The content within this chapter is included in our Instructions for Use.

​

Host System Configuration

Lumi Prepare and Lumi Segment are cloud-based and can be used via a web browser with internet connection and/or a HoloLens 2.

 

Supported web browsers and operating systems include:

• Google Chrome (version 90.0.4430 and later) running on Windows 10 and later

• Microsoft Edge (version 83.0.0 and later) running on Windows 10 and later

• Mozilla Firefox (version 78.10 and later) running on Windows 10 and later

 

Supported AR HMDs include:

• Microsoft HoloLens 2 Mixed Reality Headset

 

Minimum network requirements:

• Minimum download speed 6.12 Mbps / upload speed 1.05 Mbps use with Microsoft HoloLens 2 (joint session feature)

​​

User Configuration

Only users with accounts provided by Augmedit can use the Lumi software. User can access the Lumi software with a personal login.

​

Segmentation Functionalities

For the conversion of 2D image data into 3D structures of the head, the following functionalities may be used:

• MRI T1 Segmentation Function (T1cSF): This function transforms contrast enhanced T1-weighted MR images (MRI-T1) of the head into 3D models of the skin, brain, and ventricles. Lumi Segment's T1cSF can thereby be used for tumor segmentation only in case of a single intracranial contrast enhancing tumor, diagnosed by a neurosurgeon or a neuroradiologist, with a minimal volume of 2.0 cc (0.1 in3) and a minimal diameter in any direction of 15 mm (0.6 inch), and a maximum volume of 100cc (6.1 in3) and a maximal diameter in any direction of 75 mm (3.0 inch).

• CT non-contrast Segmentation Function (CTncSF): This function transforms non-contrast enhanced CT Scans of the head into 3D Models of the skin, skull, brain, and the ventricles.

• Manual Segmentation Function: In the 3D Viewer, Scans can be visualized as 2D slices, based on the various gray values of the anatomical and pathological structures. Using a slider, the user can adjust the thresholds, thereby selecting which parts of the Scan are displayed. A specific section of the Scan can also be isolated using sliders to focus on a particular area. All modifications are directly visualized as a preview in the 3D Viewer. Once the user has made the correct selections and adjusted the settings, the software can transform the 3D preview into an actual 3D Model.

​

Input Image Requirements for T1cSF

• Image type: T1 MRI with contrast of the cranium from the vertex to at least the foramen magnum level and maximally to the C2 level.

• Slices per series: Minimum 100 slices per series with a maximum slice thickness and slice interval of 2 mm.

​​

Input Image Requirements for CTncSF

• Image type: CT images of the cranium.

• Slices per series:  Minimum 104 slices per series with a maximum slice thickness and slice interval of 2.5 mm.

​

Input Image Requirements for Manual Segmentation

• Image type: CT or MR images of the cranium

• Slices per series: Minimum 104 slices per series with a maximum slice thickness and slice interval of 2.5 mm

​

The visualization of the image data and the output representations are based on the data quality and resolution of the original image data series. In case of poor image contrast, resolution, scan artifacts (such as patient movement) or other image related defects or inaccuracies, this can lead to inaccurate results.

 

Input Image Limitations:

The Lumi software is not compatible with: ​

• Series with more than 2048 slices

• Scans in NifTi format

• Multiframe DICOM

• Modalities other than MR or CT (such as PET, DTI, PET etc.),

• DICOM series which contains multiple subseries 

• Scans with missing slices

​

Use Environment

In case Lumi Prepare is used in combination with the HoloLens 2 stable light levels are preferable for optimal performance. However the HoloLens 2 can be used in in dark environment (20 Lux), ambient lightning environment (600 Lux, e.g. outpatient clinic) and operation room (>600 Lux). Lumi Prepare and Lumi Segment can be used in the following environments with the following features: