Would you be able to design:
a) A system (neural network) to detect and extract Fetal ECG (FECG) from (pregnant mother's) abdominal mixed signal.
b) A system that can predict the fetal status (normal or abnormal) condition, for a given database.
Note: The platform should be python (jupyter notebook).
1. The FECG extraction can be done via ICA algorithm [either JADE-ICA or FastICA] (a type of BSS).
2. The FECGSYN Database should be taken from [login to view URL] This database is in WFDB format, the following link can be used to have it work in python environment: [login to view URL]
3. The system should predict normal or abnormal fetal conditions from the FECG (after extraction) based on FQRS complex.
1. Abdominal mixed signal should be displayed before extraction.
2. FECG should be displayed after extraction.
3. Neural network (or deep learning) deployment that can read all database AFTER extraction (in training phase) and then predict the fetal condition (normal or abnormal) on the testing database. The graph should have points that says its normal or abnormal otherwise.
The reference paper of the database is attached. It also includes information about the FQRS complex, how FECG extraction can be done? .
3 freelanceri licitează în medie 204$ pentru acest proiect
Hi I am a data scientist having 9 years of experience in this field. I have developed numerous applications in machine learning deep learning and computer vision. I can complete your task efficiently. I have worked on Mai multe
Hi, there. I can implement your task using Scikit-learn. What accuracy do you need for that classification task? Let's discuss it more. Best regards, Zoltan