language: nl license: mit pipeline_tag: text-classification inference: false
A fine-tuned regression model that assigns a functioning level to Dutch sentences describing exercise tolerance functions. The model is based on a pre-trained Dutch medical language model (link to be added): a RoBERTa model, trained from scratch on clinical notes of the Amsterdam UMC. To detect sentences about exercise tolerance functions in clinical text in Dutch, use the icf-domains classification model.
|5||MET>6. Can tolerate jogging, hard exercises, running, climbing stairs fast, sports.|
|4||4≤MET≤6. Can tolerate walking / cycling at a brisk pace, considerable effort (e.g. cycling from 16 km/h), heavy housework.|
|3||3≤MET<4. Can tolerate walking / cycling at a normal pace, gardening, exercises without equipment.|
|2||2≤MET<3. Can tolerate walking at a slow to moderate pace, grocery shopping, light housework.|
|1||1≤MET<2. Can tolerate sitting activities.|
|0||0≤MET<1. Can physically tolerate only recumbent activities.|
The predictions generated by the model might sometimes be outside of the scale (e.g. 5.2); this is normal in a regression model.
- The model was fine-tuned (trained, validated and tested) on medical records from the Amsterdam UMC (the two academic medical centers of Amsterdam). It might perform differently on text from a different hospital or text from non-hospital sources (e.g. GP records).
- The model was fine-tuned with the Simple Transformers library. This library is based on Transformers but the model cannot be used directly with Transformers
pipelineand classes; doing so would generate incorrect outputs. For this reason, the API on this page is disabled.
To generate predictions with the model, use the Simple Transformers library:
from simpletransformers.classification import ClassificationModel model = ClassificationModel( 'roberta', 'CLTL/icf-levels-ins', use_cuda=False, ) example = 'kan nog goed traplopen, maar flink ingeleverd aan conditie na Corona' _, raw_outputs = model.predict([example]) predictions = np.squeeze(raw_outputs)
The prediction on the example is:
The raw outputs look like this:
- The training data consists of clinical notes from medical records (in Dutch) of the Amsterdam UMC. Due to privacy constraints, the data cannot be released.
- The annotation guidelines used for the project can be found here.
The default training parameters of Simple Transformers were used, including:
- Optimizer: AdamW
- Learning rate: 4e-5
- Num train epochs: 1
- Train batch size: 8
The evaluation is done on a sentence-level (the classification unit) and on a note-level (the aggregated unit which is meaningful for the healthcare professionals).
|mean absolute error||0.69||0.61|
|mean squared error||0.80||0.64|
|root mean squared error||0.89||0.80|
Jenia Kim, Piek Vossen