Definiens Health Image Intelligence - From Images to Clinical Decisions - Interactive 3D Segmentation and Quantification of Lung Tumors

: Figure 1: Screenshot of an MPR view depicting a segmented lung tumor of the current CT scan. In the 3D visualization, the detected lung tumor is colored yellow with a red bounding box. To help the user identify the anatomical location of the lung tumor, the 3D visualization also shows the lung (transparent and purple) as well as the high-intensity structures (e.g., bones in transparent and yellow). The figure additionally depicts the associated transversal (1a), coronal (1c), and sagittal (1d) views.

Lung cancer is one of the most common types of cancer in terms of incidence, and the most common cause of death related to cancer. Since the manual analysis and measurement of 3D structures – such as tumors – is extremely subjective, time-consuming, and cost-intensive, computer-aided methods offer great promise in today’s challenging clinical environment. However, due to significant variability in the appearance and shape of lung tumors, producing a reliable automated analysis is not an easy task.

: Figure 2: Screenshot of an MPR view depicting a segmented lung tumor of the prior CT scan. In the 3D visualization, the detected lung tumor is colored yellow with a red bounding box. To help the user identify the anatomical location of the lung tumor, the 3D visualization also shows the lung (transparent and purple) as well as the high-intensity structures (e.g., bones in transparent and yellow). The figure additionally depicts the associated transversal (2a), coronal (2c), and sagittal (2d) views.

: Figure 3: Screenshot of a comparison view depicting a segmented lung tumor of the current and prior CT scan. The figure depicts the associated transversal, coronal, and sagittal views at both time points.

Definiens Developer XD lets you create a software solution for the interactive analysis of lung tumors. Firstly, the user selects the axial image slices where the tumor of interest is largest. Clicking in the center of the tumor defines a seed point. From this seed point, Definiens Developer XD automatically calculates the tumor boundaries in all three dimensions. If the software has not yet calculated the tumor outlines accurately, multiple refinement steps are possible, such as additional seed points, blocker points, and interactive as well as manual editing.

The standard parameters for quantification are the longest diameter on a single slice of the tumor, and its perpendicular diameter (short axis). As Definiens Developer XD provides a volumetric segmentation of the tumor, more meaningful and reliable parameters – such as the tumor volume, morphology, context, and textural features – can be extracted.

The software also facilitates the volumetric analysis of lung tumors over time. This helps researchers and medical experts to detect, analyze, and quantify lung tumors, providing a highly accurate assessment of their change in size.

Use Case: Interactive 3D Segmentation and Quantification of |Lung Tumors

About Definiens