Lung Cancer
Conditions
Keywords
Lung cancer, Image, Treatment response, Prognosis, Radiomic, Deep learning
Brief summary
The investigators will evaluate the utility of computer aided image analysis in lung cancer with the aim of predicting treatment response and prognosis.
Detailed description
Tumor biological behavior is the fundamental cause of heterogeneous prognosis. The features found on medical images are also reflections of the tumor biological behavior. However, the limitations in spatial and intensity resolution of the naked eye are two inevitable shortcomings of image interpretation by naked eyes, resulting in subjective and limited analyses of images. Computer aided image analyses such as radiomic analysis and machine learning methods are emerging as promising image interpretation methods. The natural advantage of the unlimited spatial and intensity resolution of computers can overcome the shortcomings of visual inspection with the naked eye. Moreover, the massive computing power of computer is also far greater than that of humans. This study will focus on the application of computer aided analysis in predicting treatment response and prognosis in lung cancer.
Interventions
None listed
Sponsors
Study design
Eligibility
Inclusion criteria
1. Patients \>= 18 years old 2. Pathological diagnosis of NSCLC between 2012 and 2019; 3. PET/CT or CT examination before any cancer-specific treatment;
Exclusion criteria
1. A time interval between treatment and image examination greater than 1 month; 2. A history of other malignancies
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Overall survival | 2012-2021 | The interval between the date of diagnosis and death |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Progression free survival | 2012-2021 | The interval between the date of treatment initiation and disease progression |
Countries
China