CN-121983288-A - Autoimmune pancreatitis auxiliary diagnosis system based on multi-mode medical image

CN121983288ACN 121983288 ACN121983288 ACN 121983288ACN-121983288-A

Abstract

The invention discloses an autoimmune pancreatitis auxiliary diagnosis system based on multi-mode medical images, and relates to the field of medical images. The method comprises a multi-temporal image processing module, a lesion evolution track construction module and the like. The multi-temporal image processing module is used for automatically dividing, aligning and standardizing multi-mode medical images at a plurality of time points in a cross-time point mode to obtain an image sequence with consistent time sequence, the lesion evolution track construction module is used for extracting equivalent characteristics of pancreas morphology, main pancreas duct morphology, signals or metabolic parameters at each time point and constructing a lesion evolution track, the track comparison analysis module is used for comparing the evolution track of a tested person with a pre-constructed typical evolution track set to generate a similarity analysis result, the state updating module is used for fusing the characteristics of the newly added time point, maintaining and updating the state of illness state, and the auxiliary diagnosis module is used for outputting a state of illness trend prompt and a follow-up risk prompt based on the similarity analysis result and the state of illness state representation.

Inventors

ZHAI YAQI
CHEN SHENGXIN
LI MINGYANG

Assignees

中国人民解放军总医院第一医学中心

Dates

Publication Date: 20260505
Application Date: 20260206
Priority Date: 20251127

Claims (10)

1. An autoimmune pancreatitis auxiliary diagnosis system based on multi-modal medical images, characterized by comprising, The multi-temporal image processing module is used for acquiring multi-modal medical images acquired by a subject at a plurality of time points, and executing automatic segmentation, cross-time-point spatial alignment and modal standardization processing on pancreas regions and related structures on the multi-modal medical images to obtain an image sequence with consistent time sequence; the lesion evolution track construction module is used for extracting pancreatic morphology, main pancreatic duct morphology, membranous change, image signals or metabolic parameter equalization characteristics at each time point based on the image sequences with consistent time sequences, and constructing a multidimensional evolution track reflecting the change of lesions along with time according to time sequences; the track control analysis module is used for carrying out control analysis on the multidimensional evolution track reflecting the change of the lesions along with the time and a pre-constructed typical lesion evolution track set to generate analysis results representing the similarity degree of the tracks of the detected person and different lesion types; the state updating module is used for fusing the track characteristics of the time point corresponding to the image with the existing state of illness evolution state representation to form updated state of illness evolution state representation during follow-up review of the testee; The auxiliary diagnosis module is used for outputting auxiliary information for judging the diagnosis tendency of the autoimmune pancreatitis and prompting the follow-up risk based on the result of the track control analysis module and the updated state of illness evolution.
2. The multi-modality medical image-based autoimmune pancreatitis auxiliary diagnostic system of claim 1, wherein the multi-modality medical image comprises at least two of CT, MRI, DWI, MRCP or PET-CT, the multi-temporal image processing module being specifically configured to: carrying out automatic segmentation on pancreas body, main pancreas duct and peripheral membranous structures on images at different time points; aligning the images of each time point to a unified reference coordinate system through non-rigid body registration, so that the anatomical corresponding relation is kept consistent in the time dimension; And carrying out unified scale standardization processing on signals or density values among different modes.
3. The multi-modality medical image based autoimmune pancreatitis auxiliary diagnostic system of claim 1, wherein the lesion evolution trace construction module is configured to extract at each point in time at least one of the following features: pancreatic morphology features including volume, swelling extent, membranous thickness and continuity; main pancreatic duct features including stenosis length, mean stenosis diameter and degree of upstream dilation; signal or density characteristics, including CT values, MRI T1/T2 signal intensity, or DWI/ADC values; Metabolic characteristics, including PET uptake value, metabolic volume, or total metabolic amount.
4. The multi-modal medical image-based auxiliary diagnosis system for autoimmune pancreatitis according to claim 1, wherein the lesion evolution track construction module further determines the evolution trend of the lesion according to the characteristic change condition of the adjacent time points, including whether the lesion is continuously improved, tends to be stable after the improvement, is re-emphasized after the stabilization, or is intermittently fluctuated, and takes the evolution trend as a part of the evolution track.
5. The multi-modality medical image-based auxiliary diagnosis system for autoimmune pancreatitis of claim 1, wherein the trajectory control analysis module comprises a typical lesion evolution trajectory set and analysis unit: Wherein the typical lesion evolution track set is formed by track construction and pattern clustering of past cases with long-term follow-up information, and comprises track types which are represented as continuous improvement type, delay improvement type, residual lesion stability type or re-weighting type; the analysis unit is used for comparing the lesion evolution track of the detected person with the typical lesion evolution track set and generating a similarity analysis result.
6. The multi-modal medical image-based auxiliary diagnosis system for autoimmune pancreatitis according to claim 5, wherein the analysis unit analyzes differences between the subject trajectory and each typical trajectory by comprehensively considering the magnitude, the direction and the rhythm of the variation of the eigenvalue, and generates a similarity analysis result for distinguishing between different lesion evolution modes.
7. The multimodal medical image-based autoimmune pancreatitis auxiliary diagnostic system of claim 1, wherein the status update module is to: Maintaining a condition status representation for each subject reflecting past lesion changes; Fusing the lesion characteristics of the newly added time point with the state of illness representation during each review; the recent data is given a higher weight during the fusion process to highlight the impact of recent disease changes on diagnostic trend determinations.
8. The multi-modality medical image-based auxiliary diagnostic system of autoimmune pancreatitis of claim 1, wherein the lesion evolution trajectory construction module further aligns treatment-related information of the subject with the image time axis, the treatment-related information including a treatment dose, a dose change condition, a maintenance treatment time, and a drug withdrawal time point, further generates a treatment response index based on a correspondence between a treatment plan and an image change, and incorporates the index into the lesion evolution trajectory.
9. The auxiliary diagnosis system for autoimmune pancreatitis based on multi-modal medical images as in claim 1, wherein the auxiliary diagnosis module is configured to output based on the disease state representation and the similarity analysis result: An analytical cue indicating that the condition is more prone to improvement, stabilization or exacerbation; Representative track types representing the proximity of a subject's lesion track; And generating a follow-up frequency or review time interval assistance advice based on the analysis cues.
10. An autoimmune pancreatitis auxiliary diagnosis method based on multi-modal medical images is characterized by comprising the following steps: acquiring multi-mode medical images of a subject at a plurality of time points, and performing pancreas and related structure segmentation, cross-time point alignment and modal standardization processing on the multi-mode medical images; Extracting relevant characteristics of the lesions at each time point and constructing a lesion evolution track; performing control analysis on the lesion evolution track and a preset typical lesion evolution track set to form a similarity analysis result; updating the state of illness representation of the testee according to the track information of the newly added time point; and outputting prompt information for auxiliary diagnosis based on the disease state representation and the similarity analysis result.

Description

Autoimmune pancreatitis auxiliary diagnosis system based on multi-mode medical image Technical Field The invention relates to the field of medical images, in particular to an autoimmune pancreatitis auxiliary diagnosis system based on multi-mode medical images. Background Autoimmune pancreatitis (AIP) is an immune-related pancreatitis disease characterized by diffuse or focal enlargement of pancreas, membranous change of pancreas, stenosis of main pancreatic duct, etc., and clinically depends on CT, MRI, DWI, MRCP and multi-mode images such as PET-CT for comprehensive judgment. AIP has long-term fluctuation characteristic in the course of disease, patients can generally obtain better remission after primary treatment, but the AIP still has higher recurrence risk in follow-up visit, so that the AIP has important significance in longitudinal analysis of multi-temporal images. However, the existing image diagnosis mode is mainly based on qualitative observation at a single time point, lacks quantitative description of pancreas structural change, main pancreas duct form change and metabolic index change, and also lacks systematic processing and comparison mechanisms of relationships between multi-mode and multi-time point images, so that the overall trend of lesion evolution along with time cannot be effectively reflected. In addition, a technical system for extracting a lesion evolution track from long-term follow-up images of a patient and performing control analysis with a typical evolution mode is not formed in the prior art, so that auxiliary judgment on trends such as disease improvement, stability or exacerbation is difficult. For this reason, an autoimmune pancreatitis auxiliary diagnosis system based on multimodal medical images is proposed. Disclosure of Invention The invention mainly aims to provide an autoimmune pancreatitis auxiliary diagnosis system based on multi-mode medical images, which can effectively solve the problems in the background technology. In order to achieve the above purpose, the invention adopts the technical proposal that, An autoimmune pancreatitis auxiliary diagnostic system based on multimodal medical imaging, comprising, The multi-temporal image processing module is used for acquiring multi-modal medical images acquired by a subject at a plurality of time points, and executing automatic segmentation, cross-time-point spatial alignment and modal standardization processing on pancreas regions and related structures on the multi-modal medical images to obtain an image sequence with consistent time sequence; the lesion evolution track construction module is used for extracting pancreatic morphology, main pancreatic duct morphology, membranous change, image signals or metabolic parameter equalization characteristics at each time point based on the image sequences with consistent time sequences, and constructing a multidimensional evolution track reflecting the change of lesions along with time according to time sequences; the track control analysis module is used for carrying out control analysis on the multidimensional evolution track reflecting the change of the lesions along with the time and a pre-constructed typical lesion evolution track set to generate analysis results representing the similarity degree of the tracks of the detected person and different lesion types; the state updating module is used for fusing the track characteristics of the time point corresponding to the image with the existing state of illness evolution state representation to form updated state of illness evolution state representation during follow-up review of the testee; The auxiliary diagnosis module is used for outputting auxiliary information for judging the diagnosis tendency of the autoimmune pancreatitis and prompting the follow-up risk based on the result of the track control analysis module and the updated state of illness evolution. Further, the multi-modal medical image comprises at least two of CT, MRI, DWI, MRCP or PET-CT, and the multi-temporal image processing module is specifically configured to: carrying out automatic segmentation on pancreas body, main pancreas duct and peripheral membranous structures on images at different time points; aligning the images of each time point to a unified reference coordinate system through non-rigid body registration, so that the anatomical corresponding relation is kept consistent in the time dimension; And carrying out unified scale standardization processing on signals or density values among different modes. Further, the lesion evolution track construction module is configured to extract at least one of the following features at each time point: pancreatic morphology features including volume, swelling extent, membranous thickness and continuity; main pancreatic duct features including stenosis length, mean stenosis diameter and degree of upstream dilation; signal or density characteristics, including CT values, MRI T1/T2 signal intensity, or