Kyoto University · Medicine
Professor Mizuho Nishio's research lab specializes in medical image analysis and artificial intelligence applications in oncology, with a focus on improving diagnostic accuracy and efficiency in lung and abdominal cancers. The lab develops advanced deep learning models—particularly U-Net variants—combined with innovative data augmentation and generative techniques like GANs for automated organ and tumor segmentation in CT and MRI. Key research directions include radiomics, low-dose CT image reconstruction, and adaptive imaging protocols to reduce radiation exposure while maintaining diagnostic quality. The lab also emphasizes transfer learning and model generalization to enhance clinical applicability across diverse imaging datasets.
Figures are computed from collected data and may differ slightly.
Since the publication of the Radiologic Diagnostic Oncology Group Report in 1991, the clinical application of pulmonary magnetic resonance imaging (MRI) in patients with lung cancer has been limited. In contrast, MRI for lung cancer has undergone continuous development, and several promising techniques have been introduced to overcome the previously suggested limitations. In addition, comparative studies involving multidetector-row computed tomography and positron emission tomography or positron
The use of adaptive iterative dose reduction using 3D processing resulted in greater consistency of emphysema quantification by low-dose CT, with quantification by standard-dose CT.
<b>Purpose:</b> The purpose of this study was to develop and evaluate lung cancer segmentation with a pretrained model and transfer learning. The pretrained model was constructed from an artificial dataset generated using a generative adversarial network (GAN). <b>Materials and Methods:</b> Three public datasets containing images of lung nodules/lung cancers were used: LUNA16 dataset, Decathlon lung dataset, and NSCLC radiogenomics. The LUNA16 dataset was used to generate an artificial dataset f
Combinations of data augmentation methods and deep learning architectures for automatic pancreas segmentation on CT images are proposed and evaluated. Images from a public CT dataset of pancreas segmentation were used to evaluate the models. Baseline U-net and deep U-net were chosen for the deep learning models of pancreas segmentation. Methods of data augmentation included conventional methods, mixup, and random image cropping and patching (RICAP). Ten combinations of the deep learning models a
Endometrial cancer (EC) is the most common gynecological tumor in developed countries, and preoperative risk stratification is essential for personalized medicine. There have been several radiomics studies for noninvasive risk stratification of EC using MRI. Although tumor segmentation is usually necessary for these studies, manual segmentation is not only labor-intensive but may also be subjective. Therefore, our study aimed to perform the automatic segmentation of EC on MRI with a convolutiona
ULDCT with and without IR can substitute for SDCT in emphysema quantification.
This retrospective study aimed to develop and validate a deep learning model for the classification of coronavirus disease-2019 (COVID-19) pneumonia, non-COVID-19 pneumonia, and the healthy using chest X-ray (CXR) images. One private and two public datasets of CXR images were included. The private dataset included CXR from six hospitals. A total of 14,258 and 11,253 CXR images were included in the 2 public datasets and 455 in the private dataset. A deep learning model based on EfficientNet with
Pulmonary magnetic resonance (MR) imaging has been put forward as a new research and diagnostic tool mainly to overcome the limitations of computed tomography and nuclear medicine studies. However, pulmonary MR imaging has been difficult to use because of inherently low proton density, a multitude of air-tissue interfaces, which create significant magnetic field distortions and are commonly referred to as susceptibility artifacts; diminishing signal in the lung; and respiratory and/or cardiac mo
The usefulness of sparse-sampling CT with deep learning-based reconstruction for detection of metastasis of malignant ovarian tumors was evaluated. We obtained contrast-enhanced CT images (n = 141) of ovarian cancers from a public database, whose images were randomly divided into 71 training, 20 validation, and 50 test cases. Sparse-sampling CT images were calculated slice-by-slice by software simulation. Two deep-learning models for deep learning-based reconstruction were evaluated: Residual En
We aimed to develop and evaluate an automatic prediction system for grading histopathological images of prostate cancer. A total of 10,616 whole slide images (WSIs) of prostate tissue were used in this study. The WSIs from one institution (5160 WSIs) were used as the development set, while those from the other institution (5456 WSIs) were used as the unseen test set. Label distribution learning (LDL) was used to address a difference in label characteristics between the development and test sets.
Natural language processing using language models has yielded promising results in various fields. Language models can help improve the workflow of radiologists. This retrospective study aimed to construct and evaluate language models for automatic summarization of radiology reports. Two radiology report datasets from the MIMIC Chest X-ray (MIMIC-CXR) database and the Japan Medical Image Database (JMID) were included in this study. The MIMIC-CXR is an open database comprising chest radiograph re
Open papers in the app to read, cite, and organize with AI.
