MULTI-CHANNEL ORTHOGONAL CONVOLUTIONAL NEURAL NETWORKS

To provide multi-channel orthogonal convolutional neural networks.SOLUTION: Exemplary embodiments of the present disclosure include apparatus and methods to classify plaque tissue present in the coronary artery using intravascular optical coherence tomography (IVOCT) images. There is provided a meth...

Full description

Saved in:
Bibliographic Details
Main Authors NITESH KATTA, AYDIN ZAHEDIVASH, AUSTIN MCELROY, THOMAS E MILNER
Format Patent
LanguageEnglish
Japanese
Published 21.02.2024
Subjects
CALCULATING
COMPUTING
COUNTING
DIAGNOSIS
HUMAN NECESSITIES
HYGIENE
IDENTIFICATION
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
MEDICAL OR VETERINARY SCIENCE
PHYSICS
SURGERY
Online AccessGet full text

Cover

More Information
Summary:To provide multi-channel orthogonal convolutional neural networks.SOLUTION: Exemplary embodiments of the present disclosure include apparatus and methods to classify plaque tissue present in the coronary artery using intravascular optical coherence tomography (IVOCT) images. There is provided a method of improving discrimination between superficial lipid and calcium versus fibrous tissue and lipid, calcium tissue, and connective tissue, the method comprising: acquiring an image, where the image is a signal constructed from a point or linear array transducer; analyzing how the image was constructed and mathematically extracting derived data from the image; and applying an orthogonal convolutional neural network (OCNN) to optimize consideration of the derived data.SELECTED DRAWING: Figure 1 【課題】マルチチャネル直交畳み込みニューラルネットワークの提供。【解決手段】本開示の例示的実施形態は、血管内光コヒーレンス断層撮影（ＩＶＯＣＴ）画像を使用して、冠動脈内に存在するプラーク組織を分類するための装置および方法を含む。表在脂質およびカルシウム対線維組織および脂質、カルシウム組織、および結合組織間の判別を改良する方法が提供され、該方法は、画像を入手することであって、該画像は、点または線形アレイトランスデューサから構築された信号である、ことと、該画像が構築された方法を分析し、該画像から導出データを数学的に抽出することと、直交畳み込みニューラルネットワーク（ＯＣＮＮ）を適用し、該導出データの考慮点を最適化することとを含む。【選択図】図１
Bibliography:Application Number: JP20230217939