A5037666948- Medical Image Segmentation Techniques
- Brain Tumor Detection and Classification
- Advanced Neuroimaging Techniques and Applications
- Acute Ischemic Stroke Management
- Advanced Neural Network Applications
- Domain Adaptation and Few-Shot Learning
- AI in cancer detection
- ECG Monitoring and Analysis
- Face and Expression Recognition
- Medical Imaging and Analysis
- Advanced MRI Techniques and Applications
- EEG and Brain-Computer Interfaces
- Functional Brain Connectivity Studies
- Computer Science and Engineering
- Gene expression and cancer classification
- Cell Image Analysis Techniques
- Multimedia Learning Systems
- Data Mining and Machine Learning Applications
- COVID-19 diagnosis using AI
- Neural dynamics and brain function
- Neural Networks and Applications
- Currency Recognition and Detection
- Image and Signal Denoising Methods
- Edcuational Technology Systems
- Advanced Image and Video Retrieval Techniques
University of Indonesia
2011-2025
RIKEN Center for Brain Science
2021-2025
Universitas Palembang
2023-2024
Universitas Multi Data Palembang
2023-2024
Rumah Sakit Umum Pusat Nasional Dr. Cipto Mangunkusumo
2024
Kaken Pharmaceutical (Japan)
2022
RIKEN
2021-2022
University of Edinburgh
2016-2020
The prefrontal cortex (PFC) has dramatically expanded in primates, but its organization and interactions with other brain regions are only partially understood. We performed high-resolution connectomic mapping of the marmoset PFC found two contrasting corticocortical corticostriatal projection patterns: "patchy" projections that formed many columns submillimeter scale nearby distant "diffuse" spread widely across striatum. Parcellation-free analyses revealed representations gradients these...
Neural network attracts plenty of researchers lately. Substantial number renowned universities have developed neural for various both academically and industrially applications. shows considerable performance on purposes. Nevertheless, complex applications, network’s accuracy significantly deteriorates. To tackle the aforementioned drawback, lot researches had been undertaken improvement standard network. One most promising modifications applications is deep learning method. In this paper,...
Adaptive traffic signal control system is needed to avoid congestion that has many disadvantages. This paper presents an adaptive using camera as input sensor providing real-time data. Principal Component Analysis (PCA) used analyze and classify object on video frame for detecting vehicles. Distributed Constraint Satisfaction Problem (DCSP) method determine the duration of each signal, based counted number vehicles at lane. The implemented in embedded systems BeagleBoard™.
The left ventricular of ejection fraction is one the most important metric cardiac function. It used by cardiologist to identify patients who are eligible for life-prolonging therapies. However, assessment suffers from inter-observer variability. To overcome this challenge, we propose a deep learning approach, based on hierarchical vision Transformers, estimate echocardiogram videos. proposed method can without need ventrice segmentation first, make it more efficient than other methods. We...
We present MiniVess, the first annotated dataset of rodent cerebrovasculature, acquired using two-photon fluorescence microscopy. MiniVess consists 70 3D image volumes with segmented ground truths. Segmentations were created traditional processing operations, a U-Net, and manual proofreading. Code for preprocessing steps U-Net are provided. Supervised machine learning methods have been widely used automated biomedical images. While much emphasis has placed on development new network...
The primate brain has unique anatomical characteristics, which translate into advanced cognitive, sensory, and motor abilities. Thus, it is important that we gain insight on its structure to provide a solid basis for models will clarify function. Here, report the implementation features of Brain/MINDS Marmoset Connectivity Resource (BMCR), new open-access platform provides access high-resolution anterograde neuronal tracer data in marmoset brain, integrated retrograde tractography data....
Abstract Predicting the evolution of white matter hyperintensities (WMH), a common feature in brain magnetic resonance imaging (MRI) scans older adults (i.e., whether WMH will grow, remain stable, or shrink with time) is important for personalised therapeutic interventions. However, this task difficult mainly due to myriad vascular risk factors and comorbidities that influence it, low specificity sensitivity image intensities textures alone predicting evolution. Given predominantly nature...
Denoising astronomical images is a significant challenge in the field of data processing. Image acquired from sources typically contains noise various sources. The study aims to investigate denoising using an image-to-image translation approach with AttentionGAN method. This method combines attention-guided techniques Generative Adversarial Network (GAN) model improve quality noisy images. Attention-guided technique allows learn most important features image and guide generation process. has...
Magnetic resonance (MR) perfusion imaging non-invasively measures cerebral perfusion, which describes the blood's passage through brain's vascular network. Therefore, it is widely used to assess ischaemia. Convolutional Neural Networks (CNN) constitute state-of-the-art method in automatic pattern recognition and hence, segmentation tasks. But none of CNN architectures developed date have achieved high accuracy when segmenting ischaemic stroke lesions, being main reasons their heterogeneity...
Previous studies have indicated that white matter hyperintensities (WMH), the main radiological feature of small vessel disease, may evolve (i.e., shrink, grow) or stay stable over a period time. Predicting these changes are challenging because it involves some unknown clinical risk factors leads to non-deterministic prediction task. In this study, we propose deep learning model predict evolution WMH from baseline follow-up 1-year later), namely "Disease Evolution Predictor" (DEP) model,...
Abstract Pairwise image registration is a necessary prerequisite for brain comparison and data integration in neuroscience radiology. In this work, we explore the efficacy of implicit neural representations (INRs) improving performance magnetic resonance imaging. setting, INRs serve as continuous coordinate based approximation deformation field obtained through multi-layer perceptron. Previous research has demonstrated that sinusoidal representation networks (SIRENs) surpass ReLU models...
In the wake of use deep learning algorithms in medical image analysis, we compared performance algorithms, namely Boltzmann machine (DBM), convolutional encoder network (CEN) and patch-wise neural (patch-CNN), with two conventional schemes: Support vector (SVM) random forest (RF), for white matter hyperintensities (WMH) segmentation on brain MRI mild or no vascular pathology. We also all these approaches a method Lesion Segmentation Tool public toolbox named lesion growth algorithm (LGA)....
In this study, we introduce "instance loss functions", a new family of functions designed to enhance the training neural networks in instance-level segmentation and detection objects biomedical image data, particularly those varied numbers sizes. Intended be utilized conjointly with traditional functions, these proposed prioritize object instances over pixel-by-pixel comparisons. The specific instance (Linstance), center (Lcenter), false rate (Lfalse), proximity (Lproximity), serve distinct...
In the recent years, it has become readily more accepted that smart mobile phones with GPS or A-GPS enabled device, even Cell-ID enabled, among commuters, can be used as traffic sensor, which complements other traditional sensors. This development is pursued in efforts of reducing avoiding jams. Consequently, this paper attempts to find a novel way map match 2D local actual traces from phones. From number experiments, been found Virtual Detection Zone method obtain 100% matching, ensures...
Traffic plays an important role in social stability and community development. Without appropriate traffic signal control system, the possibility of congestion will be very high causes various negative impacts. The system with video camera sensor is implemented embedded systems using BeagleBoard-xM. uses Viola-Jones method Haar Training detecting vehicle object from a frame. Then, Euclidean distance kalman filter methods are used tracking vehicle. ability predicting next position feature for...
White matter hyperintensities (WMH) appear as regions of abnormally high signal intensity on T2-weighted magnetic resonance image (MRI) sequences. In particular, WMH have been noteworthy in age-related neuroscience for being a crucial biomarker all types dementia and brain aging processes. The automatic segmentation is challenging because their variable range, size shape. U-Net tackles this problem through the dense prediction has shown competitive performances not only...
The chromosome is a set of DNA structure that carry information about our life. can be obtained through Karyotyping. process requires clear image so the evaluate well. Preprocessing have to done on images enhancement. starts with background removing. will cleaned color. next step This paper compares several methods for We some method in enhancement like Histogram Equalization (HE), Contrast-limiting Adaptive (CLAHE), 3D Block Matching (HE+BM3D), and basic enhancement, unsharp masking....
In the era of Big Data, data size and security are issues that need to be solved. To address this problem, we may apply compression technique or encryption. On other hand, coding based method solution. The learn distribution reduce dimension with minimized loss information from original data. Stacked Unsupervised Extreme Learning Machine (Stacked US-ELM) is one fastest methods which can used it. problem how many stacks get optimal performance classifier. research, inspected (US-ELM) for...