+ 
-
+
-EHRoes is developped and used in collaboration with the [Morphological Unit of the Institute of Myology of Paris](https://www.institut-myologie.org/en/recherche-2/neuromuscular-exploration-and-evaluation-centre/laboratoire-dhistopathologie-dr-norma-b-romero/). A production instance is deployed to help discovering new relevant features for congenital myopathies classification and diagnosis.
+IMPatienT is developped and used in collaboration with the [Morphological Unit of the Institute of Myology of Paris](https://www.institut-myologie.org/en/recherche-2/neuromuscular-exploration-and-evaluation-centre/laboratoire-dhistopathologie-dr-norma-b-romero/). A production instance is deployed to help discovering new relevant features for congenital myopathies classification and diagnosis.
-## EHRoes Abstract
+## IMPatienT🗂️ Abstract
-**Background**
-With a growing amount of patient data such as sequencing, imaging, and medical records, electronic health records (EHR) are now central to drive research and improve diagnosis of patients by using multimodal data. Exploiting patient data is a challenge as patient data exploitation tools ecosystem is often fragmented into tools to digitize and format the data and those interpreting the data (exploration, diagnosis). Furthermore, as most of the tools are specialized in one type of data, a multitude of software is needed for multimodal approaches. There is a strong need for a simple, all-rounder and flexible platform.
-**Results**
-In this paper we present EHRoes, a web application for multimodal digitization and exploration of patient data. EHRoes has a module-based architecture, composed of four modules to: (i) create a standard vocabulary for a domain (ii) automatically digitize free-text data to a set of standard terms (iii) annotate images with standard vocabulary using automatic segmentation and (iv) generate an automatic visualization dashboard to provide insight on the data and perform automatic diagnosis suggestions. We demonstrated the utility of EHRoes by digitizing 40 artificial muscle histology reports of patients with congenital myopathies.
-**Conclusions**
-With EHRoes we created a platform for both digitization and exploration of patient data that can handle image data and free-text data. As it uses user-designed standard vocabulary, it is highly flexible to fit any domain of research. It can be used both as a patient registry with automatic diagnosis or as a research tool to explore a cohort of patients.
-A demo instance of the application is available at https://ehroes.lbgi.fr.
+**Background**
+Medical acts such as imaging leads most of the time to the production of several medical text reports to describe relevant findings. Such process induce multimodality in patient data by linking image data to free-text data. Multimodal data have become central to drive research and improve diagnosis of patients. Exploiting patients’ data is challenging as the ecosystem of tools is fragmented depending on the type of data to exploit (image, text, genetic), the task to perform (digitization, processing, exploration) and the domain of interest (clinical phenotype, histology…). There is a strong need for a simple, comprehensive, and flexible platform.
+**Results**
+In this paper, we present IMPatienT (d**I**gitize **M**ultimodal **PATIEN**t da**T**a), a free and open-source web application to digitize, process and explore multimodal patient data. IMPatienT has a modular architecture, composed of four components to: (i) create a standard vocabulary for a domain (ii) digitize and process free-text data by mapping it to a set of standard terms, (iii) annotate images and perform image segmentation and (iv) generate an automatic visualization dashboard to provide insight on the data and perform automatic diagnosis suggestions. Finally, we showcased IMPatienT on a corpus of 40 simulated muscle biopsy reports of congenital myopathy patients.
+**Conclusions**
+IMPatienT is a platform to digitize, process and explore patient data that can handle image and free-text data. As it relies on user-designed standard vocabulary, it is highly flexible to fit any domain of research and can be used as a patient registry for exploratory data analysis (EDA).
+A demo instance of the application is available at https://impatient.lbgi.fr.
## Setup guides
### (DOCKER) Developper Mode Setup (to contribute)
-[See the wiki page: Developper Mode Setup (DOCKER)]({{ url_for('auth.reset_password', token=token, _external=True) }}
If you have not requested a password reset simply ignore this message.
Sincerely,
-The EHRoes Team
+The IMPatienT Team
diff --git a/app/auth/templates/email_reset_password.txt b/app/auth/templates/email_reset_password.txt index 558e841..0d6c281 100644 --- a/app/auth/templates/email_reset_password.txt +++ b/app/auth/templates/email_reset_password.txt @@ -8,4 +8,4 @@ If you have not requested a password reset simply ignore this message. Sincerely, -The EHRoes Team \ No newline at end of file +The IMPatienT Team diff --git a/app/historeport/static/historeport.js b/app/historeport/static/historeport.js index 9c82389..c9ccdc4 100644 --- a/app/historeport/static/historeport.js +++ b/app/historeport/static/historeport.js @@ -115,7 +115,7 @@ conclusion_tag.on("input", onInputConclusion); // Tagify AJAX Function to get a list of Orphanet names var myHeaders_orpha = new Headers({ - apiKey: "ehroes", + apiKey: "impatient", }); var options_orpha = { headers: myHeaders_orpha, diff --git a/app/imgupload/static/imgupload.js b/app/imgupload/static/imgupload.js index c363786..526782a 100644 --- a/app/imgupload/static/imgupload.js +++ b/app/imgupload/static/imgupload.js @@ -11,7 +11,7 @@ conclusion_tag.on("input", onInputConclusion); // Tagify AJAX Function to get a list of Orphanet names var myHeaders_orpha = new Headers({ - apiKey: "ehroes", + apiKey: "impatient", }); var options_orpha = { headers: myHeaders_orpha, diff --git a/app/index/templates/index.html b/app/index/templates/index.html index 6a46c06..95bf727 100644 --- a/app/index/templates/index.html +++ b/app/index/templates/index.html @@ -1,34 +1,38 @@ {% extends "base.html" %} {% block content %}
-  }})
+
- EHRoes is developped for my thesis within the
- MYO-xIA project.
It features a standard vocabulary
+ IMPatienT (dIgitize Multimodal PATIENt
+ daTa) is a web application developped in the
+ MYO-xIA project for patient data digitization and exploration.
It features a standard
+ vocabulary
creator, optical character recognition (OCR), natural language processing
(NLP), image annotation and segmentation using machine learning, interactive
visualizations and automatic diagnosis prediction.
- If you are on the demo instance (https://ehroes.lbgi.fr), you can login on top
+ If you are on the demo instance (https://impatient.lbgi.fr), you can login
+ on top
right with
login: demo ; password: demo
The application is
preloaded with a database of histology reports and a standard vocabulary for
muscle biopsy. Data on this demo instance is reset once every day.
For the PDF automatic analysis, you can find a sample PDF
- HERE
+ HERE
working with the preinstalled standard vocabulary. A sample image for image
annotation can be found
- HERE.
+ HERE.
EHRoes is developped and used in collaboration with the IMPatienT is developped and used in collaboration with the Morphological Unit of the Institute of Myology of Paris. A production instance is deployed to help discovering new relevant features for congenital myopathies classification and diagnosis.
- Background
With a growing amount of patient data such as sequencing, imaging, and medical
- records, electronic health records (EHR) are now central to drive research and improve diagnosis of patients by
- using multimodal data. Exploiting patient data is a challenge as patient data exploitation tools ecosystem is often
- fragmented into tools to digitize and format the data and those interpreting the data (exploration, diagnosis).
- Furthermore, as most of the tools are specialized in one type of data, a multitude of software is needed for
- multimodal approaches. There is a strong need for a simple, all-rounder and flexible platform.
+ Background
+
+ Medical acts such as imaging leads most of the time to the production of several medical text reports to describe
+ relevant findings. Such process induce multimodality in patient data by linking image data to free-text data.
+ Multimodal data have become central to drive research and improve diagnosis of patients. Exploiting patients’ data
+ is challenging as the ecosystem of tools is fragmented depending on the type of data to exploit (image, text,
+ genetic), the task to perform (digitization, processing, exploration) and the domain of interest (clinical
+ phenotype, histology…). There is a strong need for a simple, comprehensive, and flexible platform.
Results
- In this paper we present EHRoes, a web application for multimodal digitization and exploration of patient data.
- EHRoes has a
- module-based architecture, composed of four modules to: (i) create a standard vocabulary for a domain (ii)
- automatically digitize free-text data to a set of standard terms (iii) annotate images with standard vocabulary
- using automatic segmentation and (iv) generate an automatic visualization dashboard to provide insight on the data
- and perform automatic diagnosis suggestions. We demonstrated the utility of EHRoes by digitizing 40 artificial
- muscle histology reports of patients with congenital myopathies.
+ In this paper, we present IMPatienT (dIgitize Multimodal PATIENt
+ daTa), a free and open-source web application to
+ digitize, process and explore multimodal patient data. IMPatienT has a modular architecture, composed of four
+ components to: (i) create a standard vocabulary for a domain (ii) digitize and process free-text data by mapping it
+ to a set of standard terms, (iii) annotate images and perform image segmentation and (iv) generate an automatic
+ visualization dashboard to provide insight on the data and perform automatic diagnosis suggestions. Finally, we
+ showcased IMPatienT on a corpus of 40 simulated muscle biopsy reports of congenital myopathy patients.
Conclusions
- With EHRoes we created a platform for both digitization and exploration of patient data that can handle image data
- and free-text data. As it uses user-designed standard vocabulary, it is highly flexible to fit any domain of
- research. It can be used both as a patient registry with automatic diagnosis or as a research tool to explore a
- cohort of patients. A demo instance of the application is available at https://ehroes.lbgi.fr.
+ IMPatienT is a platform to digitize, process and explore patient data that can handle image and free-text data. As
+ it relies on user-designed standard vocabulary, it is highly flexible to fit any domain of research and can be used
+ as a patient registry for exploratory data analysis (EDA). A demo instance of the application is available at https://impatient.lbgi.fr.
+
The MYO-xIA project aims to collect data from patients with congenital myopathies in order to analyse them using @@ -100,12 +106,12 @@
The main maintainer is:
Corentin Meyer - PhD Student @
CSTB Team - iCube - University Of Strasbourg co.meyer@unistra.fr
[placeholder]
-
+ 
From 9d7c48aa361f117cdcdb3356500de081b595be91 Mon Sep 17 00:00:00 2001
From: LS <20109584+lambda-science@users.noreply.github.com>
Date: Thu, 31 Mar 2022 10:24:33 +0200
Subject: [PATCH 08/13] Random State
---
app/dashapp/shapes_to_segmentations.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/dashapp/shapes_to_segmentations.py b/app/dashapp/shapes_to_segmentations.py
index 0bf518c..1d15d74 100644
--- a/app/dashapp/shapes_to_segmentations.py
+++ b/app/dashapp/shapes_to_segmentations.py
@@ -88,7 +88,7 @@ def compute_segmentations(
# do segmentation and return this
t1 = time()
clf = RandomForestClassifier(
- n_estimators=50, n_jobs=-1, max_depth=8, max_samples=0.05
+ n_estimators=50, n_jobs=-1, max_depth=8, max_samples=0.05, random_state=42
)
seg_matrix, clf = fit_segmenter(mask, features, clf)
t2 = time()
From 20a5ea60caece90bf835e0838f8728fc698501dd Mon Sep 17 00:00:00 2001
From: LS <20109584+lambda-science@users.noreply.github.com>
Date: Tue, 5 Apr 2022 23:32:04 +0200
Subject: [PATCH 09/13] Image example as default
---
.gitignore | 1 +
.pre-commit-config.yaml | 1 -
data/database/app.db.demo | Bin 3284992 -> 3284992 bytes
.../demo_patient_sample_image_histo.jpg | Bin 0 -> 534704 bytes
...ent_sample_image_histo.jpg_blend_image.png | Bin 0 -> 582682 bytes
...ient_sample_image_histo.jpg_classifier.pkl | Bin 0 -> 407978 bytes
...ent_sample_image_histo.jpg_mask_annot.json | 170 ++++++++++++++++++
...ient_sample_image_histo.jpg_mask_image.png | Bin 0 -> 24826 bytes
...ient_sample_image_histo.jpg_seq_matrix.npy | Bin 0 -> 332128 bytes
data/ontology/ontology.json.demo | 6 +-
10 files changed, 174 insertions(+), 4 deletions(-)
create mode 100644 data/images/demo_patient/demo_patient_sample_image_histo.jpg
create mode 100644 data/images/demo_patient/demo_patient_sample_image_histo.jpg_blend_image.png
create mode 100644 data/images/demo_patient/demo_patient_sample_image_histo.jpg_classifier.pkl
create mode 100644 data/images/demo_patient/demo_patient_sample_image_histo.jpg_mask_annot.json
create mode 100644 data/images/demo_patient/demo_patient_sample_image_histo.jpg_mask_image.png
create mode 100644 data/images/demo_patient/demo_patient_sample_image_histo.jpg_seq_matrix.npy
diff --git a/.gitignore b/.gitignore
index fe6bccd..e4bd1da 100644
--- a/.gitignore
+++ b/.gitignore
@@ -176,3 +176,4 @@ docker/run.sh
.gitconfig
*.tar.gz
IMPatienT
+!data/images/demo_patient
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 737449d..16a52f4 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -7,7 +7,6 @@ repos:
- id: trailing-whitespace
- id: end-of-file-fixer
- id: check-yaml
- - id: check-added-large-files
- repo: https://github.com/Lucas-C/pre-commit-hooks-bandit
rev: v1.0.5
diff --git a/data/database/app.db.demo b/data/database/app.db.demo
index 2213c8142340696890e65d42aabceaa4b248551c..3fbce5ff343b785bbbb64874070025a6ff42d99b 100644
GIT binary patch
delta 268652
zcmeEv2|!%s)i!q)h8Y-U$gm6~z_5m$VGBu&31rtuGKn#9$pR**XhMjIOEd~ttBuwK
z#Beljpw=X6A{gOymst0@5N%y!B(c_}tvY;vU0SXF=e^t9x!fTW)`9T-U%wwM%-s9F
z=RNN^&w0*sYC2wLYVbdG
