Genotype and Serotype Analysis on Potential Hepatitis B Virus as a Candidate Sequence for Hepatitis B Vaccine in Web-Based Bioinformatics

( Rifaldy Fajar ),( Dewi Mustika Sari ),( Nana Indri Kurniastuti ),( Prihantini Jupri ) 대한간학회 2020 춘·추계 학술대회 (KASL) Vol.2020 No.1

Aims: Currently there are 10 types of hepatitis B virus genotypes, from A to J, and 4 types of serotypes namely adw, adr, ayw, and ayr. This study aims to determine the genotype and serotype of the hepatitis B virus that has potential as a hepatitis B vaccine candidate. One effective method at present is bioinformatics, a multidisciplinary web-based biological science that can explore various sequences and see phylogeny. Methods: The first stage is the collection and selection of nucleotide DNA sequences or hepatitis B virus amino acids. All data on nucleotide DNA sequences and hepatitis B virus amino acids with the target genotype and serotype are accessed and collected from Genbank. Next, a kinship tree is made. This kinship tree is designed with multiple alignments, phylogeny, and tree viewers using phylogeny.fr. Results: The data obtained shows that there are 43 sequences with the same subtype, Adw, but the genotype and distribution of the spread of the hepatitis B virus are different. Genotype A originates from Somalia (Africa), and the Philippines (Asia), genotype B originates from Indonesia and China. Genotype C explains that genotype C is found around South Asia and East Asia, genotype H obtained information from America and Mexico, and genotype I originates from China. Conclusions: Sequence data that can be candidates for hepatitis B vaccine design are hepatitis B virus genotype B with subgenotype B3, genotype C with subgenotype C6 for the scope of Indonesia, while for the scope of the world obtained the potential of the Adw serotype.

Diagnosis of Liver Disease Based on Artificial Intelligence (AI) Systems Using the Decision Tree Model Algorithm Implementation

( Rifaldy Fajar ),( Nana Indri Kurniastuti ),( Dewi Mustika Sari ) 대한간학회 2020 춘·추계 학술대회 (KASL) Vol.2020 No.1

Aims: The liver is one of the most important organs of the human body. The function of the liver is to detoxify poisons in the human body and control cholesterol and fat in the human body. If the liver is damaged, then health will be disrupted and even death can occur. So an effort is needed to detect liver disease early. This study will discuss the classification of liver disease using the Decision Tree C4.5 Algorithm using the Indonesian Liver Patient Dataset. This study will also prove the most influential variable of the 11 variables that determine the liver disease. Methods: The research conducted includes processing the dataset using the help of the Rapidminer (computer program) data mining application. The dataset used in this study was taken from The Ministry of Health of the Republic of Indonesia Database. Indonesian Liver Patient Dataset contains 583 clinical data with 10 attributes with 416 positive liver output targets and 167 negatives. Based on 583 processed data, 433 data are used as training data and 150 data are used as testing data. Results: This study shows that only two variables (Almine Alminotransferase and Age) among the 11 variables in the dataset are the most influential in determining the classification of liver disease. This study also showed an accuracy of 72.7% in determining the classification of liver disease using the Dataset of the Ministry of Health of the Republic of Indonesia. Conclusions: Based on the results of this study, we can conclude that the detection or classification results can be said to be quite good based on an accuracy value of more than 70%.

Classification of Pancreatic Cancer Stadium Using Recurrent Neural Network (RNN) Model Algorithm

( Rifaldy Fajar ),( Nana Indri Kurniastuti ),( Dewi Mustika Sari ),( Prihantini Jupri ) 대한간학회 2020 춘·추계 학술대회 (KASL) Vol.2020 No.1

Aims: One way to detect the presence of pancreatic cancer is by examining it using Computed tomography (CT) scan. After a pancreatic cancer is detected, classification is done to determine the stage of cancer. In this study, we used the RNN model for the classification of Pancreatic cancer stadium. This study aimed to explain the procedure and the accuracy of the Elman tissue RNN modeling in pancreatic cancer stadium classification from the CT scan. Methods: The process carried out is to convert the image of red green blue (rgb) to a grayscale image on the CT scan data. After that the image was extracted with Gray Level Co-occurrence Matrix which was designed using Graphical User Interface with Matlab. There are 14 features, namely energy, contrast, correlation, Sum of Square, Inverse Different Moment, sum average, sum variance, sum entropy, entropy, differential variance, differential entropy, maximum probability, homogeneity, and dissimilarity. The feature is used as input, which is then divided into training data and testing data. After that, Elman network RNN modeling was carried out with data normalization, best model design, and data denormalization. The best model design was done by finding the number of hidden neurons and eliminating network inputs using the backpropagation algorithm. Results: The results of the best model training data and testing data were measured using sensitivity, specificity, and accuracy. So that from 74 training data obtained 92% accuracy rate, 96% sensitivity level as a reliable indicator when the results show pancreatic cancer, and 79% level of specificity as a good indicator when the results show normal pancreatic. While in 18 data testing showed 94% accuracy, 100% level of sensitivity, and 80% level of specificity. Conclusions: The conclusion in this study can be said that good classification results are obtained.

Lung Cancer Classification Based on Recurrent Neural Network and Recurrent Neuro-Fuzzy Model Algorithms

( Rifaldy Fajar ),( Nana Indri Kurniastuti ),( Dewi Mustika Sari ),( Prihantini Jupri ) 대한결핵 및 호흡기학회 2020 대한결핵 및 호흡기학회 추계학술대회 초록집 Vol.128 No.0

This study aims to explain the procedure, application, and accuracy of Recurrent Neural Network modeling and Recurrent Neuro-Fuzzy modeling for lung cancer nodule classification from lung photo images. Recurrent Neural Network and Recurrent Neuro-Fuzzy modeling steps are defining input and target variables, dividing data into training data and testing data, data normalization, designing the best model, and data denormalization. The input variable used is the feature of the lung photo image extraction, while the target tissue is the description of the condition from the lung photo image, namely normal lung, benign lung tumor, or malignant lung tumor. The image extraction step begins with an image transformation, namely from the original lung image (gray image) to a binary image, followed by extracting the transformed image using the Gray Level Co-occurrence Matrix Method. The design steps for the best Recurrent Neuro-Fuzzy model begin with the design steps for the best Recurrent Neural Network model followed by data clustering steps using the Fuzzy C-Means Method, learning Recurrent Neural Networks related to antecedents to fuzzy inference rules and consequent fuzzy inference rules, and Simplifying the consequent part by eliminating input and finding the consequent coefficient value of each cluster using the Least Square Estimator (LSE) Method. The Results obtained indicate that the classification of lung cancer nodules using the Recurrent Neural Network model gives better Results than the Recurrent Neuro-Fuzzy model. The sensitivity, specificity, and accuracy values of the Recurrent Neural Network model were 94%, 56%, and 81.33% for training data and 80%, 40%, and 64% for testing data, respectively. The Conclusion in this study is that the Recurrent Neural Network model is able to classify quite well compared to the Recurrent Neuro-Fuzzy model.