direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

MA: Medical Disease Classification of Videos through U-net

left: contrast enhanced time series; right: normal ultrasound image
Lupe

In the medical domain, visualizing the interior of our body is crucial for detecting structures and diseases hidden by our skin. Such visualizations are created by advanced technology like MRI, CT, Ultra Sound (Sonography), which capture fine granular information in millimeter sizes nowadays. As MRI and CT images are cost intensive, Ultra Sound and Endoscopic imaging can be performed as a cheaper solution.

The imaging gives the physician insights about the type of tissue, which might by a tumor or not. 

In this thesis, the following tasks are included:

  • Import real patient video data as e.g. DICOM format into the analysis framework
  • Testing best U-net for tissue recognition
  • Create visualizations for the algorithmic results (videos, static images and helpful diagrams)
  • Evaluating the accuracy of results 

 

Prerequisites for working on this topic are advanced knowledge in algorithmic design and good programming skills in either Java or Python.

Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Auxiliary Functions

Contact

Florian Schmidt
+49 30 314 28306
Room TEL 1205