UTFaculteitenEEMCSDisciplines & departementenDMBAssignmentsOpen AssignmentsOpen Master Assignments[B] Can you find the Gorilla in the CT scan? (2 topics: Sample and Object level)

[B] Can you find the Gorilla in the CT scan? (2 topics: Sample and Object level)

Master Assignment

[B] Can you find the Gorilla in the CT scan?  (2 topics: Sample and Object level)

Type: Master EE/CS/ITC

Period: TBD

Student: (Unassigned)

If you are interested please contact :

Background:

OOD data is data that follows a different distribution than data that a model was trained on. This can lead to erroneous results from the AI, even if an otherwise top performing model is used. The MICCAI Medical OOD Analysis challenge aims to investigate OOD performance in various situations. To achieve this it gives a standardized dataset and benchmark for anomaly detection.

You can choose one of two topics:

1)      Sample-level (i.e. patients-level) analysis, thus detecting out-of-distribution samples. For example, having a pathological condition or any other condition not seen in the training-set. This can pose a problem to classically supervised algorithms and can further allow physicians to prioritize different patients.

2)      Object-level analysis i.e. giving a score for each voxel, highlighting abnormal conditions and potentially guiding the physician.

 Resources:

  1. MICCAI Medical OOD Challenge https://www.synapse.org/Synapse:syn21343101/wiki/599515
  2. MICCAI OOD Github: https://github.com/MIC-DKFZ/mood