Skip to content

docs: fix README spelling typos #119

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
thehanslevi wants to merge 1 commit into
base: master
Choose a base branch
from
Open

docs: fix README spelling typos #119

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,16 +2,16 @@

This package provides trained U-net models for lung segmentation. For now, four models are available:

- U-net(R231): This model was trained on a large and diverse dataset that covers a wide range of visual variabiliy. The model performs segmentation on individual slices, extracts right-left lung seperately includes airpockets, tumors and effusions. The trachea will not be included in the lung segmentation. https://doi.org/10.1186/s41747-020-00173-2
- U-net(R231): This model was trained on a large and diverse dataset that covers a wide range of visual variability. The model performs segmentation on individual slices, extracts right-left lung separately includes airpockets, tumors and effusions. The trachea will not be included in the lung segmentation. https://doi.org/10.1186/s41747-020-00173-2

- U-net(LTRCLobes): This model was trained on a subset of the [LTRC](https://www.nhlbi.nih.gov/science/lung-tissue-research-consortium-ltrc) dataset. The model performs segmentation of individual lung-lobes but yields limited performance when dense pathologies are present or when fissures are not visible at every slice.

- U-net(LTRCLobes_R231): This will run the R231 and LTRCLobes model and fuse the results. False negatives from LTRCLobes will be filled by R231 predictions and mapped to a neighbor label. False positives from LTRCLobes will be removed. The fusing process is computationally intensive and can, depdending on the data and results, take up to several minutes per volume.
- U-net(LTRCLobes_R231): This will run the R231 and LTRCLobes model and fuse the results. False negatives from LTRCLobes will be filled by R231 predictions and mapped to a neighbor label. False positives from LTRCLobes will be removed. The fusing process is computationally intensive and can, depending on the data and results, take up to several minutes per volume.

- [U-net(R231CovidWeb)](#COVID-19-Web)


**Examples of the two models applied**. **Left:** U-net(R231), will distinguish between left and right lung and include very dense areas such as effusions (third row), tumor or severe fibrosis (fourth row) . **Right:** U-net(LTRLobes), will distinguish between lung lobes but will not include very dense areas. **LTRCLobes_R231** will fuse LTRCLobes and R231 results. **R231CovidWeb** is trained with aditional COVID-19 data.
**Examples of the two models applied**. **Left:** U-net(R231), will distinguish between left and right lung and include very dense areas such as effusions (third row), tumor or severe fibrosis (fourth row) . **Right:** U-net(LTRLobes), will distinguish between lung lobes but will not include very dense areas. **LTRCLobes_R231** will fuse LTRCLobes and R231 results. **R231CovidWeb** is trained with additional COVID-19 data.

![alt text](figures/figure.png "Result examples")

Expand Down Expand Up @@ -109,7 +109,7 @@ The model works on full slices only. The slice to process has to show the full l
```
lungmask INPUT OUTPUT --modelname R231CovidWeb
```
The regular U-net(R231) model works very well for COVID-19 CT scans. However, collections of slices and case reports from the web are often cropped, annotated or encoded in regular image formats so that the original hounsfield unit (HU) values can only be estimated. The training data of the U-net(R231CovidWeb) model was augmented with COVID-19 slices that were mapped back from regular imaging formats to HU. The data was collected and prepared by MedSeg (http://medicalsegmentation.com/covid19/). While the regular U-net(R231) showed very good results for these images there may be cases for which this model will yield slighty improved segmentations. Note that you have to map images back to HU when using images from the web. This [blog post](https://medium.com/@hbjenssen/covid-19-radiology-data-collection-and-preparation-for-artificial-intelligence-4ecece97bb5b) describes how you can do that. Alternatively you can set the ```--noHU``` tag.
The regular U-net(R231) model works very well for COVID-19 CT scans. However, collections of slices and case reports from the web are often cropped, annotated or encoded in regular image formats so that the original hounsfield unit (HU) values can only be estimated. The training data of the U-net(R231CovidWeb) model was augmented with COVID-19 slices that were mapped back from regular imaging formats to HU. The data was collected and prepared by MedSeg (http://medicalsegmentation.com/covid19/). While the regular U-net(R231) showed very good results for these images there may be cases for which this model will yield slightly improved segmentations. Note that you have to map images back to HU when using images from the web. This [blog post](https://medium.com/@hbjenssen/covid-19-radiology-data-collection-and-preparation-for-artificial-intelligence-4ecece97bb5b) describes how you can do that. Alternatively you can set the ```--noHU``` tag.
![alt text](figures/example_covid.jpg "COVID examples")

## jpg, png and non HU images
Expand Down