada slitlamp yang ditambahkan filter merah. Gambar yang dihasilkan kemudian dinilai secara subjektif menggunakan meiboscore oleh dua orang penilai dan persentase dropout dinilai dengan aplikasi komputer ImageJ. Hasil: Tidak ada kesesuaian penilaian meiboscore antara teknik meibografi filter merah oleh kedua merk smartphone dengan meibografi inframerah. Kesesuaian kedua teknik dalam penilaian persentase dropout menggunakan ImageJ menunjukkan nilai kesesuaian yang rendah. Namun, terdapat korelasi positif antara meiboscore dengan persentase dropout menggunakan ImageJ pada teknik meibografi filter merah oleh kedua jenis smartphone walaupun nilai korelasinya lemah. Inter-rater reliability penilaian meiboscore teknik meibografi filter merah menunjukkan tidak ada kesesuaian antara kedua penilai. Intra-rater reliability penilaian meiboscore dari teknik meibografi filter merah oleh smartphone Samsung dan iPhone menunjukkan kesesuaian yang lemah pada penilai 1 dan tidak ada kesesuaian pada penilai 2. Kesimpulan: Validitas dan reliabilitas teknik meibografi filter merah oleh smartphone kurang baik dalam menilai dropout dibandingkan dengan meibografi inframerah. Nilai intra- dan inter-rater reliability yang rendah pada semua teknik pemeriksaan meibografi menunjukkan perlunya penelitian lanjut tentang penilaian subjektif.

---

Background: Meibomian Gland Dysfunction (MGD) is defined as a diffuse abnormality of the Meibomian glands initiated through occlusion of its terminal ducts and/or changes in the glandular secretion. MGD is one of the most common disorders encountered in ophthalmic practice and a leading cause of Dry Eye Disease (DED). Meibography is a specialized imaging study developed for the purpose of directly visualizing the morphology of meibomian glands in vivo. It is important to be performed routinely for diagnosis, treatment evaluation and educational tools for patients. However, up until now meibography use infrared technique which are costly and not readily available. This study aimed to compare the validity and reliability of red filter meibography technique by smartphone to infrared meibography in evaluating Meibomian gland dropout. Methods: This is an analytical cross sectional study. A total of 35 subjects (68 eyes) with suspected MGD according to symptoms and abnormality of lid morphology were included in this study. Meibomian glands of each subject were captured using two types of smartphone (Samsung S9 and iPhone XR) through slitlamp on which we added red filter in front of the light source. Images are rated subjectively using meiboscore by two rater and dropout percentage are rated using ImageJ applications. Results: There are no agreement in meiboscore rating between red filter meibography by two smartphones with infrared meibography technique. There is also minimal level of agreement between techniques in evaluating dropout percentages using ImageJ. However, there is positive but low correlation between meiboscore and dropout percentage using ImageJ in red filter meibography by two smartphones. Inter-rater reliability of meiboscore show no agreement between two rater. Intra-rater reliability of meiboscore from red filter meibography by Samsung and iPhone smartphone demonstrated a weak level of agreement in rater 1 and no agreement in rater 2. Conclusion: The validity and reliability of red filter meibography by smartphones was not satisfactory in evaluating dropout compared to infrared meibography. The low value of intra- and inter-rater reliability on all meibography techniques indicate the need for further research on subjective assessment."

"Latar belakang: Prevalensi meibomian gland dysfunction (MGD) dilaporkan bervariasi pada rentang 3,6-69,3% karena modalitas diagnostik yang tersedia saat ini masih belum terstandar secara baku. Penilaian meibomian gland (MG) dropout secara manual masih terbatas oleh subjektivitas penilai dalam identifikasi MG, kurang akurat dalam menilai perubahan longitudinal, serta memerlukan waktu dan biaya yang lebih besar. Penelitian ini bertujuan untuk mengetahui apakah performa diagnostik dari penilaian MGD melalui meibografi dengan bantuan AI setara dengan penilaian MG dropout oleh klinisi menggunakan ImageJ. Metode: Penelitian dilakukan dengan desain cross-sectional dari pasien rawat jalan Rumah Sakit Dr. Cipto Mangunkusumo (RSCM) Kirana, Jakarta Pusat. Pengolahan data citra meibografi dilakukan dengan dua tahap preprocessing dan pengembangan model artificial intelligence (AI). Pengembangan model AI yang dilakukan menggunakan image embedding VGG16 dan model multilayer perceptron (MLP) pada Orange v3.32.0. Hasil: Dari 35 subjek penelitian dengan rerata usia 60,29±2,28 tahun, terdapat 136 data citra meibografi yang dianalisis. Nilai cut-off MG dropout yang terbaik pada nilai 33% yang mana terdapat 107 citra MGD dan 29 citra normal. Model AI menunjukkan performa AUC 83,2%, sensitivitas 89,7%, dan spesifisitas 58,6%. Kesimpulan: Penilaian meibografi dengan bantuan AI memiliki performa diagnostik yang baik dalam deteksi MGD. Pendekatan dengan AI dapat digunakan sebagai alat skrining potensial yang efektif dan efesien dalam praktik klinis.

---

Introduction: The prevalence of meibomian gland dysfunction (MGD) is reported to vary in the range of 3.6-69.3% because the currently available diagnostic modalities have not been standardized. Manual assessment through meibomian gland (MG) dropout is still has many limitations, such as the subjectivity of the assessor in identifying MG, less accuracy in assessing longitudinal abnormalities and requires more time and costs. This study aims to determine whether the diagnostic performance of MGD assessment through AI-assisted meibography is equivalent to MG dropout assessment by the clinician using ImageJ. 

Methods: The study was conducted with a cross-sectional design from outpatients at Dr. Cipto Mangunkusumo Hospital (RSCM) Kirana, Central Jakarta. The meibography image processing is conducted in two stages preprocessing and the development of artificial intelligence (AI) models. AI model development uses Orange v3.32.0 with VGG16 as image embedding and a multilayer perceptron (MLP) model. 

Results: From 35 subjects with a mean age of 60.29±2.28 years, a meibography dataset was built from 136 eyelid images. Using the MG dropout cut-off value of 33%, there are 107 MGD images and 29 normal images. The AI model showed an AUC performance of 83.2%, a sensitivity of 89.7%, and a specificity of 58.6%. 

Conclusion: AI-assisted meibography assessment has good diagnostic performance in MGD detection. The AI approach has promising potential as an effective and efficient screening tool in clinical practice."

"Mata merupakan salah satu bagian tubuh yang penting pada hidup manusia. Menggunakan bantuan mata, kita dapat menjalankan berbagai macam aktivitas dengan mudah. Namun, banyak sekali penyakit yang dapat menyerang mata, salah satunya adalah mata kering. Sebuah studi yang ada telah mengkonfirmasi bahwa sebagian besar pasien dengan penyakit mata kering dilaporkan mengalami disfungsi kelenjar meibom. Oleh karena itu, sangat penting untuk mengevaluasi kinerja kelenjar meibom pada pasien mata kering. Akan tetapi, pada kenyataannya hasil evaluasi kelenjar meibom oleh tenaga profesional masih sangat subjektif. Seorang dokter mata bisa memiliki pendapat mengenai tingkat kerusakan kelenjar meibom yang berbeda dengan dokter lainnya. Sehingga, alat diagnostik yang efektif diperlukan untuk mengevaluasi kelenjar meibom agar terhindar dari hasil penilaian tenaga profesional yang subjektif. Oleh sebab itu, pada penelitian ini dilakukan segmentasi kelenjar meibom dengan bantuan deep learning untuk menghindari penilaian tenaga profesional yang subjektif. Penelitian ini menggunakan arsitektur yang bernama U-Net. Data yang dimiliki berjumlah 139 citra meibography berasal dari pasien penyakit mata kering dari Rumah Rumah Sakit Cipto Mangunkusumo Departemen Kirana yang terdiri dari 35 citra meibography kelopak mata atas pada mata kanan, 34 citra meibography kelopak mata atas pada mata kiri, 35 citra meibography kelopak mata bawah pada mata kanan, dan 35 citra meibography kelopak mata bawah pada mata kiri. Kemudian citra meibography melalui tahapan anotasi untuk mendapatkan ground truth dan di resize menjadi ukuran 256 x 256. Selanjutnya data tersebut mengalami augmentasi dengan teknik rotasi dan teknik horizontal flip. Sehingga total data citra meibography menjadi 417 citra. Pada penelitian ini data citra meibography dibagi menjadi 3 bagian yaitu data training, data validation, dan data testing. Pada kasus pertama, jumlah data training adalah 80% dari citra meibography yang dimiliki, data validation sebanyak 10% citra meibography dari data training, dan data testing sebanyak 20% citra meibography yang dimiliki. Pada kasus kedua, pembagian data training dan data testing masih sama akan tetapi pembagian data validation adalah 20% dari data training. Pada kasus terakhir pembagian data training dan data testing masih sama akan tetapi pembagian data validation adalah 30% dari data training. Dengan melakukan 5 kali percobaan untuk masing-masing kasus pembagian data, didapat bahwa kasus pertama menghasilkan rata-rata akurasi 94,50% dan rata-rata Intersection over Union (IoU) 72,70%, kasus kedua menghasilkan nilai rata-rata akurasi 94,49% dan rata-rata Intersection over Union (IoU) yaitu 73,86%, dan kasus terakhir memiliki rata-rata akurasi 94,14% dan Intersection over Union (IoU) 72,15%.

---

The eye is one of the essential body parts in human life. With the eye's help, we can carry out various activities easily. However, many diseases can attack the sights, including dry eyes. An existing study has confirmed that most patients with dry eye disease reported meibomian gland dysfunction. Therefore, it is crucial to evaluate the performance of the meibomian glands in dry eye patients. However, the results of the evaluation of the meibomian glands by professionals are still very subjective. An ophthalmologist may have an opinion regarding the level of meibomian gland damage that is different from other doctors. Thus, an effective diagnostic tool is needed to evaluate the meibomian glands to avoid subjective professional assessment results. Therefore, in this study, segmentation of the meibomian glands was carried out with the help of deep learning to prevent subjective professional judgments. This research uses an architecture called U-Net. The data is 139 meibographic images derived from dry eye patients from Cipto Mangunkusumo Hospital Kirana Department consisting of 35 meibographic images of the upper eyelid on the right eye, 34 meibographic images of the upper eyelid on the left eye, 35 meibographic images of the lower eyelid in the right eye, and 35 meibography images of the lower eyelid in the left eye. Then the meibography image goes through the annotation stages to get the ground truth and is resized to a size of 256 x 256. Furthermore, the data is augmented using rotation techniques and horizontal flip techniques. So, the total meibography image data becomes 417 images. In this study, meibography image data is divided into three parts: training data, validation data, and testing data. In the first case, the amount of training data is 80% of the meibography image, validation data is 10% of the meibography image from the training data, and testing data is 20% of the meibography image. In the second case, the distribution of training data and testing data is still the same, but the distribution of validation data is 20% of the training data. In the last case, the training data distribution and testing data are still the same, but the distribution of validation data is 30% of the training data. By conducting five trials for each case of data division, it was found that the first case produced an average accuracy of 94.50% and an average Intersection over Union (IoU) of 72.70%, the second case made an average accuracy value of 94.49% and the average Intersection over Union (IoU) is 73.86%, and the third case has an average accuracy of 94.14% and Intersection over Union (IoU) 72.15%."

"Mata kering merupakan penyakit yang beredar pada masyarakat umum. Mata kering menyebabkan rasa tidak nyaman dan mengganggu aktivitas sehari-hari. Faktanya, lebih dari 85% penderita penyakit mata kering disebabkan kerusakan kelenjar meibom (meibomian gland dysfunction, MGD). Akibatnya mata yang memilki MGD menjadi kering karena intensitas evaporasi air mata meningkat. Untuk mendeteksi tingkat MGD dilakukanmeibography. Dari hasil meibography, klinisi (dokter spesialis mata) menilai tingkat MGD yang disebut meiboscore. Namun realitanya, penilaian meiboscore masih sangat subjektif antar para klinisi. Alat yang digunakan juga mahal dan tidak seluruh klinik mata memiliki alat tersebut. Oleh karena itu pada tugas akhir ini dilakukan deteksi tingkat kerusakan kelenjar meibom dengan pendekatan faktor-faktor potensi MGD dan machine learning. Metode machine learning yang digunakan dalam tugas akhir ini ini adalah radial basis function neural network (RBFNN). Metode machine learning dalam studi ini dilakukan Teknik SMOTE terelebih dahulu untuk menyeimbangkan jumlah data antar kelas, lalu data dibagi menjadi data training dan data testing dengan rasio sebesar 90%: 10%, 80%: 20%, 70%: 30%, dan 60%: 40% . Selain itu dilakukan pengurangan fitur-fitur yang kurang relevan menggunakan seleksi fitur Chi square. Hasil evaluasi metode RBFNN memperoleh nilai rata-rata akurasi, presisi, recall dan f1-score terbaik dicapai menggunakan data testing 20% dengan masing-masing mencapai nilai 96%, 95%, 100%, dan 95% secara berurut

---

Dry eye is a common disease happened among the public. Dry eye causes discomfort and distracts daily activities. More than 85% dry eye suffers are caused by meibomian gland dysfunction (MGD). As a result, eyes with MGD becomes dry due to high tear evaporation intensity. Detecting MGD can be done by meibography. The MGD level is scored by clinicians which is called meiboscore. However, scoring the meiboscore is still very subjective among the clinicians. The tool that is used are expensive and not all eye clinics have this tool. Therefore, this study aims to detect the MGD level with the approach of MGD potential factors and machine learning. In this study radial basis function neural network (RBFNN) is used. The machine learning method performs SMOTE technique to balance the amount of data in each class, then all data is divided into training data and testing data by90%: 10%, 80%: 20%, 70%: 30%, and 60%: 40% respectively. Moreover, irrelevant features are reduced to optimize using feature selection, Chi Square. To reduce the features that are less relevant, Chi square feature selection is performed. RBFNN method obtained the best average accuracy 96%, average precision 95%, average recall 100%, and average f1-score 95% using the 20% data testing."

"Latar belakang: Diagnosis dry eye disease (DED) ditegakkan dengan serangkaian pemeriksaan gejala subjektif dam tanda klinis objektif, namun sayangnya alat penunjang pemeriksaan tidak selalu dimiliki oleh fasilitas layanan kesehatan, sehingga kuesioner yang valid dan reliabel berperan sebagai alternatif untuk menegakkan diagnosis. Kuesioner dry eye Indonesia yang telah dikembangkan untuk populasi Indonesia masih belum tervalidasi dengan jumlah pertanyaan yang belum ideal. Tujuan: Mendapatkan item pertanyaan kuesioner dry eye Indonesia bagian diagnosis yang valid dan reliabel serta mengetahui korelasi klinis antara gejala subjektif dan tanda klinis objektif DED pada pasien. Metode: Penelitian ini terdiri dari tiga tahapan, yaitu (1) focus group discussion (FGD) untuk menilai content validity index (CVI), (2) pretesting untuk menilai cognitive debriefing, validitas, dan reliabilitas pada 30 sampel, dan (3) testing untuk menilai validitas, reliabilitas, dan korelasi klinis dengan tanda klinis objektif pada 60 partisipan. Partisipan melengkapi kuesioner dry eye Indonesia dan ocular surface disease index (OSDI), kemudian dilakukan uji tear break up time (TBUT), tear break up pattern (TBUP), pewarnaan okular, dan Schirmer. Hasil: Jumlah pertanyaan kuesioner dry eye Indonesia dari studi sebelumnya adalah 31 item. Dari hasil FGD dikerucutkan menjadi 14 item pertanyaan, dengan nilai CVI 0,98. Pada tahap pretesting, seluruh item dinyatakan dapat dipahami oleh seluruh subjek dengan nilai validitas dan reliabilitas baik. Dari hasil testing, didapatkan validitas yang dinilai dari corrected item total correlation dan reliabilitas yang dinilai dari Cronbach’s alpha yang baik pada 10 item pertanyaan kuesioner. Sensitivitas dan spesifisitas kuesioner didapatkan 91,1% dan 100% dengan AUC 98,2% (IK95% 94,4%-100%), nilai potong diagnosis DED adalah 10,5. Skor kuesioner dry eye Indonesia didapati berkorelasi positif kuat dengan skor OSDI (r=0,808; p<0,001) dan berkorelasi negatif lemah (r=- 0,339; p=0,008) dengan TBUT, namun tidak didapati korelasi yang bermakna terhadap Schirmer dan pewarnaan okular. Kesimpulan: Kuesioner dry eye Indonesia bagian diagnosis memiliki validitas, reliabilitas, serta sensitivitas dan spesifisitas yang sangat baik untuk mendiagnosis DED. Korelasi klinis antara skor kuesioner dry eye Indonesia didapatkan bermakna terhadap skor OSDI dan TBUT.

---

Background: Diagnosis of dry eye disease (DED) is established through a series of examinations of subjective symptoms and objective clinical signs. Unfortunately, diagnostic tools are not always available in healthcare facilities, making valid and reliable questionnaires an alternative for diagnosis. Indonesian dry eye questionnaire developed for the Indonesian population has not yet been validated with an ideal number of questions. Objective: To obtain valid and reliable diagnostic questions for the Indonesian dry eye questionnaire and to determine the clinical correlation between subjective symptoms and objective clinical signs of DED in patients. Methods: This study consists of three stages: (1) focus group discussion (FGD) to assess the content validity index (CVI), (2) pretesting to evaluate cognitive debriefing, validity, and reliability in 30 samples, and (3) testing to assess the validity, reliability, and clinical correlation with objective clinical signs in 60 participants. Participants completed the Indonesian dry eye questionnaire and the Ocular Surface Disease Index (OSDI), followed by testing tear break-up time (TBUT), tear break-up pattern (TBUP), ocular staining, and Schirmer. Results: Thenumber of questions in the Indonesian dry eye questionnaire from the previous study was 31 items, narrowed down to 14 items through FGD with a CVI value of 0.98. In the pretesting stage, all items were found to be understandable by all subjects with good validity and reliability. In the testing phase, 10 questionnaire items showed good validity assessed from corrected item total correlation and reliability assessed from Cronbach’s alpha. The questionnaire demonstrated a sensitivity of 91.1%, specificity of 100%, and an AUC of 98.2% (95% CI 94.4%-100%), with a diagnostic cutoff score for DED at 10.5. The Indonesian dry eye questionnaire score showed a strong positive correlation with OSDI score (r=0.808; p<0.001) and a weak negative correlation (r=- 0.339; p=0.008) with TBUT, but no significant correlation was found with Schirmer and ocular staining. Conclusion: The diagnostic section of the Indonesian dry eye questionnaire has excellent validity, reliability, sensitivity, and specificity for diagnosing DED. Clinical correlations were found between the questionnaire score and OSDI score and TBUT."

"Mata merupakan salah satu panca indra dan menjadi aset terpenting yang dimiliki oleh manusia dalam menjalani kehidupan sehari hari. Salah satu bagian terpenting dari mata adalah bagian kelopak karena terdapat sebuah kelenjar yaitu kelenjar meibom yang berfungsi untuk menyekresikan lipid dan berperan dalam menjaga kelembaban bola mata. Sehingga, permasalahan yang terjadi pada kelenjar meibom dapat menyebabkan suatu pernyakit yang disebut penyakit mata kering. Dikarenakan proses diagnosis yang dilakukan oleh dokter masih terbilang subjektif, disini penulis mengusulkan untuk menggunakan pendekatan deep learning untuk melakukan segmentasi pada citra kelenjar meibom atau citra meibography. Segmentasi dilakukan dengan membagi area kedalam 3 segmen (latar, kelenjar meibom, dan atrophy) yang diharapkan dapat membantu proses diagnosis tersebut. Metode deep learning yang digunakan dalam segmentasi ini adalah Metode SegNet yang merupakan salah satu model Convolutional Neural Network (CNN). Data yang digunakan pada penelitian ini merupakan data sekunder yang berasal dari 35 pasien penyakit mata kering di Rumah Sakit Ciptomangunkusumo (RSCM) Departemen Kirana dengan total 139 data citra yang terbagi atas 35 citra kelopak mata pada masingmasing bagian kanan atas, kanan bawah, dan kiri bawah. Sedangkan 34 citra kelopak mata bagian kiri atas. Pada tahap persiapan data, dilakukan pembuatan ground truth dengan proses anotasi. Pada tahap pre-processing, dilakukan resize citra menjadi ukuran 224 x 224 yang kemudian data dibagi menjadi 80% data training dan 20% data testing. Dari 80% data training, diambil 10% untuk dijadikan data validation yang kemudian kedua data training dan validation diterapkan teknik augmentasi yaitu rotation dan flip horizontal agar dataset yang digunakan dalam proses modelling bisa menjadi lebih banyak. Setelah augmentasi, jumlah data training, validation, dan testing berturut-turut menjadi 300, 33, dan 28 data. Kemudian dilakukan stacking pada citra asli dan one hot encoding pada ground truth. Training model dilakukan menggunakan model SegNet dengan hyerparameter model yaitu batch size 32, learning rate 0.0001, dan epoch sebanyak 300. Model juga diterapkan fungsi optimasi yaitu Adam (Adaptive moment estimation) dan fungsi loss categorical cross entropy. Proses modelling dilakukan sebanyak 10 kali percobaan dan berhasil memperoleh nilai rata-rata kinerja training model sebesar 99,31% dan 92,01% pada akurasi training dan akurasi validation-nya, diperoleh nilai 27,45% dan 44,33% pada loss training dan loss validation. Sedangkan rata-rata kinerja testing model berhasil memperoleh akurasi testing sebesar 92,99%, testing loss sebesar 0,4265 dan Mean-IoU sebesar 70,03%.

---

Eyes is one of the five senses that play a role to see a things, eyes also one of the most important asset that humans have. One of the most important parts of the eye is the eyelids, because there is a gland, called meibomian gland. Meibomian gland has a function to secrete the lipids and plays a role at keeping our eyes moist. So therefore. The problems that may occur at meibomian gland can cause a disease called dry eye disease. Because a diagnosis process that performed by doctors is still fairly subjective, right now the writer propose to use deep learning approach by segmenting meibomian gland images. Segmentation is done by dividing the area itu 3 segments (background, meibomian gland, and atrophy) which is expected to help the diagnosis process. The deep learning method used in this segmentation is the Segnet method, which is one of the Convolutional Neural Network (CNN) models. The data used in this study were the secondary data derived from 35 dry eye patients at Ciptomangunkusumo Hospital, Kirana Department with a total of 139 images data divided into 35 eyelid images on each of the upper right, lower right, and lower left. And 34 images of the upper left eyelid. During the data preparation, a ground truth was made by the annotation process which the marking area of segmentation was given directly by the relevant opthalmologists. At the pre-processing, the images and ground truths were resize to a size of 224 x 224, then divided into 80% training data and 20% testing data. From 80% of the training data, 10% is taken to used as validation data. Then both training data and validation are applied augmentation techniques, namely rotation and horizontal flip so that the dataset used in the modeling process can become more numerous. After the augmentation, the number of data for training, validation, and testing respectively become 300, 33, and 28 data. Then, images data were applied a stacking and ground truth were applied an one hot encoding. Model training was carried out by using SegNet model with hyperparameter models were batch size of 32, learning rate of 0.0001, and epoch of 300. The model also applied an optimization function, named Adam (Adaptive moment estimation) and also applied loss function called categorical cross entropy. The modelling was done by 10 times trial and the training process succeeded reach the average performance value of 99,31% and 92,01% in training and validation accuracy, reach the average performace value of 27,45% and 44,33 % in loss training and loss validation. Meanwhile the testing process succeeded reach the average performace value of 92,99% in testing accuracy, 0,4265 in testing los, and Mean-IoU of 70,03%."

"Latar Belakang: Terdapat beberapa kriteria yang diajukan sebagai standarisasi evaluasi material restorasi atau teknik operatif secara klinis yaitu US Public Health Service (USPHS) dan FDI World Dental Federation (FDI). Kriteria tersebut membantu dokter gigi memberi keputusan dalam menentukan keberhasilan suatu perawatan terutama restorasi. Adapun metode fotografi dapat menjadi metode indirek dalam menilai suatu restorasi dan sudah digunakan dalam praktik sehari-hari. Dengan media foto digital dokter gigi dapat mengidentifikasi perubahan awal pada kerusakan restorasi yang tidak terlihat jelas secara klinis. Tujuan: Untuk membandingkan kriteria USPHS-modifikasi dan FDI dalam menilai restorasi GIC gigi sulung pada media foto digital. Metode Penelitian: Penelitian analitik komparatif dilakukan di Klinik Gigi Anak RSKGM FKG UI pada 40 restorasi GIC di gigi geraham pertama bawah sulung anak usia 4-9 tahun. Gigi dibersihkan dan dilakukan evaluasi klinis menggunakan kriteria USPHS-modifikasi dan FDI sebelum pengambilan foto. Data foto yang terkumpul kemudian dievaluasi menggunakan kedua kriteria yang sama. Data hasil penilaian klinis dan foto digital diuji secara statistik. Hasil: Perbandingan penilaian evaluasi restorasi GIC gigi sulung secara klinis dan foto digital menggunakan kriteria evaluasi USPHS-modifikasi menunjukkan hasil yang berbeda bermakna dan signifikan secara statistik. Sedangkan dengan kriteria FDI menunjukkan hasil yang berbeda namun tidak bermakna dan signifikan secara statistik. Kesimpulan: Terdapat perbedaan antara kriteria USPHS-modifikasi dan FDI dalam menilai restorasi GIC gigi sulung dilihat pada media foto digital. Penilaian evaluasi dengan kriteria FDI memberikan hasil yang lebih konsisten dengan penilaian klinisnya dibandingkan kriteria USPHS-modifikasi.

---

Background: There are several criterias proposed as standardization for clinical evaluation of restoration materials or operative techniques, namely the US Public Health Service (USPHS) and FDI World Dental Federation (FDI). These criterias assist dentists in making decision to determine the success of a treatment especially restoration. The photographic method can be an indirect method in assessing a restoration and has been used in daily practice. With digital photo media, dentists can identify early changes in damaged restorations that are not clinically obvious. Objective: To compare modified USPHS and FDI criteria in assessing GIC restoration of primary teeth on digital image. Methods: A comparative analytic study was conducted at the Pedodontic Clinic, RSKGM FKG UI on 40 GIC restorations in lower primary first molars of children aged 4-9 years. The teeth were cleaned and evaluated using modified USPHS and FDI criteria prior to taking image. The collected photo datas are also evaluated using same criterias. Datas from clinical assessment and digital photo were tested statistically. Results: Comparison of clinical and digital image for GIC assessment on primary teeth using modified USPHS criteria showed statistically significant different result. Whereas FDI assessment criteria showed different result but not statistically significant. Conclusion: There is differences between modified USPHS and FDI criteria in assessing GIC restorations for primary teeth on digital image. Assessment using FDI criteria gave results that more consistent with the clinical assessment than modified USPHS criteria."

"Pterygium is an epithelial conjunctiva bulbi and connective tissue growth that could cause viston problem. Pterygium is mainly found at tropical and subtropical areas. There is no accurate data about pterygium prevalence in Indonesia.Those analyzed were respondents aged 5 years and more from Basic Health Research (RISKESAS) 2010, a cross sectional non intervention study. Diagnosis was made using flashlight and compared it to a chart. Results: The prevalence of pterygium at both eyes was 3.2% and at one eye was 1.9%. The highest prevalence of pterygium atboth eyes was at West Sumatra province (9.4%), the lowest prevalence was at Jakarta province (0. 4%). The highest prevalence of pterygium at one eye was at West Nusa Tenggara province, the lowest was at Jakarta province (0. 2%). The lowest prevalence of pterygium at both eyes as well as at one eye was at those aged 5-9 years (0. 03%) while the highest prevalence were found at age 70 years and more. The prevalence of pterygium at both eyes and the prevalence of pterygium at one eye based on gender were almost similar, the prevalence of pterygium among farmers was the highest (6.1%)and the lowest were among school children (1.0%); the highest prevalence were those with no schooling (11.0%) and the lowest were those that finished Junior High School (1.6%); the highest was at rural area for both eye (3.7%) as well as for one eye (2.2%) as compared to urban area. The prevalence of pterygium of both eyes was highest at lowest household expenditure (3.2%) while the lowest for one eye pterygium ( 1. 7%) at highest household's expenditure. Pterygium is a community health problem at rural areas especially among farmers and sailors that were used for high sunlight exposure. This type eye problem increased among those who lived in the equator. "