Invention Grant
- Patent Title: Structured representations for interpretable machine learning applications in medical imaging
-
Application No.: US18504769Application Date: 2023-11-08
-
Publication No.: US12154321B2Publication Date: 2024-11-26
- Inventor: Matthew Phillips , Andrew J. Buckler
- Applicant: ELUCID BIOIMAGING INC.
- Applicant Address: US MA Boston
- Assignee: ELUCID BIOIMAGING INC.
- Current Assignee: ELUCID BIOIMAGING INC.
- Current Assignee Address: US MA Boston
- Agency: PEARL COHEN ZEDEK LATZER BARATZ LLP
- Main IPC: G06V10/774
- IPC: G06V10/774 ; A61B5/024 ; G06T7/00 ; G06T7/73 ; G06V20/70 ; G16H50/50

Abstract:
Systems and method can be provided to transform input data (e.g., CT imaging data) into structured representations to create interpretable models. Another aspect of the current invention can be generating labels synthetically to apply to real data according to a biologically-based labelling technique to guide the model training with a priori mechanistic knowledge.
Public/Granted literature
- US20240153252A1 STRUCTURED REPRESENTATIONS FOR INTERPRETABLE MACHINE LEARNING APPLICATIONS IN MEDICAL IMAGING Public/Granted day:2024-05-09
Information query