Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988514
A. A. Iskandar, Elnora Listianto Lie, K. A. Audah, Rose Khasana Dewi
The diagnostic method for detecting cervical cancer using Pap smear can be laborious and time-consuming. Therefore, research on computer-aided diagnosis is essential. The purpose of this study is to aid the distinguishing of Pap smear images from various categories of cervical cells by creating an alternative image processing and classification method. This is so that in the future, the burden on pathologists to manually analyze many Pap smear images can be reduced. The developed method will be able to help in the detection of abnormality or cancer. The processing methods include Gaussian filtering, Otsu thresholding, Canny edge detection, and Convolutional Neural Network. The analytical methods utilized were accuracy and loss curves, and the evaluation measures of accuracy, precision, recall, and F1 measure. The most optimal trained model had an accuracy, precision, recall, and F1 measure of 93.26%, 92.55%, 91.52%, and 91.84% respectively. It was concluded that the image processing and classification method could be used to distinguish multi-cell Pap smear images. Even with some limitations, it has the potential to improve single-cell analysis and also aid in classification. In the future, this method may be used in the medical field to help diagnose cervical cancer in Indonesia.
{"title":"Cervical Cancer Image Processing with Convolutional Neural Network for Detection","authors":"A. A. Iskandar, Elnora Listianto Lie, K. A. Audah, Rose Khasana Dewi","doi":"10.1109/IBIOMED56408.2022.9988514","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988514","url":null,"abstract":"The diagnostic method for detecting cervical cancer using Pap smear can be laborious and time-consuming. Therefore, research on computer-aided diagnosis is essential. The purpose of this study is to aid the distinguishing of Pap smear images from various categories of cervical cells by creating an alternative image processing and classification method. This is so that in the future, the burden on pathologists to manually analyze many Pap smear images can be reduced. The developed method will be able to help in the detection of abnormality or cancer. The processing methods include Gaussian filtering, Otsu thresholding, Canny edge detection, and Convolutional Neural Network. The analytical methods utilized were accuracy and loss curves, and the evaluation measures of accuracy, precision, recall, and F1 measure. The most optimal trained model had an accuracy, precision, recall, and F1 measure of 93.26%, 92.55%, 91.52%, and 91.84% respectively. It was concluded that the image processing and classification method could be used to distinguish multi-cell Pap smear images. Even with some limitations, it has the potential to improve single-cell analysis and also aid in classification. In the future, this method may be used in the medical field to help diagnose cervical cancer in Indonesia.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122350711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988366
Muhammad Arfan, H. Zakaria
Visualization has always aided clinical trial diagnoses. The majority of observations are, unfortunately, performed manually. Repeatability, samples, and effort are necessary for quantitative research. More samples complicate the process. A density study of type III collagen deposition was manually performed on 105 samples using ImageJ on photomicrographs by adjusting the deposition color in a binary image. Manually examining photomicrographs for collagen fiber density is time-consuming and tiring. This study automatically quantifies the type III collagen deposition density using CellProfiler, which does not require skill in observing large samples and complex research obj ects, thus enabling a less time-consuming technique. This study equalizes illumination and reduces photomicrograph noise to help identify cells. The line and tubeness features are improved to enhance the pixel intensity and collagen fiber structure. CellProfiler processed 105 photos in eight minutes, 57 seconds, or 5,1 seconds each. ImageJ required 114 seconds per photomicrograph or 129,5 minutes total (depending on the accuracy of the researchers). CellProfiler accelerated image processing by 14,5 times. Comparing the calculations of CellProfiler and ImageJ using linear regression yielded R2 = 0,7786, indicating a strong relationship. In addition, it produced the equation y = 0.9548x + 1.2197, indicating a positive correlation. This strong relationship and positive correlation suggested that CellProfiler's automatic quantification could assist researchers in measuring complex cells like collagen fiber structure in a less time-consuming technique.
{"title":"Quantification of Type III Collagen Deposition Density from Photomicrograph of Vaginal Connective Tissue","authors":"Muhammad Arfan, H. Zakaria","doi":"10.1109/IBIOMED56408.2022.9988366","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988366","url":null,"abstract":"Visualization has always aided clinical trial diagnoses. The majority of observations are, unfortunately, performed manually. Repeatability, samples, and effort are necessary for quantitative research. More samples complicate the process. A density study of type III collagen deposition was manually performed on 105 samples using ImageJ on photomicrographs by adjusting the deposition color in a binary image. Manually examining photomicrographs for collagen fiber density is time-consuming and tiring. This study automatically quantifies the type III collagen deposition density using CellProfiler, which does not require skill in observing large samples and complex research obj ects, thus enabling a less time-consuming technique. This study equalizes illumination and reduces photomicrograph noise to help identify cells. The line and tubeness features are improved to enhance the pixel intensity and collagen fiber structure. CellProfiler processed 105 photos in eight minutes, 57 seconds, or 5,1 seconds each. ImageJ required 114 seconds per photomicrograph or 129,5 minutes total (depending on the accuracy of the researchers). CellProfiler accelerated image processing by 14,5 times. Comparing the calculations of CellProfiler and ImageJ using linear regression yielded R2 = 0,7786, indicating a strong relationship. In addition, it produced the equation y = 0.9548x + 1.2197, indicating a positive correlation. This strong relationship and positive correlation suggested that CellProfiler's automatic quantification could assist researchers in measuring complex cells like collagen fiber structure in a less time-consuming technique.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132155078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988432
K. Al-Naime, A. Al-Anbuky, G. Mawston
Abdominal cancer is one of the most frequent and dangerous cancers in the world, particularly among the elderly. Major surgery is associated with a significant deterioration in quality of life. Physical fitness and level of activity are considered important factors for patients with cancer undergoing major abdominal surgery. One of the main programmes designed to improve the patient's fitness before major surgery is physical exercises (prehabilitation). However, significant numbers of patients undergoing major surgery cannot implement the programs due to limited health service resources, or patients are living in remote locations. This paper discusses a novel IoT concept for precision prehabilitation program monitoring. The solution integrates the traditional 6-week program's follow-up mechanism with the IoT system. This in turn tracks the patient's movement activities anytime anywhere and records the significant movements specific to the program. As a result, both the patients and the health system are relieved from the restricted capacity and associated cost. A wearable sensor was placed on the participant's ankle, and a gateway and the ThingSpeak platform were developed to perform IoT remote monitoring techniques. The key outcome is the visibility of IoT system to support mixed mode prehabilitation program by reducing the barriers and obstacles of existing prehabilitation programs.
{"title":"IoT Based Pre-Operative Prehabilitation Program Monitoring Model: Implementation and Preliminary Evaluation","authors":"K. Al-Naime, A. Al-Anbuky, G. Mawston","doi":"10.1109/IBIOMED56408.2022.9988432","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988432","url":null,"abstract":"Abdominal cancer is one of the most frequent and dangerous cancers in the world, particularly among the elderly. Major surgery is associated with a significant deterioration in quality of life. Physical fitness and level of activity are considered important factors for patients with cancer undergoing major abdominal surgery. One of the main programmes designed to improve the patient's fitness before major surgery is physical exercises (prehabilitation). However, significant numbers of patients undergoing major surgery cannot implement the programs due to limited health service resources, or patients are living in remote locations. This paper discusses a novel IoT concept for precision prehabilitation program monitoring. The solution integrates the traditional 6-week program's follow-up mechanism with the IoT system. This in turn tracks the patient's movement activities anytime anywhere and records the significant movements specific to the program. As a result, both the patients and the health system are relieved from the restricted capacity and associated cost. A wearable sensor was placed on the participant's ankle, and a gateway and the ThingSpeak platform were developed to perform IoT remote monitoring techniques. The key outcome is the visibility of IoT system to support mixed mode prehabilitation program by reducing the barriers and obstacles of existing prehabilitation programs.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114386661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988631
Mir Ashib Ullah, Fazlul Rafeeun Khorshed, Md. Ruhul Amin
In this work, a simple electrode arrangement is proposed for a microfluidic channel that utilizes dielectrophoresis and fluid dynamics to separate circulating tumor cells (CTCs) from blood cells that can be used effectively in microfluidic channels. Dielectrophoresis mechanism aided the microfluidic channel that has been made considering the Clausius-Mossotti (CM) factor, electrical and other mechanical properties of white blood cell (WBC), red blood cell (RBC) and CTC particles to accumulate the rare CTCs being isolated from WBCs and RBCs to a specified outlet. A comparative analysis of the microfluidic channel for various ranges of inlet velocity and applied electric fields by computer-assisted multi-physics simulations using the Finite Element Method (FEM) with various governing parameters using COMSOL, MATLAB, and MyDEP software has been done in this study to validate the performance of the proposed microfluidic channel which showed 100% separation efficiency (SE) and separation purity (SP) for 4V peak-to-peak applied voltage on the electrodes. Analysis of the inputs and outputs from the simulation model has been done to suggest specific values of inputs for the most efficient separation of the microfluidic channel through Adaptive Neuro-Fuzzy Inference System (ANFIS).
{"title":"A Microfluidic Channel for Separation of Circulating Tumor Cells from Blood Cells Using Dielectrophoresis and Its Performance Analysis Using Adaptive Neuro-Fuzzy Inference System","authors":"Mir Ashib Ullah, Fazlul Rafeeun Khorshed, Md. Ruhul Amin","doi":"10.1109/IBIOMED56408.2022.9988631","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988631","url":null,"abstract":"In this work, a simple electrode arrangement is proposed for a microfluidic channel that utilizes dielectrophoresis and fluid dynamics to separate circulating tumor cells (CTCs) from blood cells that can be used effectively in microfluidic channels. Dielectrophoresis mechanism aided the microfluidic channel that has been made considering the Clausius-Mossotti (CM) factor, electrical and other mechanical properties of white blood cell (WBC), red blood cell (RBC) and CTC particles to accumulate the rare CTCs being isolated from WBCs and RBCs to a specified outlet. A comparative analysis of the microfluidic channel for various ranges of inlet velocity and applied electric fields by computer-assisted multi-physics simulations using the Finite Element Method (FEM) with various governing parameters using COMSOL, MATLAB, and MyDEP software has been done in this study to validate the performance of the proposed microfluidic channel which showed 100% separation efficiency (SE) and separation purity (SP) for 4V peak-to-peak applied voltage on the electrodes. Analysis of the inputs and outputs from the simulation model has been done to suggest specific values of inputs for the most efficient separation of the microfluidic channel through Adaptive Neuro-Fuzzy Inference System (ANFIS).","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127584405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988690
Aiman Muhamad Basymeleh, Bagus Esa Pramudya, Reinato Teguh Santoso
Leukemia is diagnosed by observing two indicators, bone marrow smear and peripheral blood smear with laboratory skills using a microscope for diagnosing cancer. All diagnostics tests require advanced laboratory tests and another limitations like time and pricing. With all limitations, this study compares deep learning architectures from image augmentation from HSV images for diagnosis and classification for four label outputs using Adam optimizer. As a result of this study, VGG16 achieved better evaluation results than another architecture which attained an accuracy, sensitivity, specificity, and validation accuracy of 97.50%, 99.96%, 100%, and 98.44%, respectively. For its development in real cases, the modeling can be applied directly to the relevant in the future or using a new novel method architecture.
{"title":"Acute Lymphoblastic Leukemia Image Classification Performance with Transfer Learning Using CNN Architecture","authors":"Aiman Muhamad Basymeleh, Bagus Esa Pramudya, Reinato Teguh Santoso","doi":"10.1109/IBIOMED56408.2022.9988690","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988690","url":null,"abstract":"Leukemia is diagnosed by observing two indicators, bone marrow smear and peripheral blood smear with laboratory skills using a microscope for diagnosing cancer. All diagnostics tests require advanced laboratory tests and another limitations like time and pricing. With all limitations, this study compares deep learning architectures from image augmentation from HSV images for diagnosis and classification for four label outputs using Adam optimizer. As a result of this study, VGG16 achieved better evaluation results than another architecture which attained an accuracy, sensitivity, specificity, and validation accuracy of 97.50%, 99.96%, 100%, and 98.44%, respectively. For its development in real cases, the modeling can be applied directly to the relevant in the future or using a new novel method architecture.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134303504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988223
N. N. Qomariyah, Tianda Sun, D. Kazakov
The presence of COVID-19, a respiratory disease, can be detected through medical imaging, such as Chest X-Ray (CXR) and Computed Tomography (CT) scans. These radiology images can also show how the patient's condition progresses. Radiologists need to provide a written report for each image, so that other clinicians can use it in their decision making. In this study, we applied one of the Natural Language Processing (NLP) models called IndoBERT to analyze radiology reports of COVID-19 patients written in Indonesian. We performed two tasks, clustering to group reports by meaning and understand their content, and text classification to predict one of the five possible outcomes for each patient. We show the most frequent topics in radiology reports, and word scores in each topic. The IndoBERT model was fine tuned on a medical text, ‘Kamus Kedokteran Dorland’ in an attempt to further improve it. This proved unnecessary: on one hand, there were no additional benefits, on the other, the standard model alone achieved a very satisfactory classification accuracy of over 90 %.
{"title":"NLP Analysis of COVID-19 Radiology Reports in Indonesian using IndoBERT","authors":"N. N. Qomariyah, Tianda Sun, D. Kazakov","doi":"10.1109/IBIOMED56408.2022.9988223","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988223","url":null,"abstract":"The presence of COVID-19, a respiratory disease, can be detected through medical imaging, such as Chest X-Ray (CXR) and Computed Tomography (CT) scans. These radiology images can also show how the patient's condition progresses. Radiologists need to provide a written report for each image, so that other clinicians can use it in their decision making. In this study, we applied one of the Natural Language Processing (NLP) models called IndoBERT to analyze radiology reports of COVID-19 patients written in Indonesian. We performed two tasks, clustering to group reports by meaning and understand their content, and text classification to predict one of the five possible outcomes for each patient. We show the most frequent topics in radiology reports, and word scores in each topic. The IndoBERT model was fine tuned on a medical text, ‘Kamus Kedokteran Dorland’ in an attempt to further improve it. This proved unnecessary: on one hand, there were no additional benefits, on the other, the standard model alone achieved a very satisfactory classification accuracy of over 90 %.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116974889","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988446
A. A. Iskandar, M. Jeremy, M. Fathony
Breast Cancer is one of the most common types of cancer. This research was conducted with the purpose of developing a Computer-Aided Diagnosis to detect breast cancers from mammogram images. The mammogram images were obtained from the INbreast Dataset and Husada Hospital in Jakarta. The program was developed with the usage of pre-processing which includes Median Filtering, Otsu thresholding, Truncation Normalization, and Contrast Limited Adaptive Histogram Equalization to manipulate the images and Convolutional Neural Network to classify the images into either mass or normal, or either benign or malignant. The pre-processing pipeline have provided enhanced images to be used to train and test the Convolutional Neural Network. The best model achieved reached an accuracy, precision and sensitivity of 94.1%, 100% and 85.7% in classifying the mammogram images into benign or malignant, and 88.3%, 92.6% and 83.3% in classifying the mammogram images into mass or normal. In conclusion, the algorithm was able to classify mammogram images and has provided results as high as other related researches.
{"title":"Breast Cancer Image Pre-Processing With Convolutional Neural Network For Detection and Classification","authors":"A. A. Iskandar, M. Jeremy, M. Fathony","doi":"10.1109/IBIOMED56408.2022.9988446","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988446","url":null,"abstract":"Breast Cancer is one of the most common types of cancer. This research was conducted with the purpose of developing a Computer-Aided Diagnosis to detect breast cancers from mammogram images. The mammogram images were obtained from the INbreast Dataset and Husada Hospital in Jakarta. The program was developed with the usage of pre-processing which includes Median Filtering, Otsu thresholding, Truncation Normalization, and Contrast Limited Adaptive Histogram Equalization to manipulate the images and Convolutional Neural Network to classify the images into either mass or normal, or either benign or malignant. The pre-processing pipeline have provided enhanced images to be used to train and test the Convolutional Neural Network. The best model achieved reached an accuracy, precision and sensitivity of 94.1%, 100% and 85.7% in classifying the mammogram images into benign or malignant, and 88.3%, 92.6% and 83.3% in classifying the mammogram images into mass or normal. In conclusion, the algorithm was able to classify mammogram images and has provided results as high as other related researches.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122112112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9988361
D. N. K. Hardani, H. A. Nugroho, I. Ardiyanto
Breast cancer can threaten women's health and become a cause of death. Reducing mortality from breast cancer necessitates early recognition of its signs and symptoms. An essential step in building an early detection system is to segment the breast ultrasound image (BUS). The accuracy of segmentation has a direct bearing on the effectiveness of quantitative analysis and the detection of breast tumor. However, this image segmentation becomes constrained because the BUS image has a shallow quality. Therefore, it is necessary to take preprocessing steps to improve the image. This study aims to compare the efficiency of various filtering techniques for BUS segmentation with the visual attention model. There are 12 filters tested in this study, including Mean, Median, Bilateral, Fast nonlinear, Lee, Lee-enhance, Frost, Kuan, Gamma, Wiener, Speckle Reduction Anisotropic Diffusion Filter (SRAD), and Detail Preserved Anisotropic Diffusion Filter (DPAD). The segmentation process uses a Convolutional Neural Network (CNN) based network architecture, namely Visual Geometry Group architecture with 16 layers (VGG-16). The segmentation results were analyzed using three visual attention models. The results showed that the image before filtering and after filtering showed visually significant results.
{"title":"The Impact of Filtering for Breast Ultrasound Segmentation using A Visual Attention Model","authors":"D. N. K. Hardani, H. A. Nugroho, I. Ardiyanto","doi":"10.1109/IBIOMED56408.2022.9988361","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9988361","url":null,"abstract":"Breast cancer can threaten women's health and become a cause of death. Reducing mortality from breast cancer necessitates early recognition of its signs and symptoms. An essential step in building an early detection system is to segment the breast ultrasound image (BUS). The accuracy of segmentation has a direct bearing on the effectiveness of quantitative analysis and the detection of breast tumor. However, this image segmentation becomes constrained because the BUS image has a shallow quality. Therefore, it is necessary to take preprocessing steps to improve the image. This study aims to compare the efficiency of various filtering techniques for BUS segmentation with the visual attention model. There are 12 filters tested in this study, including Mean, Median, Bilateral, Fast nonlinear, Lee, Lee-enhance, Frost, Kuan, Gamma, Wiener, Speckle Reduction Anisotropic Diffusion Filter (SRAD), and Detail Preserved Anisotropic Diffusion Filter (DPAD). The segmentation process uses a Convolutional Neural Network (CNN) based network architecture, namely Visual Geometry Group architecture with 16 layers (VGG-16). The segmentation results were analyzed using three visual attention models. The results showed that the image before filtering and after filtering showed visually significant results.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129699717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9987756
I. G. Pande Darma Suardika, I. M. Dendi Maysanjaya, Made Windu Antara Kesiman
An optic disc is an object on the retina of the eye that has the characteristics of being brightly colored and round. Optical disc segmentation is the most commonstep taken before processing a retinal fundus image. The bright characteristics of the optic disc often interfere withthe detection of other objects in the retinal fundus image. Therefore, the optic disc is the first step before processingthe fundus image of the retina. With the help of digital image processing will help in the removal of the optic discon the fundus image of the retina. Many methods can be used in optical disc segmentation, one of which is the deep learning method. The deep learning method chosen is Mask R-CNN to produce a mask from the results of object detection on the retinal fundus image. There are 3 stages in the segmentation process using the Mask R-CNN. First, the data used in the training process will be labeled. thereis 1 label given, namely optic disc. Then the model is trained using the restnet50 backbone architecture and finally, the model will be evaluated. To evaluate the results obtained from the two methods, it uses Intersection over Union (IoU) by comparing directly the results of prediction and ground truth. The data used is an IDRiD dataset containing retinal fundus images taken from eye clinics across India. As the result, Mask R-CNN can segment the optical disc with an IoU value of 0.843. it is hoped that the results of this research can help the process in processing retinal fundus images in the future.
视盘是眼睛视网膜上的一个物体，具有明亮的颜色和圆形的特征。在处理视网膜眼底图像之前，光盘分割是最常见的步骤。视盘的明亮特性常常干扰眼底图像中其他物体的检测。因此，视盘是处理视网膜眼底图像之前的第一步。借助数字图像处理将有助于去除视差的眼底图像的视网膜。用于光盘分割的方法有很多，深度学习方法是其中的一种。我们选择的深度学习方法是Mask R-CNN，从视网膜眼底图像上的目标检测结果产生一个Mask。使用掩码R-CNN的分割过程有3个阶段。首先，训练过程中使用的数据将被标记。给出了1个标签，即视盘。然后使用restnet50主干架构对模型进行训练，最后对模型进行评估。为了评价两种方法得到的结果，通过直接比较预测结果和地面真值，使用了交集优于联合(Intersection over Union, IoU)。使用的数据是一个IDRiD数据集，其中包含从印度各地眼科诊所拍摄的视网膜眼底图像。因此，Mask R-CNN可以分割光盘，IoU值为0.843。希望本研究结果能对未来视网膜眼底图像的处理过程有所帮助。
{"title":"Optic Disc Segmentation Based on Mask R-CNN in Retinal Fundus Images","authors":"I. G. Pande Darma Suardika, I. M. Dendi Maysanjaya, Made Windu Antara Kesiman","doi":"10.1109/IBIOMED56408.2022.9987756","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9987756","url":null,"abstract":"An optic disc is an object on the retina of the eye that has the characteristics of being brightly colored and round. Optical disc segmentation is the most commonstep taken before processing a retinal fundus image. The bright characteristics of the optic disc often interfere withthe detection of other objects in the retinal fundus image. Therefore, the optic disc is the first step before processingthe fundus image of the retina. With the help of digital image processing will help in the removal of the optic discon the fundus image of the retina. Many methods can be used in optical disc segmentation, one of which is the deep learning method. The deep learning method chosen is Mask R-CNN to produce a mask from the results of object detection on the retinal fundus image. There are 3 stages in the segmentation process using the Mask R-CNN. First, the data used in the training process will be labeled. thereis 1 label given, namely optic disc. Then the model is trained using the restnet50 backbone architecture and finally, the model will be evaluated. To evaluate the results obtained from the two methods, it uses Intersection over Union (IoU) by comparing directly the results of prediction and ground truth. The data used is an IDRiD dataset containing retinal fundus images taken from eye clinics across India. As the result, Mask R-CNN can segment the optical disc with an IoU value of 0.843. it is hoped that the results of this research can help the process in processing retinal fundus images in the future.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"17 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132835716","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-18DOI: 10.1109/IBIOMED56408.2022.9987817
Zahra'ul Athiyah, A. E. Permanasari, S. Wibirama
Cadaver as a learning media for manual anatomy is very important for students because it is believed to give a different impression from learning with other media. However, we still need learning support media considering the limitations of the use of corpses. It is necessary to have the role of digital media that students can use wherever they are without reducing the value or content of the material they usually learn manually. The author proposes a learning model with a mobile app-based gamification approach that can attract user's attention, increase learning motivation, and increase student interaction in the learning process. We build a Heart mobile application that is equipped with materials and quizzes. This paper presents the gamification design for the Heart application using a tetrad approach. The anatomy learning media uses 3D visualization and Augmented Reality, and the design of the learning flow uses the gamification method. The Heart program offers a simple and fun learning path. The System Usability Scale (SUS) results show a score of 72.25 and are included in the Good Usability category. Finally, gamification strategies are expected to improve users' efficacy and learning outcomes.
{"title":"Development of Gamification Design on Heart Anatomy Learning Media","authors":"Zahra'ul Athiyah, A. E. Permanasari, S. Wibirama","doi":"10.1109/IBIOMED56408.2022.9987817","DOIUrl":"https://doi.org/10.1109/IBIOMED56408.2022.9987817","url":null,"abstract":"Cadaver as a learning media for manual anatomy is very important for students because it is believed to give a different impression from learning with other media. However, we still need learning support media considering the limitations of the use of corpses. It is necessary to have the role of digital media that students can use wherever they are without reducing the value or content of the material they usually learn manually. The author proposes a learning model with a mobile app-based gamification approach that can attract user's attention, increase learning motivation, and increase student interaction in the learning process. We build a Heart mobile application that is equipped with materials and quizzes. This paper presents the gamification design for the Heart application using a tetrad approach. The anatomy learning media uses 3D visualization and Augmented Reality, and the design of the learning flow uses the gamification method. The Heart program offers a simple and fun learning path. The System Usability Scale (SUS) results show a score of 72.25 and are included in the Good Usability category. Finally, gamification strategies are expected to improve users' efficacy and learning outcomes.","PeriodicalId":250112,"journal":{"name":"2022 4th International Conference on Biomedical Engineering (IBIOMED)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130800265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：