DEEP LEARNING-BASED PANCREATIC CANCER VASCULAR INVASION CLASSIFICATION METHOD AND ANALYSIS DEVICE USING ENDOSCOPIC ULTRASOUND IMAGE

• Main Authors: SEO, Jeong Wung, PARK, Suhyun, SEO, Kang Won
• Format: Patent
• Language: English; French; Korean
• Published: 04.05.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; DIAGNOSIS; HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATIONTECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING ORPROCESSING OF MEDICAL OR HEALTHCARE DATA; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; MEDICAL OR VETERINARY SCIENCE; PHYSICS; SURGERY

The deep learning-based pancreatic cancer vascular invasion classification method using an endoscopic ultrasound image comprises steps in which an analysis device: receives a pancreatic endoscopic ultrasound image of a subject; inputs the pancreatic endoscopic ultrasound image in a first segmentation model so as to separate out a pancreatic cancer region; inputs the pancreatic endoscopic ultrasound image in a classification model so as to determine whether the pancreatic cancer has invaded blood vessels; and inputs the pancreatic endoscopic ultrasound image in a second segmentation model so as to separate out a blood vessel region located in the vicinity of the pancreatic cancer region. Le procédé de classification d'invasion vasculaire du cancer du pancréas basé sur l'apprentissage profond utilisant une image échographique endoscopique comprend des étapes lors desquelles un dispositif d'analyse : reçoit une image échographique endoscopique pancréatique d'un sujet ; saisit l'image échographique endoscopique pancréatique dans un premier modèle de segmentation de façon à séparer une région du cancer du pancréas ; saisit l'image échographique endoscopique pancréatique dans un modèle de classification de façon à déterminer si le cancer du pancréas s'est propagé dans des vaisseaux sanguins ; et saisit l'image échographique endoscopique pancréatique dans un second modèle de segmentation de façon à séparer une région de vaisseau sanguin située à proximité de la région du cancer du pancréas. 초음파 내시경 영상을 이용한 딥러닝 기반 췌장암의 혈관 침범 분류 방법은 분석장치가 대상자의 췌장 초음파 내시경 영상을 입력받는 단계, 상기 분석장치가 상기 췌장 초음파 내시경 영상을 제1 세그멘테이션 모델에 입력하여 췌장암 영역을 구분하는 단계, 상기 분석장치가 상기 췌장 초음파 내시경 영상을 분류 모델에 입력하여 췌장암의 혈관 침범 여부를 판단하는 단계 및 상기 분석장치가 상기 췌장 초음파 내시경 영상을 제2 세그멘테이션 모델에 입력하여 상기 췌장암 영역의 주변에 위치한 혈관 영역을 구분하는 단계를 포함한다.