論文 - 林田　哲

Fluorodeoxyglucose-positron emission tomography as a potential alternative tool for functional diagnosis of glycogen storage disease type I.

Sato T, Inokuchi M, Nakano S, Iwabuchi Y, Hayashida T, Ishii T, Hasegawa T

Radiology case reports 18 （ 1 ） 91 - 93 2023年01月

共著,  ISSN  1930-0433

　概要を見る

A 43-year-old woman with genetically confirmed glycogen storage disease type Ib was suspected to have left breast cancer. Fluorodeoxyglucose-positron emission tomography showed high fluorodeoxyglucose accumulation in the whole liver as well as left mammary gland. We consider that high fluorodeoxyglucose accumulation in the liver of patients with glycogen storage disease type I is caused by impaired glucose-6-phosphate metabolism due to the congenital deficiency of glucose-6-phosphatase activities in hepatocytes. This study describes fluorodeoxyglucose-positron emission tomography as a potential alternative tool to diagnose glycogen storage disease type I functionally.

• Access to Document (DOI)

Establishment of a deep-learning system to diagnose BI-RADS4a or higher using breast ultrasound for clinical application.

Hayashida T, Odani E, Kikuchi M, Nagayama A, Seki T, Takahashi M, Futatsugi N, Matsumoto A, Murata T, Watanuki R, Yokoe T, Nakashoji A, Maeda H, Onishi T, Asaga S, Hojo T, Jinno H, Sotome K, Matsui A, Suto A, Imoto S, Kitagawa Y

Cancer science 113 （ 10 ） 3528 - 3534 2022年10月

共著,  ISSN  1347-9032

　概要を見る

Although the categorization of ultrasound using the Breast Imaging Reporting and Data System (BI-RADS) has become widespread worldwide, the problem of inter-observer variability remains. To maintain uniformity in diagnostic accuracy, we have developed a system in which artificial intelligence (AI) can distinguish whether a static image obtained using a breast ultrasound represents BI-RADS3 or lower or BI-RADS4a or higher to determine the medical management that should be performed on a patient whose breast ultrasound shows abnormalities. To establish and validate the AI system, a training dataset consisting of 4028 images containing 5014 lesions and a test dataset consisting of 3166 images containing 3656 lesions were collected and annotated. We selected a setting that maximized the area under the curve (AUC) and minimized the difference in sensitivity and specificity by adjusting the internal parameters of the AI system, achieving an AUC, sensitivity, and specificity of 0.95, 91.2%, and 90.7%, respectively. Furthermore, based on 30 images extracted from the test data, the diagnostic accuracy of 20 clinicians and the AI system was compared, and the AI system was found to be significantly superior to the clinicians (McNemar test, p < 0.001). Although deep-learning methods to categorize benign and malignant tumors using breast ultrasound have been extensively reported, our work represents the first attempt to establish an AI system to classify BI-RADS3 or lower and BI-RADS4a or higher successfully, providing important implications for clinical actions. These results suggest that the AI diagnostic system is sufficient to proceed to the next stage of clinical application.

• Access to Document (DOI)

Comparison between a new assay system, Elecsys(®) Anti-p53, and conventional MESACUP™ for the detection of serum anti-p53 antibodies: A multi-institutional study.

Suzuki T, Oshima Y, Shiratori F, Nanami T, Yajima S, Sumazaki M, Ushigome M, Sugita H, Eberl M, Ogata H, Hayashida T, Nakamura S, Nakagawa T, Shimada H

Molecular and clinical oncology 17 （ 2 ） 130 2022年08月

共著,  ISSN  2049-9450

　概要を見る

The sensitivity and specificity of a new automated electrochemiluminescence immunoassay system, Elecsys® Anti-p53 (Elecsys), were compared with that of the conventional serum anti-p53 antibody (s-p53-Ab) enzyme-linked immunosorbent assay kit [MESACUP anti-p53 test (MESACUP)]. Elecsys and MESACUP were used to analyze the levels of s-p53-Abs in patients with esophageal, colorectal and breast cancer. A total of 532 controls and 288, 235 and 329 patients with esophageal, colorectal and breast cancer, respectively, were enrolled. Additionally, the sera of patients with benign diseases of the esophagus, colorectal system and breast, patients with autoimmune diseases and healthy volunteers were analyzed as controls. Sensitivity and specificity were compared between the two assay systems. Positive agreement rates were 58.7% in all samples, 71.2% in esophageal samples, 73.6% in colorectal samples and 35.1% in breast samples. Negative agreement rates for the different cancer types were ≥97.1% and the overall agreement rates were ≥92.3%. When the specificities of the two assays were aligned for all samples, Elecsys demonstrated higher sensi-tivities for all types of analyzed cancer together, as well as for esophageal, colorectal and breast cancer, respectively. Although positive concordance between the two assay systems was low in terms of specificity, Elecsys had a higher sensitivity than the MESACUP.

• Access to Document (DOI)

Observational study of axilla treatment for breast cancer patients with 1 to 3 positive micrometastases or macrometastases in sentinel lymph nodes.

Yasojima, H; Imoto, S; Nagashima, T; Onishi, T; Takashima, T; Kitada, M; Kawada, M; Hayashida, T; Naoi, Y; Aihara, T; Wada, N; Kawabata, H; Yoshida, M; Toh, U; Yoneyama, K; Yamada, A; Tsuda, H; Masuda, N; Saito-Oba, M; Sakamoto, J

JOURNAL OF CLINICAL ONCOLOGY 40 （ 16 ）  2022年06月

共著,  ISSN  0732-183X

Feasibility study on collecting patient-reported outcomes from breast cancer patients using the LINE-ePRO system.

Hayashida T, Nagayama A, Seki T, Takahashi M, Matsumoto A, Kubota A, Jinno H, Miyata H, Kitagawa Y

Cancer science 113 （ 5 ） 1722 - 1730 2022年05月

共著,  ISSN  1347-9032

　概要を見る

Due to the increasing complexity of cancer treatment, ensuring safety and maintaining the quality of life during treatment are important issues. Patient-reported outcomes (PROs) in oncology are essential for assessing patient symptoms. A feasibility study was undertaken on breast cancer patients by building a PRO data collection system based on LINE, one of the most popular social network service applications in Japan. In this study, one or more predefined PRO questions for each breast cancer patient's clinical situation were sent to the patient’s LINE application daily. The patient selected a predefined answer by tapping the screen, but no free-text answers were allowed. Seventy-three patients were enrolled. The median observation period was 435 days (84-656 days), and the total number of PROs collected was 16,417, with a mean of 224.9 reports per patient. Patients on adjuvant endocrine therapy were notified of 2.5 questions per week, and the median number of responses per week and response rate were 2.387 (1.687-11.627) and 95.5%, respectively. Analyzing the results by age group, the number of responses from those aged 60 and above was equal to or higher than that of the younger age group. It was also possible to track each patient’s PROs accurately. These results suggested that the design of the system, based on an application used daily, instead of using specifically prepared applications for collecting electronic PROs, was the reason for the favorable acceptance from patients and the satisfactory response rate from all age groups, including the elderly.

• Access to Document (DOI)

Spatiotemporal Video Highlight by Neural Network Considering Gaze and Hands of Surgeon in Egocentric Surgical Videos

Yoshida K., Hachiuma R., Tomita H., Pan J., Kitani K., Kajita H., Hayashida T., Sugimoto M.

Journal of Medical Robotics Research 7 （ 1 ）  2022年03月

共著,  ISSN  2424905X

　概要を見る

In the medical field, surgical videos can be used to introduce surgical skills. Medical students and residents watch the videos to study the surgical skills and increase learning speed by compensating for the lack of experience in surgical rooms due to limited opportunity to join in surgery. To record egocentric surgical videos by a wearable camera is a solution to record surgical skills of a surgeon in detail. However, most egocentric surgical videos are of quite long duration. For example, in the case of tumor removal in breast surgery, a video recording time often reaches 2h. With that length, it is time consuming to see important scenes in the video, particularly because many surgical videos include nonessential scenes such as sterilization and preparation of tools. For extracting specific scenes from a long video, we can apply scene estimation by machine learning. Furthermore, it is important to know where the surgeon is looking to observe the area of the incision in detail. In particular, it is vital to be able to zoom in on key elements, allowing viewers to see the incision area and the fine details of the necessary surgical skills. In this study, we aimed to highlight incision scenes from egocentric surgical videos in the spatiotemporal domain by utilizing two neural networks for the temporal and spatial highlights. For the temporal highlights, we designed a neural network that estimates the incision scenes by learning gaze speed, hand movements, number of hands, and background movements in egocentric surgical videos. For the spatial highlights, in order to estimate the important area to zoom in, we designed a neural network that learns the surgeon's gaze on natural features of surgical scenes to form a probability map as a representation of the estimated gaze area. The estimated gaze area was also used to calculate the appropriate zoom-in position and zoom-in ratio. To control the highlighted parameters in accord with user preferences, we also made a user interface that allows for the selection of playback speed gain and zoom ratio gain. For the evaluation, we verified the performance of the networks by a quantitative assessment and conducted a user study with medical doctors by showing an actual surgical video to obtain a qualitative assessment on the proposed system.

• Access to Document (DOI)

The relationship between work-related outcomes and symptoms in early breast cancer survivors receiving adjuvant endocrine therapy.

Nakao M, Komatsu H, Hayashida T, Takahashi M, Seki T, Yagasaki K

Asia-Pacific journal of oncology nursing 9 （ 3 ） 174 - 178 2022年03月

共著,  ISSN  2347-5625

　概要を見る

Objective: This study examined the relationship between symptom burdens and work-related outcomes, including work participation and overall work impairment (OWI) among breast cancer survivors (BCS) receiving adjuvant endocrine therapy (AET). Methods: This was a cross-sectional study with 140 BCS of working age receiving AET. Data were collected using self-report questionnaires that included an assessment of symptoms and their employment status, and OWI. Data were analyzed using descriptive statistics and multiple logistic regression analysis. Results: A total of 111 (79%) survivors reported being employed at the time of the survey. Symptom burdens were not associated with unemployment. Of the 110 working BCS receiving AET, symptom burdens were significantly related to a higher degree of OWI (OR ​= ​2.14, 95% CI, 1.58–2.89, P ​≤ ​0.001). Conclusions: Participating BCS receiving AET continued to work while experiencing symptoms, with survivors who experienced high symptom burdens being negatively affected in their work life. Healthcare providers need to assess and manage symptoms and their impact on work, with the help of employers, to improve the quality of work life of BCS receiving AET.

• Access to Document (DOI)

乳房超音波検査におけるAI診断システムの開発

林田 哲, 小谷 依里奈, 永山 愛子, 関 朋子, 高橋 麻衣子, 北川 雄光

超音波検査技術抄録集 （一般社団法人 日本超音波検査学会）  47 （ 0 ） S82 - S82 2022年

共著

• Access to Document (DOI)
• CiNii Research

特集 AIは外科にどこまで役立つか I. 総論 2.手術映像におけるAIによる時間的・空間的推定技術の確立と応用

杉本 麻樹, 梶田 大樹, 林田 哲, 北川 雄光

外科 （南江堂）  83 （ 11 ） 1160 - 1165 2021年10月

共著,  ISSN  0016593X

• Access to Document (DOI)
• CiNii Research

Appropriate use of cancer comprehensive genome profiling assay using circulating tumor DNA.

Sunami K, Bando H, Yatabe Y, Naito Y, Takahashi H, Tsuchihara K, Toyooka S, Mimori K, Kohsaka S, Uetake H, Kinoshita I, Komine K, Takeda M, Hayashida T, Tamura K, Nishio K, Yamamoto N, Working Group of a Joint Task Force of Three Academic Societies for the Promotion of Cancer Genomic Medicine

Cancer science 112 （ 9 ） 3911 - 3917 2021年09月

共著,  ISSN  1347-9032

　概要を見る

Comprehensive genomic profiling (CGP) is being increasingly used for the routine clinical management of solid cancers. In July 2018, the use of tumor tissue-based CGP assays became available for all solid cancers under the universal health insurance system in Japan. Several restrictions presently exist, such as patient eligibility and limitations on the opportunities to perform such assays. The clinical implementation of CGP based on plasma circulating tumor DNA (ctDNA) is also expected to raise issues regarding the selection and use of tissue DNA and ctDNA CGP. A Joint Task Force for the Promotion of Cancer Genome Medicine comprised of three Japanese cancer-related societies has formulated a policy proposal for the appropriate use of plasma CGP (in Japanese), available at https://www.jca.gr.jp/researcher/topics/2021/files/20210120.pdf, http://www.jsco.or.jp/jpn/user_data/upload/File/20210120.pdf, and https://www.jsmo.or.jp/file/dl/newsj/2765.pdf. Based on these recommendations, the working group has summarized the respective advantages and cautions regarding the use of tissue DNA CGP and ctDNA CGP with reference to the advice of a multidisciplinary expert panel, the preferred use of plasma specimens over tissue, and multiple ctDNA testing. These recommendations have been prepared to maximize the benefits of performing CGP assays and might be applicable in other countries and regions.

• Access to Document (DOI)

Recovery of lower oesophageal barrier function: a pilot study comparing a mixture of atropine and neostigmine and sugammadex <i>A randomised controlled pilot study</i>

Suganuma, E; Ishikawa, T; Kitamura, Y; Hayashida, T; Matsumura, T; Fujie, M; Nozaki-Taguchi, N; Sato, Y; Isono, S

EUROPEAN JOURNAL OF ANAESTHESIOLOGY 38 （ 8 ） 856 - 864 2021年08月

共著,  ISSN  0265-0215

• Access to Document (DOI)

A transposon screen identifies enhancement of NF-κB pathway as a mechanism of resistance to eribulin.

Teng X, Hayashida T, Murata T, Nagayama A, Seki T, Takahashi M, Kitagawa Y

Breast cancer (Tokyo, Japan) 28 （ 4 ） 884 - 895 2021年07月

共著,  ISSN  1340-6868

　概要を見る

Background: Eribulin mesylate (eribulin) is an efficient microtubule inhibitor that is used for metastatic breast cancer. However, breast cancer can develop resistance to eribulin. This resistance mechanism needs to be elucidated. Methods: A transposon mutagenesis screen was conducted using a pPB-SB-CMV-puro-SD plasmid and pCMV-PBase transposase. Viability and cytotoxicity were analyzed by MTT assay and flow cytometry, respectively. Real-time PCR and western blot were used for gene expression analysis. In addition, vivo study was also designed to analyze therapy efficiency. Results: TAB2, which is part of the nuclear factor-kappa B (NF-κB) pathway, was identified as a candidate eribulin-resistant gene. TAB2 down-regulation resulted in significantly lower cell viability and higher cytotoxicity of cells treated with eribulin, while TAB2 up-regulation showed opposite results. Similarly, combination of NF-κB inhibitors [Bay-117082 and QNZ (quinazoline derivative)] with eribulin showed significantly lower cell viability and higher drug cytotoxicity than single agent treatment with eribulin in MDA-MB-231 cells. However, QNZ increased NF-κB activity in MCF7 cells by up-regulating TAB2, which reduced the sensitivity to eribulin. Furthermore, combination of Bay-117082 with eribulin induced greater regression of MDA-MB-231 tumors compared to eribulin monotherapy in vivo. Conclusions: These results consistently illustrated that TAB2-NF-κB pathway may increases resistance to eribulin in breast cancer models. Moreover, these results support the use of a combination strategy of eribulin with NF-κB inhibitors, and provide evidence that transposon mutagenesis screens are capable of identifying drug-resistant genes.

• Access to Document (DOI)

Comprehensive analysis of the homeobox family genes in breast cancer demonstrates their similar roles in cancer and development.

Nakashoji A, Hayashida T, Yamaguchi S, Kawai Y, Kikuchi M, Yokoe T, Nagayama A, Seki T, Takahashi M, Kitagawa Y

Breast cancer research and treatment 186 （ 2 ） 353 - 361 2021年04月

共著,  ISSN  0167-6806

　概要を見る

Background: The homeobox (HOX) family consists of 39 genes whose expressions are tightly controlled and coordinated within the family, during development. We performed a comprehensive analysis of this gene family in cancer settings. Methods: Gene correlation analysis was performed using breast cancer data available in The Cancer Genome Atlas (TCGA) and data from the patients admitted to our hospital. We also analyzed the data of normal breast tissue (GSE20437). We next collected gene expression and prognosis data of breast cancer patients (GSE11121, GSE7390, GSE3494, and GSE2990) and performed unsupervised hierarchal clustering by the HOX gene expression pattern and compared prognosis. We additionally performed this analysis to leukemia (available in TCGA) and sarcoma (GSE20196) data. Results: Gene correlation analysis showed that the proximal HOX genes exhibit strong interactions and are expressed together in breast cancer, similar to the expression observed during development. However, in normal breast tissue, less interactions were observed. Breast cancer microarray meta-data classified by the HOX gene expression pattern predicted the prognosis of luminal B breast cancer patients (p = 0.016). Leukemia (p = 0.00016) and sarcoma (p = 0.018) presented similar results. The Wnt signaling pathway, one of the major upstream signals of HOX genes in development, was activated in the poor prognostic group. Interestingly, poor prognostic cancer presented stronger correlation in the gene family compared to favorable prognostic cancer. Conclusion: Comprehensive analysis of the HOX family demonstrated their similar roles in cancer and development, and indicated that the strong interaction of HOX genes might be specific to malignancies, especially in the case of poor prognostic cancer.

• Access to Document (DOI)

Estimating copy number using next-generation sequencing to determine ERBB2 amplification status.

Nakamura K, Aimono E, Oba J, Hayashi H, Tanishima S, Hayashida T, Chiyoda T, Kosaka T, Hishida T, Kawakubo H, Kitago M, Okabayashi K, Funakoshi T, Okita H, Ikeda S, Takaishi H, Nishihara H

Medical oncology (Northwood, London, England) 38 （ 4 ） 36 2021年03月

共著,  ISSN  1357-0560

　概要を見る

Assessing Erb-b2 receptor tyrosine kinase 2 (ERBB2) amplification status in breast and gastric cancer is necessary for deciding the best therapeutic strategy. Immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) are currently used for assessing protein levels and gene copy number (CN), respectively. The use of next-generation sequencing (NGS) to measure ERBB2 CN in breast cancer is approved by the United States Federal Drug Administration as a companion diagnostic. However, a CN of less than 8 is evaluated as “equivocal”, which means that some ERBB2 amplification cases diagnosed as “HER2 negative” might be false-negative cases. We reviewed the results of gene profiling targeting 160 cancer-related genes in breast (N = 90) and non-breast (N = 19) cancer tissue, and compared the ERBB2 CN results with the IHC/FISH scores. We obtained an estimated CN from the measured CN by factoring in the histological proportion of tumor cells and found that an ERBB2-estimated CN of 3.2 or higher was concordant with the combined IHC/FISH outcome in 98.4% (88/90) of breast cancer cases, while this was not always evident among non-breast cancer cases. Therefore, NGS-estimated ERBB2 CN could be considered a diagnostic test for anti-HER2 therapy after the completion of adequate prospective clinical trials.

• Access to Document (DOI)

Predictors of invasive disease in patients preoperatively diagnosed with ductal carcinoma without stromal invasion, with breast magnetic resonance imaging (MRI) and ultrasound (US).

Maeda H, Hayashida T, Watanuki R, Kikuchi M, Nakashoji A, Yokoe T, Seki T, Takahashi M, Kitagawa Y

Breast cancer (Tokyo, Japan) 28 （ 2 ） 398 - 404 2021年03月

共著,  ISSN  1340-6868

　概要を見る

Background: A preoperative diagnosis of ductal carcinoma in situ (DCIS) is sometimes upstaged to invasive disease postoperatively. Our objective was to clarify the predictive factors of invasive disease using preoperative imaging and to investigate the positive ratio of sentinel lymph nodes (SLN) and the incidence of invasive disease. Methods: The subjects were 402 patients with preoperatively diagnosed ductal carcinoma without stromal invasion who underwent breast surgery with concomitant SLN surgery in January 2007 to December 2016. Of the 306 included patients, all 306 patients underwent preoperative MRI and US assessment. Outcomes were analyzed for significance using univariate and multivariate analyses. Results: Of the 306 patients, 115 (37.6%) had invasive disease and 191 (62.4%) had DCIS only. Of the 115 patients with invasive disease, 5 (4.4%) and 4 (3.5%) had macro- and micrometastases in SLN. On the other hand, of the 191 patients with DCIS, only 1 (0.5%) had a micrometastasis. Predictors of invasive disease in the univariate analysis included having a palpable mass, were varied by biopsy method, having a US hypoechoic mass, MRI enhancement, or MRI large enhanced lesion; the size of the mass enhancement ≥ 1.1 cm or a spread of non-mass enhancement ≥ 3.1 cm (P = 0.003). Predictors of invasive disease in the multivariate analysis included US hypoechoic mass and MRI large enhanced lesion. Conclusion: We need to perform SLN biopsy for preoperatively diagnosed DCIS when patients have predictors of invasive disease, but SLN biopsy will no longer be essential for patients when they have no predictors of invasive disease.

• Access to Document (DOI)

乳がん患者の化学療法誘発性末梢神経障害と身体活動および生活の質の関連

杉本 美希, 小松 浩子, 林田 哲

日本がん看護学会誌 （一般社団法人 日本がん看護学会）  35 （ 0 ） n/a 2021年02月

共著,  ISSN  09146423

　概要を見る

<p>本研究の目的は，タキサン系薬剤を用いた乳がん患者の化学療法誘発性末梢神経障害（CIPN）と，身体活動および生活の質（HRQOL）の関連を明らかにすることである．タキサン系薬剤を用いた乳がん患者43名に対し，基本情報，疾患・治療関連情報，CIPN（Ntx subscale），HRQOL（FACT-G），身体活動（IPAQ-SF），抑うつ・不安（K6）について自記式質問紙および診療録で調査した．分析の結果，CIPNはHRQOL（<i>r</i>＝0.55, <i>p</i> ＜ 0.001），HRQOLの身体面（<i>r</i>＝0.72, <i>p</i> ＜ 0.001），精神面（<i>r</i>＝0.41, <i>p</i> ＜ 0.01），機能面（<i>r</i>＝0.47, <i>p</i> ＜ 0.01）と有意な正の相関がみられた．CIPNと身体活動，身体活動とHRQOLの相関は有意ではなかった．CIPNの症状が強い19名において，身体活動は7名（36.8％）がinactive，12名（63.2％）がactiveであった．CIPNは乳がん患者の生活に影響する症状であり，症状があっても自己管理を行い，症状に耐えて個々の役割や環境に応じて身体活動を行っている者がいることが考えられた．看護師は，CIPNによる日常生活と身体活動への影響について具体的に確認し，活動する背景について受け止めを聞き，支援することが重要である．そして，CIPNをかかえて安全に身体活動を行う方法を個別に検討することが必要である．</p>

• Access to Document (DOI)
• CiNii Research

Clinical practice guidance for next-generation sequencing in cancer diagnosis and treatment (edition 2.1).

Naito Y, Aburatani H, Amano T, Baba E, Furukawa T, Hayashida T, Hiyama E, Ikeda S, Kanai M, Kato M, Kinoshita I, Kiyota N, Kohno T, Kohsaka S, Komine K, Matsumura I, Miura Y, Nakamura Y, Natsume A, Nishio K, Oda K, Oda N, Okita N, Oseto K, Sunami K, Takahashi H, Takeda M, Tashiro S, Toyooka S, Ueno H, Yachida S, Yoshino T, Tsuchihara K, Japanese Society of Medical Oncology, Japan Society of Clinical Oncology, Japanese Cancer Association

International journal of clinical oncology 26 （ 2 ） 233 - 283 2021年02月

共著,  ISSN  1341-9625

　概要を見る

Background: To promote precision oncology in clinical practice, the Japanese Society of Medical Oncology, the Japanese Society of Clinical Oncology, and the Japanese Cancer Association, jointly published “Clinical practice guidance for next-generation sequencing in cancer diagnosis and treatment” in 2017. Since new information on cancer genomic medicine has emerged since the 1st edition of the guidance was released, including reimbursement for NGS-based multiplex gene panel tests in 2019, the guidance revision was made. Methods: A working group was organized with 33 researchers from cancer genomic medicine designated core hospitals and other academic institutions. For an impartial evaluation of the draft version, eight committee members from each society conducted an external evaluation. Public comments were also made on the draft. The finalized Japanese version was published on the websites of the three societies in March 2020. Results: The revised edition consists of two parts: an explanation of the cancer genomic profiling test (General Discussion) and clinical questions (CQs) that are of concern in clinical practice. Particularly, patient selection should be based on the expectation that the patient's post-test general condition and organ function will be able to tolerate drug therapy, and the optimal timing of test should be considered in consideration of subsequent treatment plans, not limited to treatment lines. Conclusion: We expect that the revised version will be used by healthcare professionals and will also need to be continually reviewed in line with future developments in cancer genome medicine.

• Access to Document (DOI)

【乳癌診療におけるAIの応用:画像診断、病理診断、データ解析の分野など】乳房超音波AIの実用化に向けた研究開発

林田 哲

乳癌の臨床 36   31 - 37 2021年

共著

• CiNii Research

Camera Selection for Occlusion-Less Surgery Recording via Training with an Egocentric Camera

Saito Y., Hachiuma R., Saito H., Kajita H., Takatsume Y., Hayashida T.

IEEE Access 9   138307 - 138322 2021年

共著,  ISSN  2169-3536

　概要を見る

Recording surgery is an important technique for education and the evaluation of medical treatments. However, capturing targets such as the surgical field, surgical tools, and the surgeon's hands, is almost impossible since these targets are heavily occluded by the surgeon's head and body during a surgery. We used a recording system in which multiple cameras are installed in the surgical lump, supposing at least one camera would capture the target without occlusion. As this system records multiple video sequences, we address the task to select a best view camera automatically. Recently, learning-based approaches in a fully supervised manner have been proposed for this task, but these previous approaches completely rely on manual annotation of the training data. In this paper, we focus on the eye tracker mounted on the surgeon's head, which can capture the recording targets without occlusion. Employing this first-person-view video synchronized with multiple videos of the surgical lump, we propose a novel camera selection approach using a self-supervised learning framework. In experiments, we created a dataset composed of four different breast surgery. Our extended experiments showed that our approach successfully switched to the best camera view without manual annotation and achieved competitive accuracy compared with conventional supervised methods. Also, our approach yielded effective visual representations comparable to state-of-the-art self-supervised learning frameworks.

• Access to Document (DOI)

Impact of neoadjuvant and adjuvant chemotherapy on invasive lobular carcinoma: A propensity score-matched analysis of SEER data.

Watanuki R, Hayashida T, Yokoe T, Kawai Y, Kikuchi M, Nakashoji A, Nagayama A, Seki T, Takahashi M, Kitagawa Y

The breast journal 26 （ 9 ） 1765 - 1770 2020年09月

共著,  ISSN  1075-122X

• Access to Document (DOI)