Department of Intelligent Information Engineering, Research Promotion Unit,
School of Medical Sciences, Fujita Health University

News & Topics

  • June 1, 2021
 Our graduate student received the Student Research Encouragement Award from the Tokai Branch of the IEICE.
  • April 17, 2021
 Our bachelor student received Technical Newcomer Award from the Japanese Society of Radiological Technology.
  • April 17, 2021
 Our graduate student received Doi Award from the Journal of Radiological Physics and Technology.
  • April 17, 2021
 Our graduate student received the CyPos Gold and Bronze Award at the 76th Annual Meeting of the Japanese Society of Radiological Technology.
  • April 1, 2021
 Five master's students, one doctoral student, and eight undergraduate students have newly joined our lab.

Introduction

ICT technologies, such as artificial intelligence and information and communication systems, are important technologies that support modern society, and they have become essential technologies in the medical field as well. This field focus on the elemental and applied technologies to support diagnosis and decision making using various types of biological data.

Members

Faculty members

Radiological technology

Graduate students

Doctor’s course: Yujiro Doi (D3), Yoshitaka Isobe(D2)
Master’s course: Yuki Oshita (M2), Fumiaki Oba (M2), Ayana Sugiura (M2), Masahiro Tsukijima (M2), Yudai Higashi (M2), Patipipittana Supanuch (M1), Terumasa Kondo(M1), Yuta Suganuma(M1), Gakuto Hirano(M1)

Undergraduate students

Fourth grade : Amase Ito, Nanami Ogino, Momoko Kawajiri, Airi Kurata, Hina Kotani, Nozomi Sakakibara, Momona Tsunooka, Hiroki Nakagawa, Nahoko Nakamura, Mizuki Mori

Research Theme

・Computer Aided Diagnosis (CAD) (Atsushi Teramoto)

Computer Aided Diagnosis

The evolution of diagnostic imaging equipment has made it possible to obtain a large amount of image data in a short period of time with a single imaging session. However, the increase in the number of images has resulted in an ever-increasing burden on the physicians who perform the diagnosis, and there are concerns that this may lead to a decrease in the accuracy of the diagnosis.
Computer Aided Diagnosis, Computer Aided Detection (CAD) is a technology that uses a computer to automatically detect suspected lesions and automatically analyze the size, benignness or malignancy of the lesion. It is expected to reduce the burden of reading and improve the accuracy of diagnosis by allowing doctors to refer to the results output by CAD as a "second opinion".
Since 2014, we have been working vigorously on the medical application of deep learning, which is one of the most powerful forms of artificial intelligence. Below are some of the topics we are currently working on.

CAD for Lung imaging

Lung cancer is a leading cause of the cancer death worldwide. Therefore, early diagnosis and treatment are essential. CT and PET/CT have been widely used in the checkup because of its effectiveness of early detection of lung cancer. However, the burden on the diagnosing doctor is becoming higher due to the need to observe a large number of images.
We have been developing the automated detection and benign/malignant classification system of lung nodules using CT or PET/CT images. In recent years, the performance of the system is dramatically increased by introducing the deep learning.

CAD for Breast imaging

The incidence of breast cancer in Japanese women is 1 in 11, and the mortality rate is increasing year by year, so early detection and treatment are desirable. In addition to mammography, which is mainly used for screening, various modalities such as tomosynthesis, MRI, and PET are used for imaging tests of breast cancer. We are actively working on the research and development of computer-aided detection/diagnosis techniques using these imaging modalities. We are working on the development of a method for automatic detection of construction disorder in mammogram images. We are also developing an automated analysis method using breast MR images in order to accurately analyze the invasive area of breast cancer. We are also conducting research on risk estimation and differentiation between benign and malignant using deep learning and texture analysis.

CAD for Cardiac Imaging

Cardiovascular diseases account for about 30% of all deaths, and improvements in the accuracy of treatment and diagnosis of cardiovascular diseases are required. Early detection is essential to improve the prognosis while reducing the burden on patients with these diseases. Especially, myocardial infarction may cause the risk of fatal complications such as arrhythmia and heart failure, and early detection is important to improve the prognosis of patients and prevent sudden death. Therefore, our laboratory is developing automated detection techniques of myocardial infarction on echocardiography, which is a non-invasive examination, for the purpose of early detection of myocardial infarction.

CAD for Digestive Imaging

Gastric cancer is the second leading cause of death among Japanese men and the fourth leading cause of death among women, and early detection and treatment are required. Endoscopy has come to be used in medical examinations due to its reported effectiveness in reducing gastric cancer mortality. However, endoscopy requires the physician to detect lesions while examining the moving stomach, and there is a possibility that gastric cancer may be overlooked. In our laboratory, we are developing a system for automatic detection of gastric cancer from endoscopic images using artificial intelligence, which has been used in various fields in recent years.

CAD for Pathology

Pathological diagnosis to confirm a disease requires observation of a large number of specimens and cells, which is a high burden for screeners and pathologists. Therefore, we are conducting research to analyze pathological images using deep learning to automatically classify benign/malignant and cancer histological types.

・Efficient Visualization and Analysis of Clinical Information Base on Artificial Intelligence(AI) and Data Science (Tomoko Tateyama)

While CAD provides valuable information for diagnosis and clinical treatment, it is daily collecting medical images, biopsy information, and many other types of data of various types and properties, too. The effective establishment, storage, processing, and representation of the data will have a significant contribution to the enhancement of AI and CAD applications and their accuracy in the future.
Our research focuses on the following issues based on informatics, data science, and artificial intelligence:
・Data visualization and analysis in clinical scenes, and database publication (Fundamental research)
・Medical Data fusion and Analysis of multimodality based on Artificial Intelligence (Fundamental research)
・Assessment and Stage Estimation of Disease using 3D Morphological Changes of Organs based on AI (CAD)
・Gesture Analysis and Database Publication for Support Clinical Operations (Support for diagnostic, surgical and therapeutic)

Annual Events

Research meeting

Every week, undergraduate and graduate students present the progress of their research, and once a month we have an evening meeting with working graduate students, where we exchange various opinions to advance our research. In June, all students give presentations at a plenary seminar attended by graduates, current students, and co-researchers, where they discuss research plans and future initiatives.

Joint meeting

On-site and online joint seminars and camps are held with other universities to discuss each other's research and deepen friendships through barbecues and sports tournaments.

Technical exchange meetings with companies

We hold technology exchange meetings with several medical companies several times a year, where we learn about the technologies developed by the companies and how they work, as well as introduce our laboratory's technologies and receive advice from the company's perspective.

Academic conference presentations

Faculty members, graduate students, and undergraduate students will give presentations at domestic conferences such as the JSRT, JAMIT, MII, as well as at international conferences on medical imaging and engineering such as RSNA, IEEE, CARS, and SPIE.

Laboratory trip

Once a year, the lab members go on a trip or barbecue to deepen exchanges.

Photo Gallery

  • Research Meeting

  • Summer Seminar

  • Academic Conference (Yokohama)

  • International Conference (Hawaii)

  • International Conference (Chicago)

  • Laboratory Trip

Academic Activities

Original Papers ( English papers only )

2022

  1.  M.Yoshida, A.Teramoto, K.Kudo, S.Matsumoto, K.Saito, H.Fujita, "Automated extraction of cerebral infarction region in head MR image using pseudo cerebral infarction image by CycleGAN," Applied Sciences, Vol.12, 489, 2022.
  2. R.Muraki, A.Teramoto, K.Sugimoto, K.Sugimoto, A.Yamada, E.Watanabe, "Automated detection scheme for acute myocardial infarction using convolutional neural network and long short-term memory," PLOS ONE, Vol.17, No.2, e0264002, 2022.
  3. T.Tsukamoto, A.Teramoto, A.Yamada, Y.Kiriyama, E.Sakurai, A.Michiba, K.Imaizumi, H.Fujita, "Comparison of fine-tuned deep convolutional neural networks for the automated classification of lung cancer cytology images with integration of additional classifiers," Asian Pacific Journal of Cancer Prevention, Vol.23, No.4, pp.1315-1324.
  4. A.Watanabe, A.Teramoto, D.Hirose, "Automated Detection of Infusion and catheter Hub in Movies of Nurses’ Observation Scenes Using Deep Learning," 看護理工学会誌, in press.

2021

  1. A.Teramoto, T.Shibata, H.Yamada, Y.Hirooka, K.Saito, H.Fujita, "Automated Detection of Gastric Cancer by Retrospective Endoscopic Image Dataset Using U-Net R-CNN,” Applied Sciences, 11, 11275, 2021.
  2. Y.Doi, A.Teramoto, A.Yamada, M.Kobayashi, K.Saito, H.Fujita,“Estimating subjective evaluation of low-contrast resolution using convolutional neural networks,” Physical and Engineering Sciences in Medicine, online available, 2021.
  3. A.Teramoto, Y.Kiriyama, T.Tsukamoto, E.Sakurai, A.Michiba, K.Imaizumi, K.Saito, H.Fujita, “Weakly Supervised Learning for Classification of Lung Cytological Images Using Attention-Based Multiple Instance Learning,” Scientific Reports, 11:20317, 2021.
  4. N.Takeuchi, A.Teramoto, K.Imaizumi, K.Saito, H.Fujita, ”Analysis of Idiopathic Interstitial Pneumonia in CT Images U ing 3D U-Net,” Medical Image and Information Sciences, No.38, Vol.3, pp.126-131, 2021.
  5. H.Yamashiro, A.Teramoto, K.Saito, H.Fujita, "Development of a fully automated glioma-grading pipeline using post-contrast T1-weighted images combined with cloud-based 3D convolutional neural network,” Applied Sciences, 2021, Vol,11, 5118.
  6. T.Hayakawa, A.Teramoto, Y.Kiriyama, T.Tsukamoto, A.Yamada, K.Saito, H.Fujita, "Development of pathological diagnosis support system using micro-computed tomography," Acta Histochemica et Cytochemica, Vol.54, No.2, pp.49-55, 2021.
  7. M.Tsujimoto, A.Teramoto, M.Dosho, S.Tanahashi, A.Fukushima, S.Ota, Y.Inui, R.Matsukiyo, Y.Obama, H.Toyama, “Automated classification of increased uptake regions in bone SPECT/CT images using three-dimensional deep convolutional neural network,” Nuclear Medicine Communications, Vol.42, No.8, pp.877-883, 2021.
  8. A.Teramoto, A.Yamada, T.Tsukamoto, Y.Kiriyama, E.Sakurai, K.Shiogama, A.Michiba, K.Imaizumi, K.Saito, H.Fujita, "Mutual Stain Conversion between Giemsa and Papanicolaou in Cytological Images Using Cycle Generative Adversarial Network”, Heliyon, Vol.7, No.2, e06331, 2021.
  9. M.Tsujimoto, S.Shirakawa, M.Watanabe, A.Teramoto, M.Uno, S.Ota, R.Matsukiyo, T.Okui,Y.Kobayashi, H.Toyama, “Two- versus three-dimensional regions of interest for quantifying SPECT-CT images,” Physical and Engineering Sciences in Medicine, Vol.44, No.2, pp.365-375, 2021.
  10. R.Toda, A.Teramoto, M.Tsujimoto, H.Toyama, K.Imaizumi, K.Saito, H.Fujita, "Synthetic CT Image Generation of Shape-Controlled Lung Cancer using Semi-Conditional InfoGAN and Its Applicability for Type Classification," International Journal of Computer Assisted Radiology and Surgery, Vol.16, pp.241-251, 2021.
  11. Jiaqing Liu, Seijyu Tsujinaga, Shurong Chai, Hao Sun, Tomoko Tateyama, Yutaro Iwamoto, Xinyin Huang, Lanfen Lin, Yen-Wei Chen, “Single Image Depth Map Estimation for Improving Posture Recognition," in IEEE Sensors Journal, vol. 21, no. 23, pp. 26997-27004, 1 Dec.1, 2021.

2020

  1. R.Kimura, A.Teramoto, T.Ohno, K.Saito, H.Fujita, "Virtual Digital Subtraction Angiography Using Multizone Patch-based U-Net," Physical and Engineering Science in Medicine, Vol.43, pp.1305-1315, 2020.
  2. M.Sumitomo, A.Teramoto, R.Toda, N.Fukami, K.Fukaya, K.Zennami, M.Ichino, K.Takahara, M.Kusaka, R.Shiroki, "Deep learning using preoperative MRI information to predict early recovery of urinary continence after robot-assisted radical prostatectomy," International Journal of Urology, Vol.27, pp.922-928, 2020.
  3. A.Yamada, A.Teramoto, M.Hoshi, H.Toyama, K.Imaizumi, K.Saito, H.Fujita, "Hybrid Scheme for Automated Classification of Pulmonary Nodules Using PET/CT Images and Patient Information," Applied Sciences, Vol.10, No.12, 4225, 2020.
  4. K.Ote, F.Hashimoto, A.Kakimoto, T.Isobe, T.Inubushi, R.Ota, A.Tokui, A.Saito, T.Moriya, T.Omura, E.Yoshikawa, A.Teramoto, Y.Ouchi , “Kinetics-Induced Block Matching and 5D Transform Domain Filtering for Dynamic PET Image Denoising,” IEEE Transactions on Radiation and Plasma Medical Sciences, Vol.4, No.6, pp.720-728, 2020.
  5. T.Shibata, A.Teramoto, H.Yamada, N.Ohmiya, K.Saito, H.Fujita, ”Automated Detection and Segmentation of Early Gastric Cancer from Endoscopic Images Using Mask R-CNN,” Applied Sciences,Vol.10, No.11, 3842, 2020.
  6. Y.Onishi, A.Teramoto, M.Tsujimoto, T.Tsukamoto, K.Saito, H.Toyama, K.Imaizumi, H.Fujita, "Investigation of Pulmonary Nodule Classification Using Multi-Scale Residual Network Enhanced with 3DGAN-Synthesized Volumes,” Radiological Physics and Technology,13(2), 160-169, 2020.
  7. F.Hashimoto, K.Ote, T.Oida, A.Teramoto, Y.Ouchi, “Compressed-sensing Magnetic Resonance Image Reconstruction Using Iterative Convolutional Neural Network Approach,” Applied Sciences,10(6):1902, 2020.
  8. A.Teramoto, T.Tsukamoto, A.Yamada, Y.Kiriyama, K.Imaizumi, K.Saito, H.Fujita, "Deep learning approach to classification of lung cytological images: Two-step training using actual and synthesized images by progressive growing of generative adversarial networks,” PLoS ONE, Vol.15, No.3:e0229951, 2020.
  9. N.Matsubara, A.Teramoto, K.Saito, H.Fujita, "Bone suppression for chest X-ray image using a convolutional neural filter," Physical and Engineering Sciences in Medicine, Vol.43, pp.97–108,2020.
  10. Y.Onishi, A.Teramoto, M.Tsujimoto, T.Tsukamoto, K.Saito, H.Toyama, K.Imaizumi, H.Fujita, "Multiplanar analysis for pulmonary nodule classification in CT images using deep convolutional neural network and generative adversarial networks," International Journal of Computer Assisted Radiology and Surgery, Vol.15, pp.173-178, 2020.
  11. A.Sakai, Y.Onishi, M.Matsui, H.Adachi, A.Teramoto, K.Saito, H.Fujita,“A method for the automated classification of benign and malignant masses on digital breast tomosynthesis images using machine learning and radiomic features,” Radiological Physics and Technology, Vol.13, No.1, pp.27-36, 2020.

Conference

Please refer to Teramoto Laboratory's website for the achievements in international and domestic academic conferences.

Book (English book only)

  1. H.Fujita, T.Hara, X.Zhou, A.Teramoto, N.Kamiya, D.Fukuoka, C.Muramatsu, “Chapter 9: Function integrated diagnostic assistance based on MCA models,” Multidisciplinary Computational Anatomy: Toward Integration of Artificial Intelligence with MCA-based Medicine, M.Hashizume (eds.), Switzerland; Springer Nature Switzerland AG, 2021, in press.
  2. A.Teramoto, A.Yamada, T.Tsukamoto, K.Imaizumi, H.Toyama, K.Saito, H.Fujita, "Decision support system for lung cancer using PET/CT and microscopic images," in Deep Learning in Medical Image Analysis, Advances in Experimental Medicine and Biology 1213, G.Lee and H.Fujita (eds), Switzerland; Springer Nature Switzerland AG, pp.73-94, 2019.
  3. A. Teramoto and H. Fujita, “Automated lung nodule detection using positron emission tomography/computed tomography,” in Artificial Intelligence in Decision Support Systems for Diagnosis in Medical Imaging, K. Suzuki and Y. Chen, Eds. Cham, Switzerland; Springer International Publishing AG, pp.87-110, 2018.

Access

  • Access to Teramoto Laboratory ⇒ 410, 4th, University Building 11