Kyutacar : Kyushu Institute of Technology Academic Repository
Not a member yet
9746 research outputs found
Sort by
カモスタットメシル酸塩のCOVID-19の治療薬開発へのModel Informed Approachと臨床成績とのギャップ分析
Get PDF九州工業大学博士（情報工学）1 緒言| 2 カモスタットメシル酸塩の第I 相試験における母集団薬物動態解析と臨床有効
用量の予測| 3 カモスタットメシル酸の第III 相試験の成績とのGAP 分析| 4 総括医薬品の開発は、成功確率が低い上に開発コストも高く開発期間も長いため、その生産性の向上は長年の課題である。近年、注目されているのが、様々な課題（Research Question）に対して数理モデルを用いて解決を図る手法（Model Informed Approach）であり、Model Informed Drug Development（MIDD）と呼ばれている。一般に、生理システムを表現した数理モデルを用いるBottom-up／Mechanistic Model Informed Approach と、取得データの特徴を表す数理モデルを統計的手法により選択・構築するTop-down／Empirical Model Informed Approach に分類される。
医薬品開発初期の標的とメカニズムの理解から医薬品開発後期の薬事規制上の意思決定と、医薬品開発においてModel Informed Approach の目的は多岐にわたっており、その目的に応じてBottom-up/Top-down Model Informed Approach を選択することが重要である（Fit-for-purpose）。
Sever acute respiratory syndrome coronavirus 2（SARS-CoV-2）による感染症（COVID-19）は、2019 年末に初めて確認された新型のコロナウイルスによる感染症であり、以降、世界的に大流行した。カモスタットメシル酸塩は、宿主細胞膜上のII 型膜貫通型セリンプロテアーゼ（transmembrane protease serine 2: TMPRSS2）を阻害し宿主細胞へのウイルス侵入を抑制することでSARS-CoV-2 の増殖を抑制し、COVID-19 の治療薬となることが期待されていた。
小野薬品工業により実施された高用量のカモスタットメシル酸塩の反復投与時の安全性を確認することを目的とした健康成人対象の第I 相試験（NCT04451083）成績をもとに、高用量カモスタットメシル酸塩の薬物動態を特徴づける母集団薬物動態解析を実施した。
さらに、構築されたカモスタットメシル酸塩の母集団薬物動態モデルと、カモスタットメシル酸塩のSARS-CoV-2 の細胞侵入抑制作用を評価した非臨床試験成績を用いて、カモスタットメシル酸塩のCOVID-19 治療薬としての臨床有効用法・用量を予測した。
しかし、当該用法・用量を用いてCOVID-19 患者を対象に実施された第III 相試験
（NCT04657497）において、主評価であるSARS-CoV-2 の陰性化までの時間や副次評価であるウイルス量の推移では、プラセボ群に対してカモスタットメシル酸塩投与群で統計的に有意な差を示すことができず、期待するようなカモスタットメシル酸塩の有効性を確認することが出来なかった。
そこで、抗ウイルス薬の用法・用量を予測する際に想定すべき条件・仮説を明らかにすることを目的に、第I 相試験に成績に基づく有効用法・用量の予測結果と第III 相試験成績とのギャップ分析を行った。公表されたウイルス動態の数理モデルに、カモスタットメシル酸塩の母集団薬物動態モデルと非臨床成績を組み込むことで、カモスタットメシル酸塩を投与したときのウイルス動態を予測する数理モデルを構築した。
構築した数理モデルを用いて、カモスタットメシル酸塩の1 日あたりの投与量とSARS-CoV-2 感染を起点とした投与タイミングについて様々な条件におけるSARS-CoV-2 のウイルス動態を予測するとともに、予測値と実測値を比較することでギャップ分析を行った。
その結果、カモスタットメシル酸塩の投与量不足及び／又は投与の遅れが第III 相試験で有効性を示すことが出来なかった理由であると考えられた。投与量不足に関しては、ウイルスの増殖過程で関わる組織中の薬物動態（薬物濃度）が解析に反映できていない可能性があると考えられた。また、臨床試験を予測する際は、実臨床を加味した投与タイミングを仮定することが重要であると考えられた。九州工業大学博士学位論文 学位記番号:情工博甲第399号 学位授与年月日:令和6年6月28日令和6年度doctoral thesi
フードピックアンドプレースタスクにおけるロボティック認識向上のための合成データセットの生成
Get PDF九州工業大学博士（工学）1 INTRODUCTION| 2 LITERATURE REVIEW| 3 RESEARCH METHODOLOGY| 4 EXPERIMENTAL RESULTS AND DISCUSSION| 5 CONCLUSIONS AND FUTURE WORKThe main purpose of this research is to enhance robot vision to improve robot performance in food classification, detection, and pick and place into the lunch box.
The scale performance of neural networks can be improved by training them with large datasets. However, constructing a large-scale food dataset requires significant time and effort. This thesis introduces a Cycle Generative Adversarial Network (Cycle-GAN) as a solution to generate a large pseudo-realistic food dataset. This approach utilizes a large number of simulated images and a small number of real images, minimizing the need for traditional techniques of real data collections. To capture real RGB-D images of various food samples, an RGB-D camera and a turntable were employed from two different angles. Simulated images of 3D food models were generated using 3D modeling software and a 3D scanner, following the same configuration as the captured real images. Subsequently, a VGG-16 model was trained and evaluated using the generated RGB dataset. The results demonstrate that Cycle-GAN is an effective tool for generating pseudo realistic images, significantly reducing the efforts required for real image acquisition.
Although the proposed method successfully generates a large food scale dataset, a trained Cycle-GAN suffers from low-resolution output images due to the resolution of the input images. This limitation might contribute to the low accuracy of the neural network. Furthermore, more diverse feature variations of the food are necessary to enhance the performance of Cycle-GAN. However, providing additional variation images poses a time-consuming challenge. Moreover, the manual annotation of the dataset remains ineffective, emphasizing the need for an efficient dataset generation method with automatic annotation. Improving the accuracy of object detection is a crucial aspect to address.
To implement object detection and a pick-and-place system, the Robot Operating System (ROS) was utilized due to its system integration flexibility. The RGB images obtained from the RGB-D camera were employed for the object detection task by feeding them into the trained You Only Look At CoefficienTs (YOLACT) neural network. A PyBullet simulator software was used to generate simulated images based on 3D food models. Furthermore, to generate a larger dataset, data augmentation steps such as flipping, zooming in and out, brightening, changing contrast and coloring effects were applied to the output of simulator simulated images. YOLACT neural network was trained using the generated simulated dataset images.
For the robot demonstration, a pick-and-place test was conducted using different food object classes (real objects). The YOLACT neural network accurately recognized the objects, and the center of these objects was determined for subsequent object grasping points. Then, robot's joint values were then calculated to perform the object pick-and-place task. The results indicate that the YOLACT neural network quickly and successfully detects objects with an accuracy of up to 94.56%. This demonstrates the robot's proficiency in detecting and grasping objects, achieving high accuracy and successful pick-up rates across various object classes. In addition, our proposed YOLACT neural network succeeded in detecting the items in a group at a time.九州⼯業⼤学博⼠学位論⽂ 学位記番号：生工博甲第483号 学位授与年⽉⽇: 令和6年3⽉25⽇令和5年度doctoral thesi
画像分類と分割手法を用いた超小型衛星機上データの優先度付与の研究
Get PDF九州工業大学博士（工学）九州工業大学博士学位論文 学位記番号:工博甲第593号 学位授与年月日:令和6年9月25日令和6年度doctoral thesi
エラスターゼの効率的な近赤外検出とバイオイメージングを目指した新規スクアレイン-ペプチド複合体の研究
Get PDF九州工業大学博士（学術）1 Introduction| 2 Experimental| 3 Synthesis, BSA Interactions, In-vitro and In-vivo Studies of Newly Designed Squaraine Dyes| 4 Investigation of FRET-Based Squaraine-Peptide Conjugates for the Specific Detection of Neutrophil Elastase| 5 Synthesis, Characterization, and Evaluation of non-FRET-based Squaraine-Peptide Conjugate Probe| 6 General Conclusion and Future OutlookIn an era characterized by rapid lifestyle changes and environmental influences, the prevalence of diseases is on the rise. Chronic conditions such as diabetes, cardiovascular disorders, autoimmune/inflammatory diseases, and certain types of cancer have become increasingly prevalent, posing significant challenges to healthcare systems worldwide. Thus, the timely diagnosis and effective monitoring of diseases have emerged as the most crucial aspect of promoting public health and improving patient outcomes. In this context the existence and knowledge of Biomarkers are essential. In simple terms, biomarker refers to the presence or absence of a particular biomolecule in a specific disease. The identification and analysis of biomarkers have revolutionized the fields of disease diagnostics, disease monitoring, drug development, and personalized medicine. The growing prevalence of chronic diseases, along with the demand for more reliable, accurate, cost-efficient, and improved diagnostic solutions has propelled the market for the detection of disease biomarkers. Among various sensing methods, the optical method of sensing offers high selectivity and sensitivity, real-time monitoring, and more importantly activatable sensing. Since light is the main transducing element, optical imaging can be performed simultaneously, creating a visual representation providing crucial information on the local microenvironment. Among the various optical imaging techniques, the fluorescence mode of detection is non-invasive, easy, and cost-effective. The growing demand for reliable near-infrared (NIR) probes that maintain fluorescence in living systems and are easily compatible with biomolecules is fueled by the increasing use of NIR imaging in clinical diagnostics, owing to the enhanced spatial resolution and deeper tissue penetration in the NIR region. Five benzo[e]indole-based unsymmetrical squaraine dyes with -COOH functionalization and one without were synthesized and evaluated for their biocompatibility through BSA interactions, interactions with molecules of biological origin, in-vitro and in-vivo studies. All the dyes showed complete fluorescence quenching in aqueous solutions due to aggregation-induced quenching (AIQ). Dyes with -COOH functionalization showed strong fluorescence upon interaction with BSA, compared to the one without -COOH. Among these dyes, SQ-212 and SQ-215 showed distinct cell penetration behaviors, biocompatibility, non-toxic, and stable signal in-vitro and in-vivo, making them suitable for biomedical applications. Neutrophil Elastase (NE) and Neutrophil Extracellular Traps (NETs) are critical in immune responses. Neutrophils secrete NETs as an extracellular matrix to capture and neutralize pathogens like bacteria, fungi, and viruses. NETs are characterized by condensed DNA, Histones, NE, Cathepsin G, Myeloperoxidase, etc.
Excessive or prolonged NETs formation can contribute to Chronic inflammatory diseases and cancer. Recently, NETs have also been accounted for lung damage in COVID-19 patients. Hence, it is important to detect and monitor NETs. Neutrophil elastase (NE), is a serine protease expressed in NETs. Hence, the protease activity of NE is exploited as a fundamental mechanism for targeting NETs. Understanding NE and NETs interactions offers insights into disease development, leading to diagnostic and therapeutic advances. Foster Resonance Energy Transfer (FRET)-based squaraine-peptide conjugates Homo-APA and Hetero-APA containing the Ala-Pro-Ala as the NE recognition sequence were designed and synthesized. The probes were self-quenching (fluorescent-off) state through AIQ and FRET mechanisms. Upon incubating with NE, a tremendous increase in fluorescence intensity was seen, with Hetero-SQ-APA being better than Homo-APA. In-vitro and in-vivo studies in NETs-induced mice models showed good activity with NE and were able to detect NETs specifically. Intraperitoneal injection of Hetero-APA showed the detection of NETs and selective accumulation in the gout-induced mice model. The probes showed stable, strong, and intense ex-vivo signals while being non-toxic. However, no significant activity when tested in pathological samples. To further refine the NETs targeting probe, a non-FRET-based SQ-215-NETP incorporating the NE recognition sequence GEAIPMSIPPEVK that binds covalently to NE was designed and synthesized. When tested in-vitro with purified NE, and in-vivo in gout-induced mice SQ-215-NETP showed enhanced activity than Hetero-APA. SQ-215-NETP demonstrated high-resolution bioimaging capabilities across various ex-vivo samples, including human eye-derived NETs and human aortal thrombi aiding in detailed visualization of NET structures and NE activity. SQ-215-NETP was also able to detect NETs in flow cytometry setup as well. Thus this thesis presents the first small molecule-based probe for the detection and imaging of NETs in a wide range of samples having a great market potential in the diagnosis and monitoring of NET-induced diseases.九州工業大学博士学位論文 学位記番号:生工博甲第500号 学位授与年月日:令和6年9月25日令和6年度doctoral thesi
Effects of Anthropogenic Modification and Basin-Wide Specificities on Transverse Ribs in Steep Alluvial Fan River
Get PDF礫列は勾配1/100～1/20程度の扇状地上流にあたる急流河川に見られる河床形態である．この勾配帯では近年急速な河道掘削による人為改変が進行している．また，流域の特徴に対応して河道の様相が異なっており，潜在的な河川環境に寄り添う河道掘削工法の提案のためには，礫列の形成に対する人為改変の影響の程度と河川ごとに呈される特徴を考慮する必要がある．本研究では，急流扇状地河川における礫列への掘削の影響と流域ごとの特徴（個別性）を明らかにすることを目的とした．礫列とその形成を促進する可能性がある直径1m以上の巨石を異なる河川で観測し，環境要因を表すパラメータとの関係性を統計的に分析した．その結果，巨石の密度と礫列密度に正の相関が見られ河川ごとに傾向が異なること，河道掘削による巨石密度の低下が顕著であること，礫列に及ぼす要因は人為改変がない場合流域スケールの要因に強く影響を受ける一方，人為改変がある場合はより小さなスケールの要因が卓越することが分かった．これらより，巨石の移動・除去を抑制することが急流河川の潜在的環境形成に重要であることが示唆された．journal articl
Achieving high-Tc superconductivity in Magnéli phase based on Ti oxides: prediction by machine learning and material synthesis by high-pressure torsion processing
Get PDFWe explored superconductors with high superconducting transition temperatures (Tc) tuning the stability of Magnéli phase through high-pressure torsion (HPT). This study has started from exploring superconducting states in the Al–Mg–O ternary system along with the prediction using machine learning. We successfully found superconducting states with Tc = 4.0 and 7.3 K for a composition of Al:Ti = 1:2 in the mixture of Al and surface-oxidized Ti powders. Another magnetic anomaly was also observed at ~ 93 K, being supported by the Tc prediction using the machine learning for the Al–Ti–O system. In this study, the HPT processing was also performed on a Magnéli material Ti4O7 and such mixtures of stable materials as Al + TiO2, Al2O3 + Ti, Al2O3 + TiO2, and Al + Ti4O7. In HPT-processed Ti4O7, the metal–insulator transition was maintained even after the HPT processing. HPT experiments using stable oxides indicate the difficulty of newly stabilizing the Magnéli phase starting from thermodynamically stable materials under severe plastic deformation. A series of attempts reveals that the superconducting state in the mixture with ratio Al:Ti = 1:2 is attributed to both strained Ti-oxide created on the surface of the Ti powder and its reaction with Al under HPT processing, resulting in the stabilization of a Magnéli phase.journal articl
High pressure studies of the T-P phase diagrams of erbium and thulium up to 30 GPa by using ac magnetization experiments
Get PDFThe pressure dependence of the magnetic ordering temperatures for lanthanide ferromagnets Er and Tm has been investigated in the pressure region of up to 30 GPa through ac magnetization measurements by employing a superconducting quantum interference device. Six 4 lanthanide ferromagnets, from Gd to Tm, commonly have successive structural transitions starting from a hexagonal close-packed (hcp) structure under pressure. The experiments reported here have been performed with high sensitivity to allow for a more comprehensive discussion of the magnetostructural correlation when compared to four other 4 ferromagnets, namely Gd, Tb, Dy, and Ho. Our results conclude that only Er exhibits the increase in ferromagnetic ordering temperature against initial compression among the six lanthanide ferromagnets. The ferromagnetic magnetization anomaly for Er reduces below the detection limit before the structural transformation from Sm-type to the dhcp structures, and that for Tm exhibits similar behavior before the structure transformation from the hcp to Sm-type structures. However, the helimagnetic magnetization anomalies of both Er and Tm remain stable at least within the Sm-type structure below 30 GPa.journal articl
Initial bone tissue reactions of hydroxyapatite/collagen–(3-glycidoxypropyl)trimethoxysilane injectable bone paste
Get PDFWe have previously reported that a novel bioresorbable self-setting injectable bone paste composed of hydroxyapatite/collagen bone-like nanocomposite (HAp/Col) and (3-glycidoxypropyl)trimethoxysilane (GPTMS) was successfully prepared and was replaced with new bone within 3 months of implantation in defects created in porcine tibia. In this study, the HAp/Col-GPTMS paste was implanted into bone defects in rat tibiae to investigate the initial kinetics and bone tissue response. Even though more than 35% of GPTMS molecules should be eluted rapidly from directly injected pastes according to previously reported cell culture tests, in this study, energy-dispersive X-ray spectrometry did not detect Si (GPTMS) deposition in tissues surrounding the paste at 1 day postimplantation. Further, no abnormal inflammatory responses were observed in the surrounding tissues over the test period for both directly injected and prehardened pastes. Companying these observations with the results of the previous animal test (in which the paste was fully resorbed and was substituted with new bone), the eluted GPTMS resolved in no harm in vivo from the initial to final (completely resorbed) stages. Material resorption rates calculated from X-ray microcomputed tomography (μ-CT) images decreased with increasing in GPTMS concentration. Histological observations indicated that tartrate-resistant acid phosphatase (TRAP) active cells, (assumed to be osteoclasts), exist on the periphery of pastes. This result suggested that the paste was resorbed by osteoclasts in the same way as the HAp/Col. Since a good correlation was observed between TRAP active areas in histological sections and material resorption rate calculated from μ-CT, the TRAP activity coverage ratio offers the possibility to estimate the osteoclastic resorption ratio of materials, which are replaced with bone via bone remodeling process.journal articl
Robot Development and Biofeedback in the Medical and Welfare Field
Get PDF本稿は，日本における高齢化と介護人材不足の深刻化に対し，ロボット技術と人工知能（AI）の導入がもたらす影響と可能性について解説し，今後バイオフィードバックの概念が，医療・福祉分野でますます重要になってくることを指摘するものである．介護現場における技術活用の現状や課題，特に見守り・コミュニケーションロボットの効果が紹介され，職員の負担軽減や高齢者の生活支援における利点が論じられている．また，著者自身の経験をもとに，パーキンソン病患者に向けた歩行アシスト技術の開発も紹介されている．This article provides an analysis of the impact and potential of introducing robotics and artificial intelligence (AI) in response to the accelerating aging population and the growing shortage of caregiving personnel in Japan. It highlights the increasing importance of the biofeedback concept in the medical and welfare fields. The current state and challenges of utilizing technology in caregiving settings are discussed, with a focus on the effectiveness of monitoring and communication robots, emphasizing their benefits in reducing the burden on caregivers and supporting the daily lives of the elderly. Additionally, based on the author’s personal experience, the development of walking assistive technology for patients with Parkinson’s disease is introduced.journal articl