Merge pull request #144 from nhsengland/aib-tag-ontology

tag ontology implemented
nhsengland · Dec 12, 2024 · 2fe1bc5 · 2fe1bc5
2 parents 768baa0 + e1540f7
commit 2fe1bc5
Show file tree

Hide file tree

Showing 45 changed files with 142 additions and 48 deletions.
diff --git a/docs/our_work/a_and_e_forecasting_tool.md b/docs/our_work/a_and_e_forecasting_tool.md
@@ -2,7 +2,7 @@
 title: 'Accident and Emergency (A&E) Forecasting Tool'
 summary: 'A probabilistic model which gives a three-week forecast for A&E departments, predicting their expected admissions loads.'
 origin: ''
-tags: ['MODELLING', 'HOSPITAL', 'MONTE CARLO',  'URGENT CARE', ]
+tags: ['FORECASTING', 'HOSPITAL',  'URGENT CARE', 'WIP', 'PYTHON']
 ---
 
 ![Image showing plot of historical and predicted admissions rates. The plot shows the confidence intervals for the model's fit to historical data, and for its predictions.](../images/a_and_e_forecasting/forecast.png)

diff --git a/docs/our_work/adrenal-lesions.md b/docs/our_work/adrenal-lesions.md
@@ -3,7 +3,7 @@ title: 'Using deep learning to detect adrenal lesions in CT scans'
 summary: 'This project explored whether applying AI and deep learning augment the detection of adrenal incidentalomas in patients’ CT scans.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['CLASSIFICATION','LESION DETECTION','COMPUTER VISION','AI']
+tags: ['CLASSIFICATION','COMPUTER VISION', 'ACCURACY']
 ---
 
 <figure markdown >

diff --git a/docs/our_work/ai-deep-dive.md b/docs/our_work/ai-deep-dive.md
@@ -3,7 +3,7 @@ title: 'AI Deep Dive Workshops'
 summary: 'The NHS AI Lab Skunkworks team have developed and delivered a series of workshops to improve confidence working with AI.'
 category: 'Playbooks'
 origin: 'Skunkworks'
-tags : ['AI', 'GUIDANCE', 'BEST PRACTICE']
+tags : ['BEST PRACTICE']
 ---
 
 <figure markdown >

diff --git a/docs/our_work/ai-dictionary.md b/docs/our_work/ai-dictionary.md
@@ -3,7 +3,7 @@ title: 'AI Dictionary'
 summary: 'A simple dictionary of common AI terms with a health and care context.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags : ['AI', 'DICTIONARY', 'JAVASCRIPT', 'REACT']
+tags : ['DOCUMENTATION', 'WEBDEV']
 ---
 
 [![Image of a browser showing the AI dictionary.](../images/ai-dictionary.png)](https://nhsx.github.io/ai-dictionary)

diff --git a/docs/our_work/ai-ethics.md b/docs/our_work/ai-ethics.md
@@ -2,7 +2,7 @@
 title: 'AI Ethics in Practice at NHS England'
 summary: 'Defining ethical AI development best practice for data practitioners in the NHS'
 origin: 'NHS England'
-tags: ['AI', 'ETHICS', 'TRANSPARENCY', 'QUALITY', 'DOCUMENTATION', 'RESEARCH', 'GUIDANCE', 'BEST PRACTICE']
+tags: ['ETHICS', 'QUALITY', 'DOCUMENTATION', 'RESEARCH', 'BEST PRACTICE']
 ---
 !!! warning
 

diff --git a/docs/our_work/ai-skunkworks.md b/docs/our_work/ai-skunkworks.md
@@ -3,7 +3,7 @@ title: 'NHS AI Lab Skunkworks'
 summary: 'The NHS AI Lab Skunkworks team demonstrates the potential for AI in health and social care through practical experience'
 category: 'Overview'
 origin: 'Skunkworks'
-tags: ['CLASSIFICATION','LESION DETECTION','AI', 'PYTHON']
+tags: ['CLASSIFICATION', 'PYTHON']
 ---
 
 ![AI Skunkworks website homepage](../images/ai-skunkworks.png)

diff --git a/docs/our_work/ambulance-delay-predictor.md b/docs/our_work/ambulance-delay-predictor.md
@@ -3,7 +3,7 @@ title: 'Ambulance Handover Delay Predictor'
 summary: 'Predict ambulance delays at hospital, with reasons, to allow them to influence hospitals'' behaviour to mitigate against queues before they happen.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['AMBULANCE','PREDICTION','RANDOM FOREST', 'CLASSIFICATION', 'TIME SERIES', 'PYTHON']
+tags: ['AMBULANCE','FORECASTING','MACHINE LEARNING', 'CLASSIFICATION', 'TIME SERIES', 'PYTHON']
 ---
 
 ![Ambulance Handover Delay Predictor screenshot showing the handover times expected for different hospitals, with the high times highlighted in orange.](../images/ambulance-delay-predictor.png)

diff --git a/docs/our_work/bed-allocation.md b/docs/our_work/bed-allocation.md
@@ -3,7 +3,7 @@ title: 'Bed allocation'
 summary: 'Machine learning to effectively aid bed management in Kettering General Hospital.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['HOSPITAL','BAYESIAN FORECASTING','MONTE CARLO','GREEDY ALLOCATION', 'PYTHON']
+tags: ['HOSPITAL','FORECASTING', 'PYTHON', 'WEBDEV']
 ---
 
 ![Browser showing the dashboard for Kettering General Hospital that shows the forecasting of their bed occupancy.](../images/bed-allocation.png)

diff --git a/docs/our_work/c245_synpath.md b/docs/our_work/c245_synpath.md
@@ -3,7 +3,7 @@ title: Building the Foundations for a Generic Patient Simulator (SynPath)
 summary: Developing an agent-based simulation for generating synthetic patient pathways and scenario modelling for healthcare specific implementations.
 category: Projects
 permalink: c245_synpath.html
-tags: ['SYNTHETIC DATA', 'PATHWAYS','SIMULATION']
+tags: ['SYNTHETIC DATA','SIMULATION']
 ---
 
 ![](../images/c245fig1.png)

diff --git a/docs/our_work/c338_poud.md b/docs/our_work/c338_poud.md
@@ -2,7 +2,7 @@
 title: How to Assess the Privacy of Unstructured Data
 summary: What are the privacy considerations that need to be addressed when dealing with unstructured healthcare text data
 permalink: c338_poud.html
-tags: ['UNSTRUCTURED DATA', 'PRIVACY', 'PII', 'BEST PRACTICE']
+tags: ['UNSTRUCTURED DATA', 'ETHICS', 'PII', 'BEST PRACTICE', 'STRUCTURED DATA']
 ---
 
 <figure markdown>

diff --git a/docs/our_work/c339_sas.md b/docs/our_work/c339_sas.md
@@ -2,7 +2,7 @@
 title: Creating a Generic Adversarial Attack for any Synthetic Dataset
 summary: Can the privacy of a generated dataset be assessed through downstream adversarial attacks to highlight the risk of re-identification
 permalink: c339_sas.html
-tags: ['SYNTHETIC DATA', 'GAN','TABULAR DATA']
+tags: ['SYNTHETIC DATA','STRUCTURED DATA']
 ---
 
 ![](../images/sas.png)

diff --git a/docs/our_work/c399_privfinger.md b/docs/our_work/c399_privfinger.md
@@ -2,7 +2,7 @@
 title: Building a Tool to Assess the Privacy Risk of Text Data
 summary: Can we generate usable privacy scores for text data to support understanding of privacy concerns and the anonymisation process
 permalink: c399_privfinger.html
-tags: ['TEXT DATA', 'LLM','PYTHON', 'PRIVACY']
+tags: ['TEXT DATA', 'LLM','PYTHON', ]
 ---
 
 This work was undertaken as an external commission aiming to build a pipeline of components which firstly generated unstructured medical notes using a structured output from [Synthea:tm:](https://github.com/synthetichealth/synthea) and then running these through [GPT-3.5](https://platform.openai.com/docs/models/gpt-3-5) models to transform these into human readable notes.

diff --git a/docs/our_work/casestudy-recruitment-shortlisting.md b/docs/our_work/casestudy-recruitment-shortlisting.md
@@ -3,7 +3,7 @@ title: 'Examining whether recruitment data can, and should, be used to train AI
 summary: 'Identify where bias has potential to occur when using machine learning for shortlisting interview candidates and mitigate it'
 category: 'CaseStudies'
 origin: 'Skunkworks'
-tags: ['NLP', 'NEURAL NETWORKS']
+tags: ['NLP', 'NEURAL NETWORKS', 'SYNTHETIC DATA']
 ---
 
 ## Info

diff --git a/docs/our_work/casestudy-synthetic-data-pipeline.md b/docs/our_work/casestudy-synthetic-data-pipeline.md
@@ -3,7 +3,7 @@ title: 'Exploring how to create mock patient data (synthetic data) from real pat
 summary: 'The generation of safe and effective synthetic data to be used in technologies that improve health and social care.'
 category: 'CaseStudies'
 origin: 'Skunkworks'
-tags: ['SYNTHETIC DATA','VAE','PRIVACY','QUALITY','UTILITY','AI', 'PYTHON']
+tags: ['SYNTHETIC DATA','NEURAL NETWORKS','ETHICS','QUALITY','PYTHON', 'MIMIC']
 ---
 
 ![Example graphs studying the fidelity of the synthetic data to the artificial data.](../images/example_report_output.png)

diff --git a/docs/our_work/ct-alignment.md b/docs/our_work/ct-alignment.md
@@ -3,7 +3,7 @@ title: 'CT Alignment and Lesion Detection'
 summary: 'A range of classical and machine learning computer vision techniques to align and detect lesions in anonymised CT scans over time from George Eliot Hospital NHS Trust.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['CT','COMPUTER VISION','IMAGE REGISTRATION','PYTHON']
+tags: ['CT','COMPUTER VISION','PYTHON', 'ACCURACY']
 ---
 
 ![CT Alignment and Lesion Detection screenshot of the interface for identifying lesions.](../images/ct-alignment.png)

diff --git a/docs/our_work/data-lens.md b/docs/our_work/data-lens.md
@@ -3,7 +3,7 @@ title: 'Data Lens'
 summary: 'Data Lens brings together information about multiple databases, providing a fast-access search in multiple languages.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['NLP', 'SEMANTIC SEARCH', 'SCRAPING','JAVASCRIPT','PYTHON']
+tags: ['NLP','JAVASCRIPT','PYTHON', 'WEBDEV']
 ---
 
 ![Image of a browser showing the data lens search front end.](../images/data-lens.png)

diff --git a/docs/our_work/data-linkage-hub/linkage-projects/better-matching.md b/docs/our_work/data-linkage-hub/linkage-projects/better-matching.md
@@ -3,7 +3,7 @@ title: 'Probabilistic Linkage Model'
 summary: 'This project is creating a probabilistic linkage model using Splink, in order to improve linkage outcomes, and by extension, patient outcomes. The aim is for this to be used to link data in a range of NHS datasets.'
 category: 'Projects'
 origin: 'NHSD'
-tags: ['LINKAGE', 'PYTHON', 'PROBABILISTIC MODEL']
+tags: ['LINKAGE', 'PYTHON', 'PII', 'WIP','STRUCTURED DATA']
 ---
 
 ## Crafting a model that suits NHS England data linkage needs

diff --git a/docs/our_work/data-linkage-hub/linkage-projects/cop.md b/docs/our_work/data-linkage-hub/linkage-projects/cop.md
@@ -3,7 +3,7 @@ title: 'Data Linkage Community of Practice (DL CoP)'
 summary: 'We are creating and leading a community of practice to help people do the best linkage they can, with support from the data linkage team, but also from fellow analysts who are actively working on data linkage.'
 category: 'Projects'
 origin: 'NHSD'
-tags: ['BEST PRACTICE','EXPLAINABILITY','LINKAGE']
+tags: ['BEST PRACTICE','EXPLAINABILITY','LINKAGE', 'WIP']
 ---
 
 ## Why do we want a Community of Practice?

diff --git a/docs/our_work/data-linkage-hub/linkage-projects/mps-handbook.md b/docs/our_work/data-linkage-hub/linkage-projects/mps-handbook.md
@@ -3,7 +3,7 @@ title: 'MPS Documentation - the Person_ID handbook'
 summary: 'Documenting how the Person_ID is generated via the Master Person Service (MPS), to make the current process of linking data in the NHS more transparent and easy to understand.'
 category: 'Projects'
 origin: 'NHSD'
-tags: ['BEST PRACTICE','EXPLAINABILITY','LINKAGE']
+tags: ['BEST PRACTICE','EXPLAINABILITY','LINKAGE', 'COMPLETE']
 ---
 
 The Person_ID is a unique patient identifier used by NHS England with the objective of standardising the approach to patient-level data linkage across different data sets.

diff --git a/docs/our_work/data-linkage-hub/linkage-projects/qaf.md b/docs/our_work/data-linkage-hub/linkage-projects/qaf.md
@@ -3,7 +3,7 @@ title: 'Quality Assurance Framework for Data Linkage'
 summary: 'This project aims to create, test, and distribute a quality assurance framework for data linkage to ensure robust, transparent and auditable results.'
 category: 'Projects'
 origin: 'NHSD'
-tags: ['BEST PRACTICE','EXPLAINABILITY','LINKAGE']
+tags: ['BEST PRACTICE','EXPLAINABILITY','LINKAGE', 'WIP']
 ---
 
 Data Linkage is a business-critical process within many government organisations, including NHS England. Research publications, official statistics, but also many direct care applications depend on data linkage. Its importance is further amplified when considering privacy preserving principles that require to minimise the use of patients' personal identifiable information. Consequently, data linkage is initiated early in the data lifecycle, establishing a substantial **reliance of downstream applications on the quality of the linkage process**.

diff --git a/docs/our_work/ds251_RAG.md b/docs/our_work/ds251_RAG.md
@@ -2,7 +2,7 @@
 title: 'Retrieval Augmented Generation'
 summary: 'Investigating Advanced RAG, collaborating with efforts to evaluate LLM outputs.'
 origin: ''
-tags: ['NLP','LLM','GENAI']
+tags: ['NLP','LLM']
 ---
 
 

diff --git a/docs/our_work/ds255_privacyfp.md b/docs/our_work/ds255_privacyfp.md
@@ -1,7 +1,7 @@
 ---
 title: "Building a Tool to Assess the Privacy Risk of Text Data - Extended"
 summary: Can we generate a modular tool to score the privacy risk of healthcare free-text data using open-source LLMs and NERs.
-tags: ['TEXT DATA', 'LLM','PYTHON', 'PRIVACY', 'NAMED ENTITY RECOGNITION', 'UNSTRUCTURED DATA']
+tags: ['TEXT DATA', 'LLM','PYTHON', 'ETHICS', 'NLP', 'UNSTRUCTURED DATA', 'SYNTHETIC DATA']
 ---
 
 !!! warning

diff --git a/docs/our_work/index.md b/docs/our_work/index.md
@@ -2,15 +2,16 @@
 <h3 align="center"><i>Explore our comprehensive portfolio of ongoing and completed projects that harness the power of data to drive insight.</i></h3>
 
 <script>
-    var sites = ['long-stay', 'c245_synpath', 'nursing-placement-optimisation', 'ai-dictionary', 'p14_mcr', 'a_and_e_forecasting_tool', 'p21_synthvae', 'casestudy-recruitment-shortlisting', 'open-safely', 'synthetic-data-pipeline', 'ds218_rap_community_of_practice', 'p31_txtrayalign2', 'p24_lime', 'p11_synpathdiabetes', 'c338_poud', 'swpclab', 'p22_txtrayalign', 'data-lens', 'ambulance-delay-predictor', 'index', 'p12_synthvae', 'ai-deep-dive', 'ratings-and-reviews', 'parkinsons-detection', 'bed-allocation', 'ct-alignment', 'sde_data_validation', 'long-stay-baseline', 'ai-skunkworks', 'adrenal-lesions', 'p23_stm', 'c339_sas', 'p32_phmdiabetes', 'p33_patientsafetylms', 'casestudy-synthetic-data-pipeline', 'c399_privfinger', 'nhs-resolution', 'renal-health-prediction', 'ds255_privacyfp', 'c250_nhscorpus', 'ds251_RAG', 'index', 'qaf', 'better-matching', 'mps-handbook', 'cop'];
+    var sites = ['ai-dictionary', 'ai-skunkworks', 'adrenal-lesions', 'c338_poud', 'data-lens', 'a_and_e_forecasting_tool', 'casestudy-synthetic-data-pipeline', 'p24_lime', 'parkinsons-detection', 'renal-health-prediction', 'ds251_RAG', 'ratings-and-reviews', 'p42_hypergraphs2', 'synthetic-data-pipeline', 'p32_phmdiabetes', 'index', 'p23_stm', 'ct-alignment', 'c245_synpath', 'swpclab', 'casestudy-recruitment-shortlisting', 'p22_txtrayalign', 'nursing-placement-optimisation', 'ds218_rap_community_of_practice', 'ambulance-delay-predictor', 'p12_synthvae', 'ai-deep-dive', 'p31_txtrayalign2', 'ai-ethics', 'c339_sas', 'c399_privfinger', 'p11_synpathdiabetes', 'ds255_privacyfp', 'long-stay-baseline', 'p14_mcr', 'nhs-resolution', 'p34_hypergraphs', 'open-safely', 'c250_nhscorpus', 'bed-allocation', 'p43_medcat', 'sde_data_validation', 'p33_patientsafetylms', 'p21_synthvae', 'long-stay', 'data-linkage-hub/index', 'data-linkage-hub/linkage-projects/mps-handbook', 'data-linkage-hub/linkage-projects/qaf', 'data-linkage-hub/linkage-projects/cop', 'data-linkage-hub/linkage-projects/better-matching'];
 
     function randomSite() {
         var i = parseInt(Math.random() * sites.length);
         location.href = sites[i];
     }
 </script>
 
-<button class="hero md-button md-button--primary" onclick="randomSite()" >Take me to a random project :material-puzzle:</button> [Explore our projects by topic :fontawesome-solid-tags:](./tags.md){ .md-button }
+<button class="hero md-button md-button--primary" onclick="randomSite()" >Take me to a random project :material-puzzle:</button> [Explore our projects by categories & tags :fontawesome-solid-tags:](./tags.md){ .md-button }
+
 
 <p align="center">
   <img src="../images/ourwork.jpeg" width=700 alt="" ></img>

diff --git a/docs/our_work/long-stay-baseline.md b/docs/our_work/long-stay-baseline.md
@@ -3,7 +3,7 @@ title: 'Long Stayer Risk Stratification Baseline Models'
 summary: 'Baseline machine learning models using historical data from Gloucestershire Hospitals NHS Foundation Trust to predict how long a patient will stay in hospital upon admission.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['LOS','RISK MODEL', 'REGRESSION', 'CLASSIFICATION','PYTHON']
+tags: ['LOS', 'REGRESSION', 'CLASSIFICATION','PYTHON', 'SQL', 'F1', 'ACCURACY']
 ---
 
 Long Stayer risk stratification baseline models was selected as a project to run in tandem with the [Long Stayer Risk Stratification](long-stay.md) project, and started in March 2022.

diff --git a/docs/our_work/long-stay.md b/docs/our_work/long-stay.md
@@ -3,7 +3,7 @@ title: 'Long Stayer Risk Stratification'
 summary: 'Machine learning using historical data from Gloucestershire Hospitals NHS Foundation Trust to predict how long a patient will stay in hospital upon admission.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['LOS','NEURAL NETWORKS','RISK MODEL','PYTHON', 'GAN']
+tags: ['LOS','NEURAL NETWORKS','PYTHON', 'SQL', 'WEBDEV', 'STRUCTURED DATA', 'ACCURACY']
 ---
 
 ![browser with the Long Stayer Risk Stratification dashboard showing a zoom in to the current risk stratification score currently at: Level 2](../images/long-stay.png)

diff --git a/docs/our_work/nhs-resolution.md b/docs/our_work/nhs-resolution.md
@@ -3,7 +3,7 @@ title: 'Predicting negligence claims with NHS Resolution'
 summary: 'This project investigated whether it is possible to use machine learning AI to predict the number of claims a trust is likely to receive and learn what drives them in order to improve safety for patients.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['CLASSIFICATION','PREDICTION', 'AI']
+tags: ['CLASSIFICATION','FORECASTING']
 ---
 
 <figure markdown>

diff --git a/docs/our_work/nursing-placement-optimisation.md b/docs/our_work/nursing-placement-optimisation.md
@@ -3,7 +3,7 @@ title: 'Nursing Placement Schedule Optimisation Tool'
 summary: 'Optimisation problem developed with Imperial College Healthcare Trust to produce optimised schedules for student nurses going on placement within the trust.'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['OPTIMISATION','GENETIC ALGORITHM', 'PYTHON']
+tags: ['PYTHON']
 ---
 
 This project is an example of the AI Skunkworks team offering capability resources to produce proof-of-concepts which could be applicable to the NHS at large. The project ran from January 2022 to May 2022.

diff --git a/docs/our_work/open-safely.md b/docs/our_work/open-safely.md
@@ -3,7 +3,7 @@ title: 'Working with a Trusted Research Environment - the NHS @Home programme'
 summary: 'An exploration of OpenSafely'
 category: 'Projects'
 origin: 'Skunkworks'
-tags: ['TRE', 'PYTHON', 'OPENSAFELY']
+tags: ['PYTHON', 'OPENSAFELY']
 ---
 
 OpenSAFELY gives trusted researchers restricted levels of access to the server to run analysis on real data and obtain aggregate results, without having sight of the patient level data. Aggregate results are checked to ensure there are no disclosure risks before being released from the server. This highly secure way of working enables researchers to have access to large and sensitive datasets in a safe manner.

diff --git a/docs/our_work/p11_synpathdiabetes.md b/docs/our_work/p11_synpathdiabetes.md
@@ -3,7 +3,7 @@ layout: base
 title: Applying our SynPath Simulator to a Diabetes Pathway
 summary: Exploration work into incorporating learning into a pathway simulator for diabetes.  This work has fed our current SynPathGo project to create synthetic patient pathways and a foundation for agent based modelling in the NHS.
 permalink: p11_synpathdiabetes.html
-tags: ['SIMULATION']
+tags: ['SIMULATION', 'SYNTHETIC DATA']
 ---
 
 <figure markdown>

diff --git a/docs/our_work/p12_synthvae.md b/docs/our_work/p12_synthvae.md
@@ -3,7 +3,7 @@ layout: base
 summary: The initial creation of a variational autoencoder with differential privacy for generating single table tabular gaussian data.  This work demonstrated the feasibility of this approach for healthcare and fed into further interactions of the code base.
 title: Investigating Differential Privacy in a Variational AutoEncoder for Synthetic Data Generation
 permalink: p12_synthvae.html
-tags: ['VAE', 'SYNTHETIC DATA', 'PYTHON']
+tags: ['NEURAL NETWORKS', 'SYNTHETIC DATA', 'PYTHON']
 ---
 
 ![Diagram showing the flow of data in a variational autoencoder. Starts with an input labeled as "x" on the left, which is passed through an NN arrow. This encoder outputs two posterior parameters: μ and variance σ². These are then used to sample a latent variable "z" from a normal distribution z ~ N(μ, σ²). "z" is passed to an NN decoder, which reconstructs the input, producing the reconstruction labeled as "x̂" on the right.](../images/vae.png)

diff --git a/docs/our_work/p14_mcr.md b/docs/our_work/p14_mcr.md
@@ -3,7 +3,7 @@ layout: base
 title: Using Model Class Reliance to Understand the Impact of Commerical Data on Predictions
 permalink: p14_mcr.html
 summary: How to assess the value that commercial sales data of over-the-counter prescriptions has on respiratory death predictions
-tags: ['MCR', 'PYTHON', 'MORTALITY', 'RESPIRATORY']
+tags: ['MCR', 'PYTHON', 'MORTALITY', 'RESPIRATORY', 'RANDOM FOREST']
 ---
 
 The primary aim of the project was to apply the novel variable importance technique, [model class reliance](https://papers.nips.cc/paper/2020/hash/fd512441a1a791770a6fa573d688bff5-Abstract.html), to machine learning models which could predict registered respiratory deaths in the UK. The objective was to assess the value of commercial health data in healthcare predictions compared to other available datasets.

diff --git a/docs/our_work/p21_synthvae.md b/docs/our_work/p21_synthvae.md
@@ -3,7 +3,7 @@ layout: base
 title: Developing our SynthVAE code
 permalink: p21_synthvae2.html
 summary: Improving our variational autoencoder to consider fairness and to run on non-gaussian distributions
-tags: ['VAE', 'PYTHON', 'GAUSSIAN MIXTURE MODEL', 'DAG']
+tags: ['NEURAL NETWORKS', 'PYTHON', 'SYNTHETIC DATA']
 ---
 
 ![DAG showing the relationships between variables that may influence job opportunity. Five nodes labeled as Prior Experience, Qualifications, Networking, Gender, and Job Opportunity. Prior Experience, Qualifications, Networking, and Gender all have arrows pointing toward Job Opportunity. Additionally, Networking is influenced by Prior Experience, and Gender influences Networking.](../images/dag_job_opportunity.png)

diff --git a/docs/our_work/p22_txtrayalign.md b/docs/our_work/p22_txtrayalign.md
@@ -3,7 +3,7 @@ layout: base
 title: Descriptive text from X-Ray Images (TxtRayAlign)
 permalink: p22_txtrayalign.html
 summary: Generating descriptive text from X-Ray images using contrastive learning on multi-modal data
-tags: ['NLP', 'MULTI-MODAL', 'RETRIEVAL', 'CONTRASTIVE-LEARNING', 'PYTHON']
+tags: ['NLP', 'MULTI-MODAL', 'RETRIEVAL', 'DEEP LEARNING', 'PYTHON']
 ---
 
 ![](../images/p22fig1.png)

diff --git a/docs/our_work/p23_stm.md b/docs/our_work/p23_stm.md
@@ -3,7 +3,7 @@ layout: base
 title: Text Analysis using Structural Topic Modelling
 permalink: p23_stm.html
 summary: An open reusable tool for topic modelling of survey responses
-tags: ['NLP', 'STRUCTURAL TOPIC MODELLING', 'R']
+tags: ['NLP', 'R']
 ---
 
 ![Screenshot of the STM Insights dashboard, top left has an interopic distance map, top left includes a graph for most relevant terms, beneath are various other low res graphs.](../images/stminsights_lowquality.png)