Computer-aided Diagnosis in Chest Images Based on Machine Learning
Principal Investigator
Name
wanle chi
Degrees
Ph.D.
Institution
WenZhou PolyTechnic
Position Title
Assistant Professor
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-957
Initial CDAS Request Approval
Sep 6, 2022
Title
Computer-aided Diagnosis in Chest Images Based on Machine Learning
Summary
Recently, the deep cross-fusion applications between artificial intelligence (AI) and medical imaging have attracted wide attention of many researchers with the continuous growth of massive image data, continuously iterative update of intelligent diagnostic model algorithms, substantial improvement of computing power and strong support of national policies. In the academia, artificial intelligence, especially deep learning, has made considerable progress on many long-term problems such as lesion detection, segmentation and diagnosis. The research on medical image-assisted diagnosis shows a booming development. In the industry, AI medical imaging is rapidly transitioning from the experimental stage to the clinical trial stage with the development advantages of medical image big data and deep learning technology. AI medical imaging has become one the most closely integrated fields of artificial intelligence and medical industry.
This dissertation focuses on the problems of imbalanced data distribution and high misdiagnosis rate in the chest imaging- assisted diagnosis scenarios. A lesion often presents many signs in the chest images. Generally, doctors can capture the signs to make judgment of lesions. But, the doctors are insufficient which leads to misdiagnosis. This study aims to aid the diagnosis process and reduce the misdiagnosis by providing a computer aided diagnosis framework in chest images based on machine learning technology.
We carry out researches on generative adversarial learning- based data synthesis over-sampling and transfer learning . The main research work and contribution of this dissertation include the following parts.
1. Two-steps-GAN-based synthetic oversampling method for semantic classification in chest images.
2. Auxiliary diagnosis of benign and malignant lung lesion based on improved soft parameter sharing deep learning of transfer learning.
To solve the medical problems in the small-sample field is more important for the development of artificial intelligence in medical imaging, because the cost of the collection and labeling of these medical data is very high and small-sample scenarios often occupy most of the actual scenes. The methods of the generative adversarial synthesis learning and semantic task association transfer are the core to solve the problems of small data. The improvement of the accuracy and efficiency in medical image-assisted diagnosis plays an important role and practical significance in alleviating medical resources, promoting graded diagnostic treatment services and promoting telemedicine services.
This dissertation focuses on the problems of imbalanced data distribution and high misdiagnosis rate in the chest imaging- assisted diagnosis scenarios. A lesion often presents many signs in the chest images. Generally, doctors can capture the signs to make judgment of lesions. But, the doctors are insufficient which leads to misdiagnosis. This study aims to aid the diagnosis process and reduce the misdiagnosis by providing a computer aided diagnosis framework in chest images based on machine learning technology.
We carry out researches on generative adversarial learning- based data synthesis over-sampling and transfer learning . The main research work and contribution of this dissertation include the following parts.
1. Two-steps-GAN-based synthetic oversampling method for semantic classification in chest images.
2. Auxiliary diagnosis of benign and malignant lung lesion based on improved soft parameter sharing deep learning of transfer learning.
To solve the medical problems in the small-sample field is more important for the development of artificial intelligence in medical imaging, because the cost of the collection and labeling of these medical data is very high and small-sample scenarios often occupy most of the actual scenes. The methods of the generative adversarial synthesis learning and semantic task association transfer are the core to solve the problems of small data. The improvement of the accuracy and efficiency in medical image-assisted diagnosis plays an important role and practical significance in alleviating medical resources, promoting graded diagnostic treatment services and promoting telemedicine services.
Aims
We carry out researches on generative adversarial learning- based data synthesis over-sampling and transfer learning . The main research work and contribution of this dissertation include the following parts.
1. Two-steps-GAN-based synthetic oversampling method for semantic classification in chest images.
2. Auxiliary diagnosis of benign and malignant lung lesion based on improved soft parameter sharing deep learning of transfer learning.
Collaborators
Yun Huoy Choo, Ong Sing Gho