As shown by Koenig and Zhu, the quotient of a triangulated category by a cluster tilting subcategory becomes an abelian category. On the other hand, as is well known since 1980s, the heart of any $t$-structure is abelian. In this talk, we make a simultaneous generalization of these two constructions. We consider a pair $(\mathcal{U},\mathcal{V})$ of subcategories with some conditions, which generalizes a $t$-structure and a cluster tilting subcategory. As a main theorem, we associate a subquotient abelian category to any such a pair. This gives back the heart and the quotient abelian category, respectively in the above two cases. Moreover, we can construct an essentially surjective cohomological functor from the original triangulated category to the subquotient abelian category. This functor generalize the canonical cohomology functor for a $t$-structure.
Information on the morphological changes in dental panoramic
radiograph is needed to detect various diseases, including
osteoporosis. Doctors often perform a cortical bone width measurement
manually to estimate the risk of certain bone diseases for a patient.
The greatest obstacle in the manual measurement is the different
measurement of the variable tendency of each doctor. Beside that, It
will take a long time process. One part is quite difficult to measure
automatically the width of cortical bone because of uneven
illumination and noise disturbance to the image.
This paper proposes a new system that can automatically measure the
width of cortical bone by using Active Shapes Model (ASM) method.
Measurements were taken at the boundary between the results of
fitting statistical models generated by the ASM method and the
object in the image. In this paper also discussed how the optimal
number of sample points to form a shape of the object of cortical
bone.
In experiment of 100 dental panoramic images obtained by the
optimal number of points in the training set is 50 points. The
correlation between measurement of this system and manual
measurement is 90%.
The inelastic nature and the resulting energy exchange in collisions of intrinsic localized modes, or discrete breathers, has been an open problem for quite a while. In this talk, a regularized PDE model for discrete breathers in Fermi-Pasta-Ulam chain will be proposed. Breather solutions are obtained via a variational approach, and their interactions are explored numerically. Then, based on the modified conservation law, a simplified ODE model is derived from the PDE model. The direct simulations for the ODE model achieve a good agreement with the numerical results in FPU lattice.
近年、質量分析（MS）とデータベース（DB）検索を組 み合わせたプロテオーム解析により、細胞や組織で発現している数 千種類のタンパク質を同定することが可能と なった。そこでは通常、質量分析で得られるペプチドの部分配列情 報を用い、タンパク質アミノ酸配列のDBに対して確率統計的 な考えに基づく検索を行うことで、試料 中のタンパク質の同定を行っている。多くのモデル生物でゲノム DNA配列が明らかになることでDBの整備が進み、ゲノムDNA とそれらがコードする多くのタンパク質との 対応が明らかになることで、この手法が生まれた。本セミナーで は、情報技術を用いたプロテオーム解析の仕組みについて概説する とともに、光合成細菌や分裂酵母を 用いた実際のプロテオーム解析の例を示す。また、我々は、ゲノム DNAの塩基配列DBに対しての発現同定に簡単な遺伝子予測プロ グラムを組み合わせることで、遺伝子ア ノテーションの間違いを検出したり、新規遺伝子を予測したりする ことを可能にしたので、その研究について解説する。