Computer-Aided Treatment Effectiveness Assessment Based on Lung Cancer CT Screening
Principal Investigator
Name
Hsiao-Dong Chiang
Degrees
Ph.D
Institution
Cornell University
Position Title
Professor
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-346
Initial CDAS Request Approval
Sep 26, 2017
Title
Computer-Aided Treatment Effectiveness Assessment Based on Lung Cancer CT Screening
Summary
Lung cancer accounts for the highest number of cancer-related deaths, more than any other cancer. Evaluation of whether a treatment is effective or not may take months and miss the best opportunity to treat the cancer with an effective approach. In recent years, the advent of deep learning has emerged as a powerful alternative for pattern recognition by using deep neural networks that can learn a representation of data from the raw data itself.
In this project, we will address the problem of treatment effectiveness assessment. To identify lung nodules and predict the progression of cancer in evaluating the effectiveness of a treatment, we plan to train our model with tagged data such as LIDC-IDRI and part of the Spiral CT Screening dataset for a sensitive system. The progression and treatment datasets would also be necessary in training and testing our system.
In this project, we will address the problem of treatment effectiveness assessment. To identify lung nodules and predict the progression of cancer in evaluating the effectiveness of a treatment, we plan to train our model with tagged data such as LIDC-IDRI and part of the Spiral CT Screening dataset for a sensitive system. The progression and treatment datasets would also be necessary in training and testing our system.
Aims
Build an assessment system that can:
Detect and identify lung nodules from CT scans.
Predict the types of lung tumors.
Assess the effectiveness of a treatment.
Offer references to physicians with identified nodules and their assessments.
Reduce the time in deciding the effectiveness of a treatment.
Collaborators
Pumiao Yan, Cornell University