Dicom Error Handling for 3d Cnn Classification on Lidc-Idri Dataset

I am seeking a skilled freelancer to assist with an academic project focused on lung nodule classification using a 3D Convolutional Neural Network (cnn) and the lidc-idri dataset. The project's code is largely complete, but we are encountering a critical issue during the DICOM file loading process. Specifically, some dicom files within the lidc-idri dataset are causing the program to crash due to missing or malformed attributes.

The primary objective of this project is to implement robust error handling mechanisms for DICOM file loading. The ideal solution will prevent the program from failing when encountering problematic files. This can be achieved by either gracefully skipping files that cannot be processed or by assigning appropriate default values to missing attributes, ensuring the integrity and functionality of the overall data pipeline.

Key responsibilities include:
- Analyzing the existing DICOM loading code to identify points of failure related to missing attributes.
- Implementing error handling (e.g., Try-except blocks, conditional checks) to manage exceptions caused by incomplete DICOM headers.
- Developing strategies to either skip unprocessable DICOM files or infer/assign default values for critical missing attributes.
- Ensuring the implemented solution is efficient and does not significantly impact the data loading performance.
- Providing clean, well-documented code that integrates seamlessly with the existing project structure.