A5090819421- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Digital Radiography and Breast Imaging
- Advanced X-ray and CT Imaging
- MRI in cancer diagnosis
- COVID-19 diagnosis using AI
- Breast Cancer Treatment Studies
- Electron and X-Ray Spectroscopy Techniques
- Face recognition and analysis
- Prostate Cancer Treatment and Research
- Anomaly Detection Techniques and Applications
- Prostate Cancer Diagnosis and Treatment
- Traffic Prediction and Management Techniques
- Head and Neck Cancer Studies
- Bladder and Urothelial Cancer Treatments
- Generative Adversarial Networks and Image Synthesis
- Colorectal Cancer Screening and Detection
- Advanced Electron Microscopy Techniques and Applications
- Lung Cancer Diagnosis and Treatment
- Brain Tumor Detection and Classification
- Interactive and Immersive Displays
- Network Security and Intrusion Detection
- Teleoperation and Haptic Systems
- Machine Learning in Healthcare
- Image and Signal Denoising Methods
Tohoku University Hospital
2021-2025
Tohoku University
2023-2025
Keio University
2018-2023
St. Marianna University School of Medicine
2022-2023
St. Luke's International Hospital
2023
Oregon Health and Science University Hospital
2023
Jiangxi Agricultural University
2021
University of Electro-Communications
2018
Abstract Self-supervised learning (SSL) has gained attention in the medical field as a deep approach utilizing unlabeled data. The Jigsaw puzzle task SSL enables models to learn both features of images and positional relationships within images. In breast cancer diagnosis, radiologists evaluate not only lesion-specific but also surrounding structures. However, that adopt diagnostic similar human are still limited. This study aims effectiveness characterizing tissue structures for...
Background/Objectives: Delta-radiomics involves analyzing feature variations at different acquisition time-points. This study aimed to assess the utility of delta-radiomics analysis applied contrast-enhanced (CE) and non-contrast-enhanced (NE) T1-weighted images (WI) in predicting therapeutic response chemoradiotherapy (CRT) patients diagnosed with muscle-invasive bladder cancer (MIBC). Methods: Forty-three non-metastatic MIBC (cT2–4N0M0) who underwent partial or radical cystectomy after...
We investigated the value of deep learning (DL) in differentiating between benign and metastatic cervical lymph nodes (LNs) using pretreatment contrast-enhanced computed tomography (CT). This retrospective study analyzed 86 234 (non-metastatic) LNs at levels I–V 39 patients with oral squamous cell carcinoma (OSCC) who underwent preoperative CT neck dissection. were randomly divided into training (70%), validation (10%), test (20%) sets. For sets, I–II evaluated. Convolutional neural network...
Abstract Aims Current approaches to classify chronic heart failure (HF) subpopulations may be limited due the diversity of pathophysiology and co‐morbidities in HF. We aimed elucidate clusters patients with HF by data‐driven machine learning a hospital‐based registry. Methods results A total 4649 broad spectrum left ventricular ejection fraction (LVEF) CHART‐2 (Chronic Heart Failure Analysis Registry Tohoku District‐2) study were enrolled this study. Chronic classified using random forest...
Deep learning using convolutional neural networks (CNN) has achieved significant results in various fields that use images. can automatically extract features from data, and CNN extracts image by convolution processing. We assumed increasing the size interpolation methods would result an effective feature extraction. To investigate how change as number of data increases, we examined compared effectiveness augmentation inversion or rotation with when for training were small. Further,...
Ants are known to use a colony-specific blend of cuticular hydrocarbons (CHCs) as pheromone discriminate between nestmates and non-nestmates the CHCs were sensed in basiconic type antennal sensilla (S. basiconica). To investigate functional design this sensilla, we observed ultra-structures at 2D 3D Japanese carpenter ant, Camponotus japonicus, using serial block-face scanning electron microscope (SBF-SEM), conventional high-voltage transmission microscopes. Based on images 352 cross...
Background: High-resolution medical images often need to be downsampled because of the memory limitations hardware used for machine learning. Although various image interpolation methods are applicable downsampling, effect data preprocessing on learning performance convolutional neural networks (CNNs) has not been fully investigated. Methods: In this study, five different pixel algorithms (nearest neighbor, bilinear, Hamming window, bicubic, and Lanczos interpolation) were downsampling...
To assess the effectiveness of vViT model for predicting postoperative renal function decline by leveraging clinical data, medical images, and image-derived features; to identify most dominant factor influencing this prediction.
Segmentation of electron microscopic continuous section images by deep learning has attracted attention as a technique to reduce the cost annotation for researchers attempting make observations using 3D reconstruction methods. However, when observed samples are rare, or scanning circumstances unstable, pursuing generalization performance newly obtained is not appropriate. We assume transductive setting that predicts all labels in dataset from only partially while avoiding pursuit unknown...
In breast cancer diagnosis and treatment, non-invasive prediction of axillary lymph node (ALN) metastasis can help avoid complications related to sentinel biopsy.This study aims develop evaluate machine learning models using radiomics features extracted from diffusion-weighted whole-body imaging with background signal suppression (DWIBS) examination for predicting the ALN status.A total 100 patients histologically proven, invasive, clinically N0 who underwent DWIBS consisting short tau...
Predicting outcomes after intracerebral hemorrhage (ICH) may help improve patient outcomes. We developed and validated a machine learning prediction model for post-rehabilitation functional ICH. Patient selection explanatory variable settings were based on clinical significance. Functional predicted using ternary classification. The subjects patients aged > 18 years without pre-onset severe disability who primary putaminal and/or thalamic underwent an inpatient rehabilitation program. As...
Japan has four types of intensive care units (ICUs) that are divided into two categories according to the management fee charged per day: ICU fees 1 and 2 (ICU1/2) (equivalent high-intensity staffing) 3 4 (ICU3/4) low-intensity staffing). Although ICU1/2 charges a higher rate than ICU3/4, no cost-effectiveness analysis been performed for ICU1/2. This study evaluated clinical outcomes compared with those ICU3/4.This retrospective observational used nationwide Japanese administrative database...
While deep learning (DL) models have shown promise in breast cancer diagnosis using digital tomosynthesis (DBT) images, the impact of varying matrix sizes and image interpolation methods on diagnostic accuracy remains unclear. Understanding these effects is essential to optimize preprocessing steps for DL models, which can lead more efficient training processes, improved accuracy, better utilization computational resources. Our institutional review board approved this retrospective study...
Annotation of medical images, such as MRI and CT scans, is crucial for evaluating treatment efficacy planning radiotherapy. However, the extensive workload professionals limits their ability to annotate large image datasets, posing a bottleneck AI applications in imaging. To address this, we propose In-context Cascade Segmentation (ICS), novel method that minimizes annotation requirements while achieving high segmentation accuracy sequential images. ICS builds on UniverSeg framework, which...
You have accessJournal of UrologyProstate Cancer: Detection & Screening II (PD19)1 May 2024PD19-09 ARE THERE DIFFERENCES IN MRI FINDINGS BETWEEN CRIBRIFORM AND NON-CRIBRIFORM CANCER? AN ANALYSIS USING RADIOMICS DELTA-RADIOMICS Koki Enomoto, Soichiro Yoshida, Haruto Izumi, Sho Uehara, Yoh Matsuoka, Kohei Yamamoto, Daisuke Hirahara, Tatsunori Saho, Eichi Takaya, Shohei Fukuda, Yuma Waseda, Hajime Tanaka, Kenichi Ohashi, and Yasuhisa Fujii EnomotoKoki Enomoto , YoshidaSoichiro Yoshida...
You have accessJournal of UrologyImaging/Uroradiology II (MP30)1 May 2024MP30-08 DELTA-RADIOMICS ANALYSIS IN COMPARISON TO RADIOMICS USING DYNAMIC COMPUTED TOMOGRAPHY FOR PREOPERATIVE RISK STRATIFICATION UPPER URINARY TRACT UROTHELIAL CARCINOMA Motohiro Fujiwara, Daisuke Hirahara, Tatsunori Saho, Eichi Takaya, Shunya Matsumoto, Kasumi Yoshitomi, Masaki Kobayashi, Yuki Nakamura, Bo Fan, Yudai Ishikawa, Shohei Fukuda, Yuma Waseda, Hajime Tanaka, Soichiro Yoshida, and Yasuhisa Fujii...