Dissertation > Excellent graduate degree dissertation topics show

Texture Features Extraction of Chest HRCT Image Based on Granular Computing

Author: CaoTianRui
Tutor: XieGang
School: Taiyuan University of Technology
Course: Control Theory and Control Engineering
Keywords: HRCT Texture features Medical Image Segmentation Tolerance Granular Space Model Small airway disease
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 50
Quote: 1
Read: Download Dissertation

Abstract


Chronic obstructive pulmonary disease (COPD) is a group of chronic lung disease, chronic bronchitis, emphysema. According to statistics, China's COPD patients is about 32 million people, about 100 million people a year die of the disease. Small airway disease COPD early and reversible lesions, is also the main cause airflow obstruction of COPD and its timely diagnosis and treatment to prevent its development to become an important means of COPD. Multislice CT machine chest high-resolution CT (HRCT) images, is the observation of the best images of the small airway disease. Structural organization, a large range of gray distribution chest HRCT images at the same time provide a more detailed and more accurate diagnostic information, also contributed to the difficult problem of texture analysis and tissue segmentation. Currently, most of the radiologist with experience subjective judgments, mainly in the qualitative analysis of the lesion, you want to make an objective and accurate analysis very difficult, and the extent of the lesion that is less quantitative research. The chest HRCT image quantitative analysis, to help doctors diagnose small airway disease, must solve the problem is the chest HRCT texture eigenvalue extraction and lungs accurate segmentation of soft tissue. With the ongoing research of many scholars of image engineering, genetic algorithm, fuzzy sets, granular computing as the representative of the intelligent control theory is being used in medical image processing, and to provide new ideas and methods to solve practical problems. Similarly, granular computing theory also try to extract and lung tissue segmentation applied to the HRCT image texture characteristics value. Chest HRCT image as the research object, grain is calculated as the theoretical basis for the quantitative analysis of the status quo at home and abroad in the field of research objectives, extensive research, the value extracted texture features by combining the knowledge of the anatomy of the human tissue, lung tissue divided methods carried out in-depth research to complete the following work: first analyzed the HRCT image characteristics, small airway disease the HRCT signs and related texture parameters, briefly compatibility granularity space model of granular computing, based on chest HRCT image The relationship between features and compatibility granularity space model, constructed chest HRCT images space model. Texture analysis of medical images and image segmentation methods and research, texture feature extraction is introduced into the HRCT image analysis has laid a foundation for quantitative HRCT images provide effective analysis methods and data for accurate diagnosis of small airway disease ; for the complex texture of lung tissue, based on the traditional medical image segmentation method, combined with the theory of granular computing based regional Tolerance Granular Space Model growing segmentation algorithm, average gray texture features of regions of interest (ROI) degrees (mean) value automatically select seed points to improve the original region growing algorithm method of manually selected seed point, and do not need to repeatedly adjust the threshold parameter; growth improved compatibility relations (TR) criteria HRCT image can be applied. Finally, a large number of experiments and the results of evaluation to achieve the texture parameters of any straight line and region of interest (ROI) extraction, lungs accurate segmentation of soft tissue and area characteristics. Experiments show that, this article can effectively extract the characteristic parameters of the relevant texture than strong relevance and practicality of the classic texture analysis method, to obtain the data required for the diagnosis of small airway disease, provides for the radiologist to accurately diagnose the illness strong data protection.

Related Dissertations

  1. The Value of High Resolution CT and Enhanced Spiral CT in the Diagnosis of Solitary Pulmonary Nodule,R816.4
  2. Research of HRCT and Pulmonary Function Test in COPD,R563.9
  3. The Preoperative Diagnosis of Tympanosclerosis,R764
  4. The Comparative Study of High-Kilovolt Radiography and High Resolution CT in Diagnosis of Pneumoconsis,R135.2
  5. Whistar rat femur three-dimensional reconstruction of the growth phase,TP391.41
  6. The Clinical Significance of PET/CT Imaging in Differential Diagnosis of Solitary Pulmonary Nodules,R734.2
  7. Study on the Remote Sensing Monitoring of Wetland in the Dongting Lake Using ENVISAT ASAR Data,X87
  8. Chest HRCT Image Segmentation Based on Granularity Synthesis,TP391.41
  9. Optic Canal Fracture CT 3D Reconstruction Research and Clinical Application Value,R816.97
  10. Guidance of Temporal Bone Thin Slice Scanning by 256-ranked Spiral CT in Tympanoplasty,R764.21
  11. The Correlation Research to TGF-β1, FGF-1, Pulmonary Function and High Resolution Computed Tomography Score of COPD with Pulmonary Fibrosis,R563.9
  12. Labyrinth path down into the internal auditory canal and petrous apex area of microscopic anatomy of surgery,R322
  13. The Value of HRCT Applying Calcaneus Bone of Type 2 Diabetes Patients without Insulin Therapy-Morphology Study,R587.1
  14. Extraction Method of LungTexture Tree from HRCT Images,TP391.41
  15. Study of Inner Ear with Volume Rendering Technique Using Multi-slice Spiral CT and Clinical Analysis for Congenital Sensorineural Hearing Loss in Infants,R764
  16. Analysis on the Clinical Characteristics of 50 Cases with Interstitial Lung Disease,R563.9
  17. Airway wall thickness with the disease, the relationship between lung function and inflammatory markers,R562.25
  18. Assessment Airway Remodeling in Asthma by Use of High-resolution CT, Biological Makers and Pulmonary Function,R562.25
  19. Applied Study of Coronal High-Resolution Computed Temography Image and Thin Coronal Section Specimen in the Middle Ear and Nearly Structures of Temporal Bone,R816.96
  20. The Method Research of Pattern Recognition about Satellite Remote Sensing Vehicle Based on Morphological,TP75

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile