13197 research outputs found
Sort by
スパイキング ニューラル ネットワーク エノ メムキャパシティブ インメモリ ケイサン カイロ ノ オウヨウ ト ヒョウカ
Get PDF奈良先端科学技術大学院大学修士 (工学)master thesi
Segmentation of Diffuse Lung Abnormality Patterns on Computed Tomography Images using Partially Supervised Learning
No full textComputer-aided diagnostic methods that provide semantic segmentation of texture patterns of diffuse lung diseases (DLDs) on chest computed tomography (CT) are extremely useful for detecting, identifying, and quantifying lung pathologies. While a fully annotated dataset is desirable to build a semantic segmentation model, building such a dataset for DLDs is costly due to the requirements of manual segmentation and certified experts for annotation. Partially supervised learning (PSL) has been proposed recently to take advantage of the partially annotated dataset and reduce the full annotation burden. Creating a partially annotated dataset is much less expensive than creating a fully annotated dataset. Therefore, PSL has great potential to build a semantic segmentation model that only requires a feasible amount of annotation. In this study, we propose a method of PSL employing a loss function that uses both annotated and unannotated pixels of a partially annotated dataset. The proposed loss function is based on the cross entropy loss, and it uses unannotated pixels to penalize the leakage of the segmentation. A parameter that controls the balance between the two types of supervision is introduced into the loss function to allow tuning and studying of the proposed PSL. The effectiveness and characteristics of PSL for the segmentation of DLD classes (consolidation, ground grass opacity, honeycombing, emphysema, and normal) were investigated in experiments using chest CT images of 372 patients. The experimental results show that the proposed PSL improved the mean Dice score from 0.76 to 0.79, and that a higher value of the balancing parameter increased the precision of the segmentation. Using the proposed PSL, which takes full advantage of the partially annotated dataset, we improved the accuracy of DLD segmentation. Furthermore, the experimental results clarified that the proposed PSL improved the precision of the models using unannotated pixels. Our implementation of the proposed PSL is available at https://github.com/yk-szk/psl-dld.journal articl
Molecular convergence by differential domain acquisition is a hallmark of chromosomal passenger complex evolution
No full textThe chromosomal passenger complex (CPC) is a heterotetrameric regulator of eukaryotic cell division, consisting of an Aurora-type kinase and a scaffold built of INCENP, Borealin, and Survivin. While most CPC components are conserved across eukaryotes, orthologs of the chromatin reader Survivin have previously only been found in animals and fungi, raising the question of how its essential role is carried out in other eukaryotes. By characterizing proteins that bind to the Arabidopsis Borealin ortholog, we identified BOREALIN RELATED INTERACTOR 1 and 2 (BORI1 and BORI2) as redundant Survivin-like proteins in the context of the CPC in plants. Loss of BORI function is lethal and a reduced expression of BORIs causes severe developmental defects. Similar to Survivin, we find that the BORIs bind to phosphorylated histone H3, relevant for correct CPC association with chromatin. However, this interaction is not mediated by a BIR domain as in previously recognized Survivin orthologs but by an FHA domain, a widely conserved phosphate-binding module. We find that the unifying criterion of Survivin-type proteins is a helix that facilitates complex formation with the other two scaffold components and that the addition of a phosphate-binding domain, necessary for concentration at the inner centromere, evolved in parallel in different eukaryotic groups. Using sensitive similarity searches, we find conservation of this helical domain between animals and plants and identify the missing CPC component in most eukaryotic supergroups. Interestingly, we also detect Survivin orthologs without a defined phosphate-binding domain, likely reflecting the situation in the last eukaryotic common ancestor.journal articl
Diminished Reality for Sense of Movement with XR Mobility Platform
Get PDFThis paper proposes a method to reduce a sense of movement by controlling a passenger's sense of movement using a multimodal XR system mounted on an autonomous vehicle. The passenger should not be given the sense of movement from the accelerating stimulus, in order to make the space comfortable during self-driving. In the field of VR, technologies have been developed to simulate a sense of movement that does not occur, such as in a driving simulator. However, the purpose of this paper is to solve the inverse problem to reduce the sense of movement generated by real autonomous driving. By using a multimodal XR mobility platform that consists of an immersive display and a motion platform with a tilting seat, we develop a state-of-the-art method to reduce the passenger's perceived sense of movement. This method enables one to perceive the vehicle as if they are not moving, even though they are in automated driving. In the experiments, we evaluated 20 participants under the condition of the combination of visual and force acceleration stimuli. From experimental results, it was confirmed that the sense of movement was reduced significantly when both visual and force acceleration stimuli were not presented in the condition.conference pape
Pumilio homolog2 Pum2 ノ シゼン メンエキ オウトウ チョウセツ ニ オケル キノウ カイメイ
Get PDF奈良先端科学技術大学院大学修士(バイオサイエンス)master thesi
Evaluation of Biocompatible Films Using Copolymers of N-Vinylbenzamide with Cationic Moieties by Hydrolysis of N-Vinylformamide
Get PDFBiocompatible materials using poly(N-vinylamide) derivatives are developed via a recently designed synthetic route. N-Vinylbenzamide (NVB) is synthesized as typical hydrophilic N-vinylamide monomer, then copolymerized with N-vinylformamide (NVF) by free radical polymerization, which can convert to cationic polyvinylamine unit by hydrolysis. Cell toxicity on the surface of copolymer films is investigated, revealing that poly(N-vinylamide) derivatives show biocompatibility regardless of the cationic ratio.journal articl
DFusion: Denoised TSDF Fusion of Multiple Depth Maps with Sensor Pose Noises
No full textThe truncated signed distance function (TSDF) fusion is one of the key operations in the 3D reconstruction process. However, existing TSDF fusion methods usually suffer from the inevitable sensor noises. In this paper, we propose a new TSDF fusion network, named DFusion, to minimize the influences from the two most common sensor noises, i.e., depth noises and pose noises. To the best of our knowledge, this is the first depth fusion for resolving both depth noises and pose noises. DFusion consists of a fusion module, which fuses depth maps together and generates a TSDF volume, as well as the following denoising module, which takes the TSDF volume as the input and removes both depth noises and pose noises. To utilize the 3D structural information of the TSDF volume, 3D convolutional layers are used in the encoder and decoder parts of the denoising module. In addition, a specially-designed loss function is adopted to improve the fusion performance in object and surface regions. The experiments are conducted on a synthetic dataset as well as a real-scene dataset. The results prove that our method outperforms existing methods.journal articl
複数スマートフォンのMACアドレスランダム化の非同期性を用いたODデータ自動取得手法の提案と評価
Get PDF路線バスにおいて，効率的な運行のために利用者の動向把握は必要不可欠である．利用動向の把握には，OD（Origin-Destination）データの使用が最も効果的であるが，現状，OD データの取得は人手に頼った手法が主流となっている．そのため，本研究では，OD データの自動取得を目的とし，プライバシーや金銭コストの観点から，スマートフォンが発するパケットを用いた OD データの自動取得手法について着目した．しかし，プライバシー保護の観点から，ランダム MAC アドレスが使用されるようになって以降，ランダム MAC アドレスを利用するスマートフォンに対する OD データの自動取得手法は確立されていない．本研究では，Bluetooth の広告パケットから handoff や nearby といった特徴量を抽出し，特徴量の変更と MAC アドレス変更の非同期性を利用したアドレス・キャリーオーバ・アルゴリズムをベースとした，MAC アドレスランダム化の非同期性を用いた OD データ自動取得手法を提案する．京都府京都市の路線を対象に評価実験を行った結果，本提案により，ランダム MAC アドレスを使用するスマートフォンが発するパケットを用いて，OD データを自動取得できることを確認した．technical repor