Skip to Main Content

An official website of the United States government

Predicting Lung Nodule Growth Using Deep Learning Methods

Principal Investigator
Name
gao dong chen
Institution
Shan Dong University
Position Title
teaching assistant
Email
About this CDAS Project
Study
NLST (Learn more about this study)
Project ID
NLST-1354
Initial CDAS Request Approval
Nov 5, 2024
Title
Predicting Lung Nodule Growth Using Deep Learning Methods
Summary
The National Lung Screening Trial (NLST) dataset provides a comprehensive collection of lung images and associated clinical data, offering a unique opportunity to advance the field of lung cancer detection and prediction. Our project aims to leverage deep learning techniques to predict the growth of lung nodules, a critical factor in determining the aggressiveness of potential malignancies and guiding clinical interventions. By analyzing the rich dataset provided by NLST, we intend to develop a robust predictive model that can accurately forecast nodule growth patterns, ultimately contributing to improved early detection and treatment strategies for lung cancer.
Aims

Aim 1: Data Preprocessing and Feature Extraction
Thoroughly preprocess the NLST dataset to standardize image data and extract relevant clinical features.
Develop a pipeline for data augmentation to enhance the diversity of the training dataset and improve model generalizability.
Aim 2: Model Development and Training
Design and implement deep learning architectures tailored for nodule growth prediction, including CNNs and RNNs to capture spatial and temporal patterns in nodule growth.
Train the models on the preprocessed NLST dataset, utilizing a stratified approach to ensure representation from various demographic and clinical groups.
Aim 3: Model Evaluation and Optimization
Evaluate the performance of the developed models.
Perform cross-validation to assess model stability and optimize hyperparameters for enhanced predictive performance.

Collaborators

Dong chen Gao, Ph.D. Student, School of Management, Shan Dong University