著者
井﨑 武士
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.34, no.2, pp.32-38, 2017

<p>In this paper, I explained the latest use cases of deep learning in several fields. Moreover, I also explained some use cases and the latest study of deep learning in medical imaging field. There are 3 reasons why deep learning adoption has been accelerated in a couple of years. One reason is GPU(Graphics Processing Unit). I explained required calculation performance in deep learning in general and how to utilize GPU power for better performance. Moreover, I also explained NVIDIA Deep Learning Platform including the latest GPU hardware, Deep Learning SDK(Software Development kit)and DIGITS(Deep Learning GPU Training System)software.</p>
著者
Hidemi KAMEZAWA Hidetaka ARIMURA Ryuji YASUMATSU Kenta NINOMIYA Shu HASEAI
出版者
MEDICAL IMAGING AND INFORMATION SCIENCES
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.37, no.4, pp.66-74, 2020-12-23 (Released:2020-12-25)
参考文献数
47

We have investigated a non-invasive approach for predicting parotid gland cancer (PGC) malignancy grade based on radiomic biomarkers in preoperative magnetic resonance (pMR) images using six conventional machine learning (cML) and five deep learning (DL) algorithms. 39 patients were divided into 70% (27 patients) for a training dataset and 30% (12 patients) for a test dataset. A total of 972 hand-crafted features were extracted from cancer regions on the twodimensional T1- and T2-weighted pMR images, and then hand-crafted biomarkers were obtained by a least absolute shrinkage and selection operator (LASSO) in the training dataset for six cML models. Five DL models were constructed by transfer learning of pre-trained DL architectures, i.e. AlexNet, GoogLeNet, VGG-16, ResNet-101, and DenseNet-201. Highgrade versus intermediate- plus low-grades malignant PGCs was predicted using the eleven prediction models for the test dataset. The VGG-16-based DL model demonstrated a highest accuracy of 85.4% among the eleven models for the test dataset, which was a higher accuracy than the histological diagnostic accuracy of 79.5% using fine needle aspiration cytology (FNAC). The MR-based DL approach could be feasible for preoperatively and non-invasively predicting the grades of PGC malignancy.
著者
山田 朋奈 李 鎔範 長谷川 晃
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.36, no.2, pp.83-87, 2019-06-30 (Released:2019-06-28)
参考文献数
19

The purpose of this paper is to develop a computerized classification method for right or left and directions of arms in forearm X-ray images using a deep convolutional neural network(DCNN). 648 radiographs were obtained by using X-ray lower arm phantoms. These images were downsized to 213×256 pixels and used as training and test images in the DCNN. AlexNet and GoogLeNet were used as the DCNN. All radiographs were classified to eight categories by the DCNN. Classification accuracies were obtained by nine-fold cross validation tests. The accuracies using AlexNet and GoogLeNet were 79.3% and 92.6%, respectively. GoogLeNet would be useful to classify forearm radiographs automatically. The proposed method may contribute to quality assurance for medical images.
著者
町田 治彦
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.30, no.4, pp.76-82, 2013 (Released:2013-12-25)
参考文献数
30

With flat panel detector radiography or x-ray TV systems, digital tomosynthesis (DT) has been recently introduced as limited-angle CT, which can remove overlying structures, enhance local tissue separation, and provide depth information about structures of interest by providing tomographic images of high quality. DT is advantageous in various clinical applications, including chest, head and neck, orthopedic, fluoroscopic, and mammographic studies. Specifically, compared to conventional radiography or fluoroscopy, DT can improve the detection of pulmonary nodules, the diagnosis of paranasal sinusitis, and the delineation of subtle fractures, particularly in complex anatomic structures, and gastrointestinal and breast tumors by less tissue overlap. Compared to CT, DT can reduce radiation exposure by limited sweep angle and improve the assessment of fine bony changes by better in-plane resolution and the postoperative evaluation in patients with metallic prostheses and osteosynthesis materials by less metallic artifact. DT is also useful for functional, weight-bearing and stress tests by more arbitrary patient positioning. To accomplish adequate patient management, a comprehensive understanding of the clinical utilities and limitations of those DT applications is important for the improvement of diagnostic quality, workflow, and cost-effectiveness.
著者
芳野 由利子 陸 慧敏 金 亨燮 村上 誠一 青木 隆敏 木戸 尚治
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.36, no.2, pp.77-82, 2019-06-30 (Released:2019-06-28)
参考文献数
14

A temporal subtraction image is obtained by subtracting a previous image, which are warped to match between the structures of the previous image and one of a current image, from the current image. The temporal subtraction technique removes normal structures and enhances interval changes such as new lesions and changes of existing abnormalities from a medical image. However, many artifacts remain on a temporal subtraction image and these can be detected as false positives on the subtraction images. In this paper, we propose a 3D-CNN after initial nodule candidates are detected using temporal subtraction technique. To compare the proposed 3D-CNN, we used 7 model architectures, which are 3D ShallowNet, 3D-AlexNet, 3D-VGG11, 3D-VGG13, 3D-ResNet8, 3D-ResNet20, 3D-ResNet32, with these performance on 28 thoracic MDCT cases including 28 small-sized lung nodules. The higher performance is showed on 3D-AlexNet.
著者
町田 好男
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.32, no.4, pp.ii-v, 2015-12-25 (Released:2015-12-26)
参考文献数
14

The purpose of this article is to describe the recent development of imaging techniques in MRI. After the brief review of the basics of MR imaging, new techniques such as simultaneous multi‐slice techniques, compressed sensing (CS) MRI and MR fingerprinting are discussed.
著者
本谷 秀堅
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.34, no.2, pp.39-44, 2017-06-30 (Released:2017-06-30)
参考文献数
42

Deep neural networks(DNNs)play important roles in medical image processing : Given a set of enough number of training data, one can construct a high-performance pattern recognition/regression machine by using a DNN. In this article, the author briefly describes a research trend of DNN in a field of medical image processing.
著者
齋藤 哲史
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.25, no.3, pp.37-39, 2008-07-01 (Released:2008-10-03)
参考文献数
7

日本の医療体制は崩壊の瀬戸際にある, といっても過言ではありません. 政府による医療費抑制策が原因です. しかし, 医療費は本当に抑制し なければならないのでしょうか. 医療は, 健康・長寿を実現するのに必要なサービスです. また, 高齢化や所得水準の向上に伴って支出に占める割合が上昇するのは必然です . それは他の先進国の例からも明らかでしょう. それにもかかわらず, 政府は財政的見地から医療費抑制策を30 年近くも続けてきました. 現在の 状況は, その報いと言えるのかもしれません. 今回の講演では, 医療崩壊の経済学と題して, 「医療費抑制の是非」, 「後期高齢者医療制度」, 「将来の医療保険設計」の3 点について言及し ましたが, 内容を一部変更・省略しております. 何卒ご容赦ください.
著者
大田 英揮
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.30, no.1, pp.12-17, 2013 (Released:2013-03-26)
参考文献数
14

The number of CT examinations has increased over the last decade using multidetector CT. With the development of technology, multidetector CT, enabling scan faster and wider, can be used as a one-stop-shop procedure in many clinical settings. Routine CT scan protocols are necessary to meet the growing clinical demands. However, we often encounter situations where routine protocols are not enough to assess a patient's condition and specify the diagnosis. Tips for CT image acquisition that can help diagnosis and treatment are discussed through case presentations.
著者
新川 慶明
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.30, no.3, pp.49-52, 2013 (Released:2013-08-28)
参考文献数
13

In late years Autopsy imaging (Ai) attracts attention. Ai is the concept that matched autopsy with postmortem imaging in a wide sense. On the other hand, Ai is postmortem imaging itself in a narrow sense. I explain a role of Ai when a clinician makes a death certificate. Specifically, I introduce examples of Ai which I experienced, tips about examination for Ai-CT and interpretation.
著者
神島 保
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.33, no.1, pp.4-7, 2016-03-25 (Released:2016-03-26)
参考文献数
8

Current magnetic resonance imaging (MRI) techniques and high‐quality coil system can provide exquisite anatomic detail of the wrist and fingers. They present a valuable tool for assessment and characterization of traumatic injuries to small structures, various arthropathies, and neoplastic processes. As imaging quality continues to improve, the clinical significance of MRI of this field becomes more apparent in the evaluation and management of these entities.
著者
名誉顧問 長谷川 伸
出版者
MEDICAL IMAGING AND INFORMATION SCIENCES
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.25, no.2, pp.22-25, 2008

本学会の大会が150回を迎えました. 50年後に予想される300回大会の折り返し点, 半世紀先のその日にも「医用画像情報学会」は輝きを保っているすばらしい学会と思います. 東京オリンピックの年, 学会を立ち上げて第1回を開催された先達をはじめ, 今日まで44年の長期に亘り大会開催に学会運営にと尽力された役員の方々, 何より研究活動を通して学会を支え, 大会を盛り上げてこられた会員各位にお祝いとお礼を申し上げます. ここでは今後のご参考になろうかと, 古い時代に運営を手伝った立場から歴史を振り返りますが, 記憶違いや偏見もあろうと思います. お許しください.
著者
久保 走一
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.3, no.2, pp.57-61, 1986 (Released:2012-08-27)

This is a brief introduction to image science and engineering education at Chiba University. A short history and the present status of the education are given, and an outline of the newly established PhD program in image science is described.
著者
清水 聡 杉田 純一 アザバジック アンジェラ テイラー ロバート
出版者
MEDICAL IMAGING AND INFORMATION SCIENCES
雑誌
医用画像情報学会雑誌 = Japanese journal of imaging and information sciences in medicine (ISSN:09101543)
巻号頁・発行日
vol.24, no.2, pp.55-60, 2007-10-01

The evolution of a Modarity represented by MSCT brings the explosion of the quantity of inspection data. It is easy to treat these large data intuitively interactively quickly, and the system which a technologist and a doctor can treat is bought. However, there is a limit in conventional PACS system and WS, and expensive H/W to realize transfer time and the high handling of data set is necessary in large quantities. We think, the solution which is cheap, and can use these volume data browsing feature while all doctors have 3D performance same as a work station with every computer being an ideal solution. Furthermore, we developed a new backbone platform called Aquarius iNtuition to solve a problem in the workflow.
著者
齋藤 哲史
出版者
MEDICAL IMAGING AND INFORMATION SCIENCES
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.25, no.3, pp.37-39, 2008

日本の医療体制は崩壊の瀬戸際にある, といっても過言ではありません. 政府による医療費抑制策が原因です. しかし, 医療費は本当に抑制し なければならないのでしょうか. 医療は, 健康・長寿を実現するのに必要なサービスです. また, 高齢化や所得水準の向上に伴って支出に占める割合が上昇するのは必然です . それは他の先進国の例からも明らかでしょう. それにもかかわらず, 政府は財政的見地から医療費抑制策を30 年近くも続けてきました. 現在の 状況は, その報いと言えるのかもしれません. 今回の講演では, 医療崩壊の経済学と題して, 「医療費抑制の是非」, 「後期高齢者医療制度」, 「将来の医療保険設計」の3 点について言及し ましたが, 内容を一部変更・省略しております. 何卒ご容赦ください.
著者
小森 優
出版者
医用画像情報学会
雑誌
医用画像情報学会雑誌 (ISSN:09101543)
巻号頁・発行日
vol.23, no.4, pp.108-111, 2006 (Released:2006-12-05)
参考文献数
12

Tactile information are utilized almost everywhere in medical practice. Although the information is important, it is difficult to learn/teach, reproduce and record them. Recent virtual reality technology in medical field can handle a part of tactile information. This paper summarizes tactile sensory of human and progress in haptic devices, and introduces some recent medical applications with haptics. Technical aspects in haptic rendering, delay/latency in haptic response and time/accuracy trade-off in physical elastic simulation are mentioned.