Papers by tati rajab Mengko
2020 IEEE REGION 10 CONFERENCE (TENCON), 2020
Deep learning has been proposed as one of the automated solutions for diabetic retinopathy (DR) s... more Deep learning has been proposed as one of the automated solutions for diabetic retinopathy (DR) severity classification problem. However, most of the successful deep learning models are based on large convolutional neural network (CNN) architectures, requiring a vast volume of training data as well as dedicated computational resources. In this study, we used MobileNetV2 architecture, which was considered a small-scale architecture (4.2 million trainable parameters), to perform DR classification task in APTOS 2019 dataset (3662 color retinal images). We used the generic MobileNetV2 pre-trained weights from ImageNet as initialization and implemented data augmentation and resampling during training. We further optimized our model by combining it with an SVM classifier, resulting a hybrid and computationally efficient deep learning model, MobileNetV2-SVM. This model obtained quadratic weighted kappa score of 0.925, 85% accuracy, and area under receiving operating characteristic (AUROC) of 1.00, 0.82, 0.94, 0.94, 0.93 for normal, mild, moderate, severe, and proliferative DR classes, respectively; which is better or at least comparable with the larger architecture performance on the same dataset. Our result shows that with proper optimization strategy, a relatively small and generic CNN architecture, can achieve promising DR classification performance, and even outperform the performance of CNN model with larger architecture.
2020 IEEE REGION 10 CONFERENCE (TENCON), 2020
Determination of material properties is one of essential stages to more understanding the charact... more Determination of material properties is one of essential stages to more understanding the characteristics of biological sample especially in biomedical research. In this paper, a method of transmission phase-shift is proposed to determine complex permittivity of biological sample which is performed using a WR90 type X-band rectangular waveguide. Some samples of chicken meat, liver, and skin are applied as biological materials for the experimentation. The irregular shape of biological sample is examined by placing it into a thin container to obtain a flat surface and putting the container in inside of the rectangular waveguide. The complex relative permittivity of each sample is extracted from measured S-parameters and then determined using the method. The results show that the method could successfully determine the complex permittivity of biological sample. In addition, the water content in the material has become a critical issue to be considered in the examination especially for the biological sample with high permittivity.
2015 4th International Conference on Instrumentation, Communications, Information Technology, and Biomedical Engineering (ICICI-BME), 2015
Heart sound recording device prototype Detection of S1-S2 signals of heart sounds Analysis of... more Heart sound recording device prototype Detection of S1-S2 signals of heart sounds Analysis of heart sounds using discrete wavelet transform and teager energy operator Figure A. The block diagram of the heart sound recording and automatic S1-S2 waves detecting system Purpose: The second leading cause of death in the world is cardiovascular diseases. Diagnosis of vast majority of cardiovascular diseases is made by listening to heart sounds by specialists (auscultation method). However, since the method of auscultation depends on the experience and hearing ability of the specialist, obtained results can be subjective. Therefore, digitization and visualization of heart sounds enables accurate, rapid and economical diagnosis of cardiovascular diseases, especially heart valve diseases. The most effective parameter for the diagnosis of heart valve diseases is the location of the S1-S2 heart sounds. For this purpose, a device prototype that collects the heart sound from human body and records collected data on digital environment was designed also, a medical decision support system to detect the S1-S2 locations to assist physicians in their diagnosis was established. Theory and Methods: In the first part of the study a device prototype that collects heart sounds and records them on digital environment was developed using capacitive microphone, filter and amplifier circuits and stereo jack. In order to test the working accuracy of the designed device, clinical applications were carried out and obtained recordings were examined. In the second part of the study, in order to detect the location of S1-S2 heart sounds, obtained heart sounds are first filtered by discrete wavelet transform. Then, the S1-S2 waves in the filtered signal are made evident by the teager energy operator and rule-based algorithm. Results: As a result, it is decided that the developed prototype works functionally likewise with the use of the algorithm S1-S2 locations in normal and pathological data were detected with 98.67% sensitivity, 97.69% specificity and 98.18% accuracy. Conclusion: In conclusion, physicians can easily detect the waves of S1 and S2 with high accuracy using the developed prototoype and established algorithm. In addition, it is thought that this system can be used to diagnose heart valve diseases.
Usually, physicians diagnosing lung diseases by listening to the lung sound using stethoscope. Th... more Usually, physicians diagnosing lung diseases by listening to the lung sound using stethoscope. This technique is known as auscultation. Some lung diseases produce unique lung sounds, which is refers to special recognized pattern. But the main problems are the the frequency of lung sounds that are low (20 – 2000 Hz), low amplitude, interference from other sounds, ear sensitiveness, and low variety of the pattern of lung sounds that are almost similar. These factors remain to the false diagnosing of lung disease if the auscultation procedures aren't conducted correctly. Actually, false diagnoses can be minimized by a software that can classify lung sounds automatically. Input for the software are recorded lung sounds in *.wav, mono, and sampling frequency of 8000 Hz, with one respiration cycle. Lung sounds recognition is built base on wavelet analysis. Also, lung sounds are decomposed using wavelet packet up to 5 levels. The decomposition scenario includes breaking down lung sound...
International Journal of E-Health and Medical Communications, 2015
The automated diagnosis of cervical cancer in Pap smear images is a difficult though extremely im... more The automated diagnosis of cervical cancer in Pap smear images is a difficult though extremely important procedure. In order to obtain reliable diagnostic information, the nuclei and their characteristics must be correctly identified and evaluated. However, the presence of inflammatory and overlapping cells in these images complicates the detection process. In this work, a segmentation algorithm is developed to extract the inflammatory cells and enable accurate nuclei detection. The proposed algorithm is based on the combination of gray level thresholding and the definition of a distance rule, which entails in the identification of inflammatory cells. The results indicate that our method significantly simplifies the nuclei detection process, as it reduces the number of inflammatory cells that may interfere.
Machine Vision Applications, 2000
This paper proposes a method to improve the performance of Fractal lmage Compression (FIC) techni... more This paper proposes a method to improve the performance of Fractal lmage Compression (FIC) technique by combining FIC (lossy compression) and another lossless method (in this case Huffman coding is used). This proposed method takes advantage on each techniques, the high compression ratio that can be provided by FIC and the infinite peak-to-peak signal to-noise ratio (PSNR) that can be
International Conference on Instrumentation, Communication, Information Technology, and Biomedical Engineering 2009, 2009
We propose a new Hole-filling algorithm by improving the Olympic operator, and we also apply it t... more We propose a new Hole-filling algorithm by improving the Olympic operator, and we also apply it to generate the volume in our freehand 3D ultrasound reconstruction of the spine. First, the ultrasound frames and position information are compounded into a 3D volume using the Bin-filling method. Then, the Hole-filling method is used to repair gaps in the volume. The conventional Olympic operator defines the empty voxels by sorting the neighboring voxels, removing the n% of the upper and lower values, and averaging them to attain the value to fill the empty voxels. The empty voxel estimation can be improved by thresholding the range width of its neighboring voxels and adjusting it to the average values. The method is tested on a holemanipulated volume derived from a cropped 3D ultrasound volume of a part of the spine. Our MAE calculation on the proposed technique shows improved result compared to all tested existing methods.
International Journal of E-Health and Medical Communications, 2010
Hole-filling in ultrasound volume reconstruction using freehand three-dimensional ultrasound esti... more Hole-filling in ultrasound volume reconstruction using freehand three-dimensional ultrasound estimates the values for empty voxels from the unallocated voxels in the Bin-filling process due to inadequate sampling in the acquisition process. Olympic operator, as a neighbourhood averaging filter, can be used to estimate the empty voxel. However, this method needs improvement to generate a closer estimation of the empty voxels. In this paper, the authors propose an improved Olympic operator for the Hole-filling algorithm, and apply it to generate the volume in a 3D ultrasound reconstruction of the spine. The conventional Olympic operator defines the empty voxels by sorting the neighbouring voxels, removing the n% of the upper and lower values, and averaging them to attain the value to fill the empty voxels. The empty voxel estimation can be improved by thresholding the range width of its neighbouring voxels and adjusting it to the average values. The method is tested on a hole-manipula...
International Journal of E-Health and Medical Communications, 2010
Phase Unwrapping (PU) is reconstruction of absolute phase data from its wrapped phase. The absolu... more Phase Unwrapping (PU) is reconstruction of absolute phase data from its wrapped phase. The absolute phase cannot be extracted from the wrapped phase data directly. Without phase noise, singularity, and aliasing problems, the phase information can be unwrapped easily. However, the phase data are always contaminated by noise and discontinuities, making the PU process more complicated. Therefore, a suitable PU algorithm is required to address the problems properly. In this method, the energy difference between neighborhood pixels in level 3 is counted, followed by getting the probability value to obtain its total fringes. The capability of the proposed method to unwrap simulated and actual MRI phase images is also demonstrated. In actual MRI phase image, PU can be implemented for water and fat separation.
International Journal of E-Health and Medical Communications, 2013
The RGB color retinal image has an interesting characteristic, i.e. the G channel contains more i... more The RGB color retinal image has an interesting characteristic, i.e. the G channel contains more important information than the other ones. One of the most important features in a retinal image is the retinal blood vessel structure. Many diseases can be diagnosed based on in the retinal blood vessel, such as micro aneurysms that can lead to blindness. In the G channel, the contrast between retinal blood vessel and its background is significantly high. The authors explore this retinal image characteristic to construct a more suitable image coding system. The coding processes are conduct in three schemes: weighted R channel, weighted G channel, and weighted B channel coding. Their hypothesis is that allocating more bits in the G channel will improve the coding performance. The authors seek for image quality assessment (IQA) metrics that can be used to measure the distortion in retinal image coding. Three different metrics, namely Peak Signal to Noise Ratio (PSNR), Structure Similarity ...
International Journal of E-Health and Medical Communications, 2010
Accurate blood vessel segmentation plays a crucial role in non-invasive blood flow velocity measu... more Accurate blood vessel segmentation plays a crucial role in non-invasive blood flow velocity measurement based on complex-valued magnetic resonance images. We propose a specific snake active contour model-based blood vessel segmentation framework for complex-valued magnetic resonance images. The proposed framework combines both magnitude and phase information from a complex-valued image representation to obtain an optimum segmentation result. Magnitude information of the complex-valued image provides a structural localization of the target object, while phase information identifies the existence of flowing matters within the object. Snake active contour model, which models the segmentation procedure as a force-balancing physical system, is being adopted as a framework for this work due to its interactive, dynamic, and customizable characteristics. Two snake-based segmentation models are developed to produce a more accurate segmentation result, namely the Model-constrained Gradient Ve...
IEEE APCCAS 2000. 2000 IEEE Asia-Pacific Conference on Circuits and Systems. Electronic Communication Systems. (Cat. No.00EX394)
In this study, several image processing techniques have been studied for mineral identification. ... more In this study, several image processing techniques have been studied for mineral identification. The image data are obtained from a thin section under a polarizing microscope through a video camera with a frame grabber card. The data have been collected for isotropic (garnet), anisotropic non-pleochroic (quartz), and anisotropic pleochroic (biotit) minerals, during parallel- and cross-nicol observation. Using the image processing
Asia-Pacific Conference on Circuits and Systems
Characteristic of interference color in rock forming mineral image has been studied for the miner... more Characteristic of interference color in rock forming mineral image has been studied for the mineral identification. The interference color data are obtained from thin section images under polarizing microscope through video camera and a frame grabber card. The data have been collected for the isotropic (garnet), anisotropic non pleochroic (quartz), and anistropic pleochroic (biotite) minerals. The method of analysis is
MAKARA of Technology Series, 2010
Proses rekonstruksi data fasa dari bentuk tutupannya disebut Phase Unwrapping (PU). Secara ideal,... more Proses rekonstruksi data fasa dari bentuk tutupannya disebut Phase Unwrapping (PU). Secara ideal, tanpa adanya derau fasa, singularitas, dan masalah aliasing, informasi fasa dapat di-unwrap secara mudah. Namun kenyataannya, data fasa sebenarnya selalu mengalami gangguan derau dan diskontinuitas. Proses PU menjadi lebih rumit dan membutuhkan algoritma PU yang lebih sesuai untuk mengatasi masalah yang muncul. Untuk itu pada penelitian ini dikembangkan suatu algoritma PU lokal dengan menggunakan pendekatan minimisasi energi piksel-piksel yang bertetanggaan pada orde 1. Pada metode ini dihitung selisih energi dari empat piksel yang bertetanggaan, kemudian dihitung nilai probabilitas untuk mendapatkan jumlah lipatannya. Dari hasil pengujian menggunakan citra sintesis dan InSAR dengan koherensi 0,8 didapatkan nilai Peak Signal to Noise Ratio (PSNR) 30,5373 dB pada 20 iterasi.
OBJECTS TRACKING IN A VIDEO SEQUENCE. This paper presents the result of implementing a tracking s... more OBJECTS TRACKING IN A VIDEO SEQUENCE. This paper presents the result of implementing a tracking system for identifying objects in a video sequence. The main objective of this research is to keep track of objects movement and their activities which are then analyzed whether the activities related to suspicious activities or not. At this stage the research is concentrated on the keep track of the objects once the objects enter the scene. The objects tracking are done by identify objects' movement from video sequence using frame by frame analysis. In order to avoid tracking unnecessary objects a method is implemented to eliminate such objects. In this research a method to eliminate such objects is to use spatial objects information. Based on the described method the research shows that objects tracking in a video sequence can be implemented. Moreover, the research is also trying to isolate objects so that the object size and its activities can be analyzed. Finally, this research h...
Annals of Biomedical Engineering, 2010
Unreliable spinal X-ray radiography measurement due to standing postural variability can be minim... more Unreliable spinal X-ray radiography measurement due to standing postural variability can be minimized by using positional supports. In this study, we introduce a balancing device, named BalancAid, to position the patients in a reproducible position during spinal X-ray radiography. This study aimed to investigate the performance of healthy young subjects' standing posture on the BalancAid compared to standing on the ground mimicking the standard X-rays posture in producing a reproducible posture for the spinal X-ray radiography. A study on the posture reproducibility measurement was performed by taking photographs of 20 healthy young subjects with good balance control standing on the BalancAid and the ground repeatedly within two consecutive days. We analyzed nine posterior-anterior (PA) and three lateral (LA) angles between lines through body marks placed in the positions of T3, T7, T12, L4 of the spine to confirm any translocations and movements between the first and second day measurements. No body marks repositioning was performed to avoid any error. Lin's CCC test on all angles comparing both standing postures demonstrated that seven out of nine angles in PA view, and two out of three angles in LA view gave better reproducibility for standing on the BalancAid compared to standing on the ground. The PA angles concordance is on average better than that of the LA angles.
J. Biomed. Pharm. …, 2007
IEEE Access, 2022
The sequential labeling model is commonly used for time series or sequence data where each instan... more The sequential labeling model is commonly used for time series or sequence data where each instance label is classified using previous instance label. In this work, a sequential labeling model is proposed as a new approach to detect the type and index mutations simultaneously, using DNA sequences from lung cancer study cases. The methods used are One Dimensional Convolutional Neural Network (1D-CNN), Bidirectional Long Short-Term Memory (BiLSTM), and Bidirectional Gated Recurrent Unit (Bi-GRU). Each nucleotide in the patient's DNA sequence is classified as either normal or with a certain type of mutation in which case, its index mutation is predicted. The mutation types detected are either substitution, insertion, deletion, or delins (deletion insertion) mutations. Based on the experiments that were conducted using EGFR gene, BiLSTM and Bi-GRU displayed better performance and were more stable than 1D-CNN. Further tests were carried out on the TP53, KRAS, CTNNB1, SMARCA4, CDKN2A, PTPRD, BRAF, ERBB2, and PTPRT gene. The proposed model reports F1-scores of 0.9596, and 0.9612 using Bi-GRU and BiLSTM, respectively. Based on the results the model can successfully detect the type and index mutations in the DNA sequence more accurately and faster without the need for other supporting data and tools, and does not require realignment to reference sequences. This will greatly facilitate the user in detecting type and index mutations faster by entering only the DNA sequence.
2018 2nd International Conference on Biomedical Engineering (IBIOMED), 2018
Brachial-ankle pulse wave velocity (baPWV) is a vascular parameter used to assess the stiffness o... more Brachial-ankle pulse wave velocity (baPWV) is a vascular parameter used to assess the stiffness of arterial segments between the brachial and femoral arteries, including aorta. The baPWV value is determined by the distance and the time difference between two wave patterns in the upper arm and ankle. This research studied six calculation methods to find specific reference point on the oscillometric waveform (OMW) from cuff pressure measurement to calculate time difference between two OMW patterns. The methods used were maximum value, minimum value, maximum first derivative, maximum second derivative, intersection between tangent of maximum first derivative and minimum value points, and intersection between tangent of maximum second derivative and minimum value points. baPWV calculation was performed during three levels of cuff pressure. At each pressure level, at least 3 waves were used for calculation. The best baPWV calculation method was determined by the smallest value of data variance. The method test was performed on 81 subjects in which 77 subjects were for general test and 4 subjects were for repetition test. General test results showed that the maximum second derivative method provides the most consistent baPWV value, meanwhile the repetition test results from 6 OMW data for each subject showed that the maximum first derivative method also gave the most consistent baPWV value with insignificant differences compared to the maximum second derivative method.
Bulletin of Electrical Engineering and Informatics, 2021
This paper aims to conduct an analysis of the SARS-CoV-2 genome variation was carried out by comp... more This paper aims to conduct an analysis of the SARS-CoV-2 genome variation was carried out by comparing the results of genome clustering using several clustering algorithms and distribution of sequence in each cluster. The clustering algorithms used are K-means, Gaussian mixture models, agglomerative hierarchical clustering, mean-shift clustering, and DBSCAN. However, the clustering algorithm has a weakness in grouping data that has very high dimensions such as genome data, so that a dimensional reduction process is needed. In this research, dimensionality reduction was carried out using principal component analysis (PCA) and autoencoder method with three models that produce 2, 10, and 50 features. The main contributions achieved were the dimensional reduction and clustering scheme of SARS-CoV-2 sequence data and the performance analysis of each experiment on each scheme and hyper parameters for each method. Based on the results of experiments conducted, PCA and DBSCAN algorithm achi...
Uploads
Papers by tati rajab Mengko