syngo®.via for MI

Conventional planar lung perfusion and ventilation studies to determine regions and volumes are often inaccurate due to the limitations of 2D imaging.

• SPECT/CT studies provide 3-dimensional displays of lung perfusion and ventilation, allowing more accurate visualization and analysis of anatomy and function compared to a 2D planar image
• Incorporation of CT enables more precise anatomical segmentation which enhances the accuracy of lobar lung volumes estimation
• Manual delineation of the lung lobes is time consuming and may be prone to error

3D is more accurate than 2D, and may help optimize surgical management of lung cancer.

Auto Lung 3D provides accurate estimates to optimize treatment

• An AI-driven tool to precisely measure lung ventilation and perfusion using SPECT/CT data
• Provides a more accurate perfusion and ventilation estimate to help guide lung resection decisions
• Supports thoracic surgeons and pulmonologists in making more precise, data-driven treatment decisions

Auto Lung 3D is designed to²:

• Segment and quantify 3D anatomy
• Automatically locate and contour lungs
• Segment right/left lung lobes and quantify ventilation/perfusion
• Generate a report containing the % of total perfusion and volumes

• AI-based deep-learning algorithm automatically detects and consistently segments 5 lung lobes freeing up time to analyze more studies
• Automation and quality improvement may boost the volume of ventilation/perfusion studies, helping to increase profitability
• 3D evaluation better reflects patient anatomy than 2D planar and can improve patient management and surgical planning

Automatic fusion of SPECT and CT images allows use of Auto Lung 3D without a SPECT/CT combined system.

Lesion Scout with Auto ID automatically assesses an entire patient data set, segments all uptake, identifies the foci of interest, and proposes a selection of foci to be excluded and included from findings creation.
Reproducibility of functional tumor burden calculations is potentially improved and the possibility of missing a lesion may be minimized.

• Lesion Scout finds multiple tumors automatically based on SUV threshold
• Auto ID identifies physiological and non-physiological uptake 
• Proposed classifications are color coded and presented in the preview mode for review and confirmation by the user prior to final acceptance

• Seamless integration of AI
• Identify and propose all foci to be included and excluded from findings generation 
• Quickly deliver comprehensive results such as whole-body tumor burden

• Metabolic tumor volume (MTV) and total lesion glycolysis (TLG) reflect the total viable cancer volume in the patient
• Literature shows that MTV and TLG may offer comprehensive quantification of disease
• Lesion Scout with Auto ID automatically proposes findings to be included in the TLG and MTV calculation. After acceptance, a report is generated that can be communicated to the referring physician

Lesion Scout with Auto ID utilizes a deep-learning AI algorithm that was trained with PET/CT images annotated by clinical experts who classified uptake into physiological and non-physiological (considered for whole-body tumor burden).

“The fully automated…classification of…PET uptake patterns in foci suspicious and nonsuspicious for cancer in patients with lung cancer and lymphoma…is feasible and achieves high diagnostic performance when both CT and PET images are used.”

“For PET experts, reading PET/CT is not  a challenge. For a non-expert like me, reading PET/CT is always a question – is it a hot spot due to tumor or normal variation…without AI it would be extremely tedious.”