Figure 1 | Scientific Reports

Figure 1

From: An interpretable data-driven prediction model to anticipate scoliosis in spinal muscular atrophy in the era of (gene-) therapies

Figure 1

(a) Data processing flow in the predictive modeling process.1230 visits of 86 genetically confirmed SMA patients were cleaned and merged on time. The 333 visits were labeled with corresponding scoliosis labels from available spine examinations for supervised training of a RandomForestClassifier (RFC). (b) Schematic visualization of data subsets for training and validation. The model’s predictions were tested on visits without scoliosis labels and patient subsets where the scoliosis was unknown. (c) Table 1 summarizes the demographics and features in the training and testing data set used during model development. Table 2 summarizes the demographics of the validation data set used to validate the model after training.

Back to article page