公开(公告)号：US20240225588A1

公开(公告)日：2024-07-11

申请号：US18615351

申请日：2024-03-25

Applicant: Hsiao-Ching Nien ,  YUAN High-Tech Development Co., Ltd.

Inventor： Hsiao-Ching Nien ,  Ta-Hsiang Lin ,  Pei-Lien Chou

IPC: A61B8/08 ,  A61B8/00 ,  G06T7/00 ,  G06V10/764 ,  G06V10/774 ,  G06V20/70 ,  G16H15/00 ,  G16H30/40 ,  G16H50/30

CPC classification number: A61B8/085 ,  A61B8/4427 ,  A61B8/463 ,  A61B8/54 ,  G06T7/0014 ,  G06V10/764 ,  G06V10/774 ,  G06V20/70 ,  G16H15/00 ,  G16H30/40 ,  G16H50/30 ,  G06T2207/10132 ,  G06T2207/20081 ,  G06T2207/30056 ,  G06T2207/30096 ,  G06V2201/031

Abstract: Provided is a method and apparatus of intelligent analysis for liver tumors, including an analysis module for receiving YOLOR-based training to acquire sufficient intelligence to detect and locate liver tumor automatically and attain a mAP score as high as 0.56 required to distinguish lesions of benignant and malignant liver tumors in medical images from each other, attaining a mAP score of 0.628 for tumors at least 5 cm in size or a mAP score of 0.33 for tumors less than 5 cm in size. Thus, the area under the liver tumor differentiation curve of the analysis module and the mAP score reach 0.9 and 0.56 respectively. The values equal those of the effect of the diagnosis rate of liver tumors with CT and MRI in practice. The method is advantageous in terms of higher speed and thus can diagnose liver tumors earlier, preclude delays and radiation, but incur low cost.

公开(公告)号：US20210153838A1

公开(公告)日：2021-05-27

申请号：US16952238

申请日：2020-11-19

Applicant: Hsiao-Ching Nien ,  YUAN High-Tech Development Co., Ltd.

Inventor： Hsiao-Ching Nien ,  Ta-Hsiang Lin ,  Pei-Lien Chou

IPC: A61B8/08 ,  A61B8/14 ,  A61B8/00 ,  G16H50/30 ,  G06N20/00

Abstract: A method of Intelligent Analysis is provided for liver tumor. It is a method using scanning acoustic tomography (SAT) with a deep learning algorithm for determining the risk of malignance for liver tumor. The method uses the abundant experiences of abdominal ultrasound specialists as a base to mark pixel areas of liver tumors in ultrasound images. The parameters and coefficients of empirical data are trained with the deep learning algorithm to establish a categorizer model reaching an accuracy rate up to 86 percent. Thus, with an SAT image, a help to doctor or ultrasound technician is obtained to determine the risk of malignance for liver tumor through the method, and to further provide a reference base for diagnosing liver tumor category.