This research develops a 3D modeling approach to comprehensively virtualize multiple pulmonary nodules throughout the whole lung, aiming to improve the diagnosis and treatment of early stage lung cancer patients. The key questions are how to accurately reconstruct nnodule distribution and interplay with lung tissue. Recent advance in deep learning and computer vision enable more accurate AI-assisted detection and segmentation of lung nodules.
However, limitation persist regarding whole lung modeling and the spatial relationships between multiple nodules. This research offers progress through a 3D reconduction technique for the entire lung. Integrating AI-driven medical imaging and visualization with a specialist clinical diagnosis and treatment is critical.
Technologies like deep learning for segmentation volumetric modeling, virtual augmented reality for 3D visualization, and multimodal data fusion are advancing whole lung modeling and multi-nodule assessment to enhance clinical decision making. This research established a effective 3D modeling approach for visualizing the distribution and the spatial relationships of multiple pulmonary nodules across the whole lung value. Key innovations include lung contour extraction, nodule reconstruction in 3D space, and interactive whole lung visualization.
This enables more accurate diagnosis and treatment planning for early lung cancer patients. Developing an accurate 3D modeling approach for visualizing whole lung nodule partners provide a new capability for comprehending disease progression. This can enable earlier diagnosis, personalized treatment plans, and improved outcomes for lung cancer patients.
The findings lay the groundwork for expanding whole lung modeling using multI-model data and advancing clinical translation. To begin, copy the patient's Digital Imaging and Communications in Medicine or DICOM data to a defined working directory. Using the file browser, examine each file directory to identify the image sequence with the highest number of scanning layers for analysis.
Employ the DICOM function within MATLAB by providing DICOM files as input to extract essential parameters, such as slice thickness and pixel spacing directly in the MATLAB environment. Then retrieve the location data for each image and access the information via info. SliceLocation in MATLAB workspace.
Next, use the SliceLocation function to save the location data into a variable and generate a plot for it. Click the Data tips button in the GUI's upper right corner to add data points and enhance the plot. Then use the volume resort function to organize all images and extract the images ranging from the first to the maximum location.
Safeguard the volume data from the valid images along with their sorted index. Using the size function in MATLAB, examine the three dimensional scale of a 3D volume. To view the 3D volume using the SliceView command function, record the sequence scan range obtaining the lungs from 60 to 340.
Then use the command to obtain a 3D volume containing all the data of the entire lung. Employ the MATLAB command function DICOM info to obtain the slice thickness of the image sequence. Use the command to calculate the number of Z axes for the isovoxel transformation.
Then use the MATLAB command function imresize3 to perform isovoxel transformation on V1.Employ the ThreeD_Slice_View function to view the isovoxel-transformed 3D volume. To remove the noise interference, use the Data Tips button to add continuous data points within the interactive interface. Next, right click on the data tips and select Export Cursor Data to Workspace to export the reference boundary for spatial filtering to the MATLAB workspace.
Invoke the Noise_Clean function to apply spatial filtering to V2 using the input parameter CI from the workspace. Use the Slice_View command function to visualize the resulting volume. Select a template slice such as the two 30 second image for the image segmentation design and assign it to a variable.
Then open the MATLAB image segmenter GUI by executing the imagesegmenter1 command. Select the Auto Cluster tool from the toolbar at the top and click the left mouse button to execute the command. Next, click the Show Binary button in the upper right corner to display the image in black and white binary.
To make the lung region white, select the Invert Mask button from the top toolbar and click the left mouse button for the execution of the command. To eliminate the white color outside the lung area, select the Clear Borders button to the top toolbar and click the left mouse button. Initiate the 3D lung volume function within the MATLAB workspace.
Next, select Maximize in the dropdown menu in the top right corner of the fourth view. Select MIP Projection. And then choose the jet color map from the built-in color maps option below.
Once more, invoke the Slice_View function, but this time, input the entire lung's 3D volume. Within the resulting GUI, use the bottom scroll bar to navigate to the region where the dominant lung nodules are situated, spanning scans 48 to 70. Use the ThreeDlung_Horizon function to conduct a 3D reconstruction of the region of interest, encompassing sections 48 to 70 from the whole lung's 3D volume.
