I am building a simple system to detect sleep-apnea events in children under the age of 10 using three signals from the same single person: SpO₂ (oxygen level), ECG (heart signal), and snoring audio. We do not have any datasets, so you will need to find suitable data on your own and clearly provide the dataset source links or files that you used. I need you to build an RCNN model and include preprocessing steps to clean and align the three signals. The model should be trained only on data from one person, and the final output should show how accurately it detects apnea. Please report the model’s accuracy, precision, recall, F1-score, ROC-AUC, and confusion matrix. You should also explain the types of noise present in each signal and the types of filters used to remove those noises (for example, low-pass, high-pass, band-pass, or notch filters). The final deliverables include: Commented Python code for preprocessing, training, and evaluation A short README explaining setup and retraining A demo notebook or short video showing the results on the single-person data A note or document listing dataset sources, noise types, and filters applied