Algoritm segmentacii OKT izobrazhenij makuly dlya analiza pacientov s vozrastnoj makulyarnoj degeneraciej

نویسندگان

چکیده

Odnoj iz osnovnyh prichin slepoty i slabovideniya u lic starshego trudosposobnogo vozrasta yavlyaetsya vozrastnaya makulyarnaya degeneraciya (VMD), dlya diagnostiki kotoroj krajne vazhny rezul'taty opticheskoj kogerentnoj tomografii (OKT). Sozdanie rekomendatel'noj sistemy analiza OKT-snimkov pozvolit sokratit' vremya na obrabotku vizual'noj informacii snizit' veroyatnost' oshibok v processe raboty vracha. Cel'yu issledovaniya bylo razrabotat' algoritm segmentacii dannyh OKT makuly pacientov s VMD, pozvolyayushchij, osnovyvayas' forme vydelennyh patologij, korrektno predskazyvat' stadiyu razvitiya VMD. Razrabotana programma yazyke programmirovaniya Python ispol'zovaniem biblioteki PyTorch TensorFlow. Kachestvo programmy ocenili OKT-izobrazheniyah 51 pacienta VMD rannej, promezhutochnoj pozdnej stadij. Razrabotan OKT-snimkov, osnovannyj svertochnoj nejronnoj seti. V kachestve arhitektury seti byla vybrana set' UNet. Nejronnaya obuchena OKT-snimkah 125 (197 glaz). Avtorskij prodemonstriroval 98,1% verno segmentirovannyh oblastej OKT-snimkah, naibolee vazhnyh opredeleniya stadii Vzveshennaya chuvstvitel'nost' specifichnost' klassifikatora stadij sostavili sootvetstvenno 83,8% 84,9%. Razrabotannyj perspektiven sistemy, realizuyushchej klassifikaciyu osnove dannyh, sposobstvuyushchej prinyatiyu reshenij o taktike lecheniya pacientov.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

FAST: A Functional Algoritm Simulation Testbed

In this paper we extend the practical range of simulations of parallel executions by what we call \func-tional algorithm simulation," that is, simulation without actually performing most of the numerical computations involved. We achieve this by introducing a new approach for generating and collecting communication and computation characteristics for a class of parallel scientiic algorithms. We...

متن کامل

A Parsing Algoritm That Extends Phrases

It is desirable for a parser to be able to extend a phrase even after it has been combined into a larger syntactic unit. This paper presents an algorithm that does this in two ways, one dealing with "right extension" and the other with "left recursion". A brief comparison with other parsing algorithms shows it to be related to the left corner parsing algorithm, but it is more flexible in the or...

متن کامل

Convergence Properties of the BFGS Algoritm

The BFGS method is one of the most famous quasi-Newton algorithms for unconstrained optimization. In 1984, Powell presented an example of a function of two variables that shows that the Polak–Ribière–Polyak (PRP) conjugate gradient method and the BFGS quasi-Newton method may cycle around eight nonstationary points if each line search picks a local minimum that provides a reduction in the object...

متن کامل

Detekcija in analiza dodatnih harmonskih komponent samodržnega vrtilnega momenta sinhronskega motorja s trajnimi magneti

Povzetek. V članku je predstavljena problematika nastanka dodatnih harmonskih komponent v samodržnem vrtilnem momentu sinhronskega motorja s trajnimi magneti (SMTM). Dodatne harmonske komponente v nekaterih primerih zmanjšujejo učinkovitost poznanih konstrukcijskih metod za znižanje samodržnega vrtilnega momenta. Vzroki za nastanek dodatnih harmonskih komponent samodržnega vrtilnega momenta so ...

متن کامل

Analiza bezbednosnih mehanizama OSPF protokola

Service or system security depends on the security of any component used on that system. Computer network attacks can jeopardize normal network functionality. There are cases where an attacker can gain unauthorized control over classified data. OSPF is the most common link state routing protocol. In this paper, we have analyzed OSPF security issues and described some protection methods. Analysi...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Vestnik RGMU

سال: 2022

ISSN: ['2070-7320', '2070-7339']

DOI: https://doi.org/10.24075/vrgmu.2022.062