Frequently asked questions (FAQ)

We’d like to understand the delivery of CoLumbo.

CoLumbo delivers pathology description and editable reports for a very short time. Then the physician will review this outcome, but time for review and corrections will be short. Some errors of omission will be avoided. CoLumbo also allows for PACS integration, including automated download, analysis and export.

Can the whole spine be covered?

Smart Soft performed some tests that will demonstrate that, in general, CoLumbo performs segmentation of C-Spine exams and even detects some types of pathology. So the next version that will be ready in 2022 and will support C-spine. In 2024 we will support CT modality as well.

Is the accuracy level so high that a radiologist just signs the report without checking it?
Up to now, not all types of pathology are detected, so a radiologist will have to check the report. The main metric is the accuracy of a radiologist using the software – it is better than a radiologist not using the software.
Shall it be done in future versions?

Smart Soft intends to include Foraminal stenosis, Modic changes, Schmorl’s nodes, Scoliosis, Fractures, Osteophytes, Disc degeneration disease,HIZ, Sequestration, Subarticular stenosis and Ligamentum Flavum Hypertrophy in the next version and reach and possibly surpass the human level. However, some very rare types of pathology will not be included soon. Smart Soft’s product is an assistant tool for physicians and not as a product that replaces the physicians. The physician may agree to the CoLumbo report but may also choose to add or remove some texts.

Are you aiming for full automation of the segmentation?

Segmentation is fully automatic. In some cases, the user might change the key points and as a consequence, some measurements will be automatically performed again.

The researcher version has many measurements, while not all of them are displayed on the display design of the version intended for release. Why?

All measurements are calculated; however, most of them are not usually included in the report by the radiologists. . For that reason, the release version includes the measurements that are correlated to the detected pathologies. Other measurements, like foramen size and spinal canal diameter, are also included. This is because they can help confirm detected pathologies or identify others that are undetected.

Are CoLumbo reports exportable?

Yes. The full report can be saved as .rtf or .pdf or send to a printer. It is also part of the entire exported study in .pdf format.

What about reach at the same time sensitivity 95% and specificity 95%?

These requirements are unlikely to be met even by a perfect algorithm. The human level is 75-85% sensitivity, 90-95% sensitivity if disagreement levels are used for calculations.

What are the prices? Will some radiologists not be willing to buy a product unless for a onetime payment?

Technically, we can propose different packages to the customers, including price per exam, per year, and unlimited. That way, we can follow the pricing schema of our partners. The basis of our pricing is the fact that we save at least 25% of the time. In the western countries, this is 25% of 50 USD that is 12.5 USD. This means 3,500 -12,000 USD savings per year.

Are you planning to obtain approval or clearance for the USA, Canada, Australia, and Japan?

Our first release version is CE mark certified. We work on FDA approval and consider entering Canadian, Australian, China, and Japanese markets after our first contract, but our main priorities are the EU and the USA.

How do you see the high burden on the regulatory side, given the large number of features that CoLumbo has?

Concerning the number of features – this means more time and effort for risk management, usability engineering, testing, incl. Clinical investigations, but we do not consider certification will be delayed due to this. Clinical trials were performed in 3 hospitals and finished successfully in Oct 2020.

Can SmartSoft provide CoLumbo output to PACS or third party viewers?

Yes, we can provide API so that the software can be easily integrated into 3rd party software. Our standalone version will be mainly for demonstration purposes so one can download the exam from the PACS and, after that, load it into Columbo. The user can even change the settings of the PACS to allow for automatic download and upload of studies.

Some hospitals do not use cloud due to national regulation or other reasons. How you will handle this?

We may deploy software on their server, PC or the working station provided minimum requirements are met.

Are there any constraints in product deployment?

The software has minimum requirements, that have to be met, for it to run. Depending on the deployment methods, there are some differences, but there are general requisites for both: the machine should have at least 8GB of RAM, 600 MB free space for the installer and 4 GB for the installed software, 2.10 GHz CPU with at least 2 cores, Windows 10 or Windows 7 as an operating system.
Depending on the deployment, the following additional requirements have to be met:

Client-Server: Windows Server 2016 operating system, installed on the server side, an NVidia GPU with 2 GB memory and CUDA installed (exactly version 10.0), internet connection with minimum speed of 4 MBs;

Standalone: a local NVidia GPU with 2GB memory and CUDА installed (exactly version 10.0);

Version with study samples: 5 GB of additional free space.

How does CoLumbo run, i.e., on what software does it run?

Windows 10, Windows 7 or Windows Server 2016. The API can be used on Linux, too and it is possible to be used in specialized hardware. For predictions on a GPU, CUDA version 10.0 is required.

What acquisition sequences do you consider? What about spin-echo sequences

Modality – MR; Body part – Lumbar spine (L-Spine); Magnetic Field Strength 1,5-3T; Age – 18 to 70 (included);
At least one sagittal T2w sequence containing 5 slices or more.
MR Acquisition Type – 2D or 3D; Scanning Sequence – SE; Repetition Time: at least 940 ms; Echo Time: from 60 ms to 300 ms; Maximum slice thickness 5mm; Min Sagittal Resolutions (Cols x Rows): 280×280; Maximum voxel volume – 1.9 mm2 for 2D series and 1.0 mm3 for 3D series.

At least one axial T2w sequence.
Min Axial Resolutions (Cols x Rows) 192×192; Maximum slice thickness 5mm; MR Acquisition Type – 2D; Scanning Sequence – SE; Repetition Time: at least 940 ms; Echo Time: from 60 ms to 300 ms; Maximum voxel volume – 1.6 mm2.
Not intended to analyze images of pregnant persons. Does not support protocols with contrast injections (even if there are series w/o contrast it usually contains fracture, infection, postoperative complications or tumor that we will support in the next version). Does not support distorted and blurry images. Does not support patients with scoliosis.

Is fat suppression used?

For fat suppression and other sequences different than T2 spin-echo, the algorithm can support them both as single or as an additional sequence but only for research purposes. These capabilities will not be tested during the official clinical trials. We need 2 weeks per new sequence, as was the case with the Dixon protocol we covered for the MICCAI challenge.

What deep-learning techniques are being used?

-ResNet for feature map extraction.
-UNet for segmentation.
-3D convolution for 3D segmentation refinement.

Regarding the deep-learning techniques: on routine spine imaging, can CoLumbo be used without having to change/modify the scanning protocols? Radiology departments typically dislike this.

Regarding the deep-learning techniques: on routine spine imaging, can CoLumbo be used without having to change/modify the scanning protocols? Radiology departments typically dislike this.

Shall the user be able to define a custom threshold so that there are minimum false negatives?

Absolutely. The user can define the positive probability of pathology so he/she can work on high sensitivity mode. These can be different for different pathologies so one can play that way with the sensitivity vs. specificity tradeoff.

Does Columbo need T1 series?

Current version no. The next version will need it for Modic changes and accuracy improvements. However, the user can still load the series manually and review it.

Can you accurately detect levels if there are transitional vertebra?

Yes, if costal facets, bifurcation of the aorta all indicate the same labeling, otherwise the full spine image is needed and we just indicate that there is a transitional vertebra and labeling is not certain.

Shall the user be able to correct the results and edit the report?

Yes, with one or two clicks, pathology can be added or removed. Also, the user may add pathologies, unsupported by the software, as well as free text. The text of the report of the verified findings can also be edited.

Shall the user be able to quickly navigate to the pathological level?

Yes, the user can see the color-coding of the level labels and quickly navigate to the chosen level by selecting it from the level dropdown menu. The user can also navigate to the most prominent slice of a certain pathology by hovering over it in the pathology panel.

Is a radiologist needed anymore?

The software is designed as an assistant type of software. The first version saves 25% of the time. The aim is for the next versions to save up to 80-90% of the time. But the radiologists will be needed for at least the next 40 years.