A5013991279- Medical Imaging and Analysis
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Medical Image Segmentation Techniques
- Dental Radiography and Imaging
- Medical Imaging Techniques and Applications
- Advanced X-ray and CT Imaging
- Brain Tumor Detection and Classification
- Lung Cancer Diagnosis and Treatment
- COVID-19 diagnosis using AI
- Infrared Thermography in Medicine
- Anatomy and Medical Technology
- Image Retrieval and Classification Techniques
- Cerebrovascular and Carotid Artery Diseases
- Retinal Imaging and Analysis
- Advanced Image and Video Retrieval Techniques
- Intracranial Aneurysms: Treatment and Complications
- Cutaneous Melanoma Detection and Management
- Robotics and Sensor-Based Localization
- Image and Object Detection Techniques
- Face and Expression Recognition
- Digital Radiography and Breast Imaging
- Inertial Sensor and Navigation
- Gene expression and cancer classification
- Spine and Intervertebral Disc Pathology
Kindai University
2018-2024
Nihon University
2018
University of Tokyo Hospital
2008-2017
The University of Tokyo
2011-2014
Tokyo University of Agriculture and Technology
2003-2006
Background The usefulness of computer‐assisted detection (CAD) for detecting cerebral aneurysms has been reported; therefore, the improved performance CAD will help to detect aneurysms. Purpose To develop a system intracranial on unenhanced magnetic resonance angiography (MRA) images based deep convolutional neural network (CNN) and maximum intensity projection (MIP) algorithm, demonstrate by training evaluating it using large dataset. Study Type Retrospective study. Subjects There were 450...
<h3>BACKGROUND AND PURPOSE:</h3> Experiences with computer-assisted detection of cerebral aneurysms in diagnosis by radiologists real-life clinical environments have not been reported. The purpose this study was to evaluate the usefulness a routine reading environment. <h3>MATERIALS METHODS:</h3> During 39 months practice environment, 2701 MR angiograms were each read 2 using system. Initial interpretation independently made without system, followed possible alteration after referring lesion...
Abstract Background Melanoma is a type of superficial tumor. As advanced melanoma has poor prognosis, early detection and therapy are essential to reduce melanoma‐related deaths. To that end, there need develop quantitative method for diagnosing melanoma. This paper reports the development such diagnostic system using hyperspectral data (HSD) convolutional neural network, which machine learning. Materials Methods HSD were acquired imager, spectrometer can simultaneously capture information...
Objective . To develop automatic visceral fat volume calculation software for computed tomography (CT) data and to evaluate its feasibility. Methods A total of 24 sets whole-body CT anthropometric measurements were obtained, with three each four BMI categories (under 20, 20 25, 25 30, over 30) in both sexes. True volumes defined on the basis manual segmentation by an experienced radiologist. Software automatically calculate was developed using a region technique based morphological analysis...
Introduction: The performance of computer-assisted detection (CAD) software depends on the quality and quantity dataset used for supervised learning. To realize continuous clinical use improvement CAD software, it is necessary to continuously collect data learning in practical improve by retraining with collected data. In this study, we investigated cerebral aneurysm based classifier through a simulation-based study. Methods: We during our retrained false positive (FP) reduction using effect...
Muography is a novel method of visualizing the internal structures active volcanoes by using high-energy near-horizontally arriving cosmic muons. The purpose this study to show feasibility muography forecast eruption event with aid convolutional neural network (CNN). In study, seven daily consecutive muographic images were fed into CNN compute probability eruptions on eighth day, and our model was trained hyperparameter tuning Bayesian optimization algorithm. By data acquired in Sakurajima...
Objective: We describe a new false positive (FP) reduction method based on surface features in our computerized detection system for lung nodules and evaluate the using clinical chest computed tomography (CT) scans. Methods: In method, nodule candidates are extracted volumetric curvature-based thresholding region growing. For various sizes of nodules, we adopt multiscale integration Hessian eigenvalues. each candidate, two calculated to differentiate FPs at vessel bifurcations. These fed...
Anatomical landmarks are useful as the primitive anatomical knowledge for medical image understanding. In this study, we construct a unified framework automated detection of distributed within human body. Our includes following three elements; (1) initial candidate based on local appearance matching technique models built by PCA and generative learning, (2) false positive elimination using classifier ensembles trained MadaBoost, (3) final landmark set determination combination optimization...
Segmentation of vertebral bones in computed tomographic data is important as a first stage image-based radiological tasks. However, it challenging problem to segment an affected spine correctly. In this study, we propose new method segmentation thoracic and lumbar bodies from thin-slice images. Especially, focus on deformable model-based scheme confirm the feasibility clinical sets with various bone diseases, such metastases scoliosis. As application algorithm, virtual straightening...
Objective: The objective of the current study was to develop a novel, artificial intelligence (AI)-based system diagnose coronavirus disease (COVID-19) using computed tomography (CT) slice images. Prior research has demonstrated that, if not focused on lungs, AI diagnoses COVID-19 information outside lungs. inclusion CT training data from multiple facilities and models may also cause with features that are irrelevant COVID-19. Thus, evaluate combination lung mask images single facility,...
Color information is an important tool for diagnosing melanoma. In this study, we used a hyperspectral imager (HSI), which can measure color in detail, to develop automated melanoma diagnosis system. recent years, the effectiveness of deep learning has become more widely accepted field image recognition. We therefore integrated convolutional neural network with transfer into our tried data augmentation demonstrate how system improves diagnostic performance. 283 lesions and 336 non-melanoma...