MRI Segmentation
MRI Knowledge Hub
Why Segmentation
One of the principal problems in therapy, Surgical or Radiotherapy planning is the localization of critical structures inside body with respect to the diseased tissue for safest possible approach. Monitoring therapy response is another important area for clinician to plan therapy. All these issues involve segmentation of anatomical and pathological boundaries in the body structures manually or automatically.
Segmentation in MRI
Segmentation of anatomical structure in MR images is challenging due to many reasons. Clinically MR image analysis of tissue boundaries plays an important role in surgery and monitoring therapy. In MR images, the intensity variation at boundaries in normal tissue (like in brain images; White matter, Gray Matter or CSF) is gradual and not sharp due to many reason like; partial volume effect, tissue inhomogeniety, variability in tissue types and imaging variabilities. Similarly, in diseased region (brain tumor), there are intensity variations within the diseased area because pixels can belong to edema, necrosis, tumor and healthy tissues. Due to these reason no single method is capable of delineating tissue boundaries in MR images. You will find lot of literature in MR segmentation by various methods. Some of methods and publications given below:
Mannual or Automatic
Gradient operator for edge detection
Threshold based on k-Mean
Publications
Segmentation of Meningiomas and Low Grade Gliomas in MRI; M. R. Kaus, etl. Link
What is Fuzzy Logic
Fuzzy logic defines variable processing that explore multiple possible truth values to be processed through the same variable. It is a form of many-valued logic in which the truth value of variables may be any real number between 0 and 1.Fuzzy set theory was first formally introduced by Zadeh (1965). Originating in ancient Greek philosophy, fuzzy set theory is an alternative to the traditional notions of set membership and logic. Fuzzy logic has been developed as a method to capture the implicit vagueness in everyday language and systems. Fuzzy logic has found many applications in diverse fields such as pattern recognition, image and signal processing, hardware design and synthesis, layout of integrated circuits, artificial intelligence, expert and decision support systems (medical diagnosis), business and social studies. Almost all the answers found in practical life are within some proximity of the absolute truth.
Website Links
Fuzzy in Medical Image Segmentation
Fuzzy C-mean clustering, fuzzy rule based system, is widely used in medical image segmentation on diseased area. The fuzzy clustering algorithm establishes the expression of the uncertainty of the sample category and can describe the ambiguity brought by the partial volume effect in MRI image. It is found suitable for brain MRI image segmentation.
- https://pubmed.ncbi.nlm.nih.gov/31812927/
- https://pubmed.ncbi.nlm.nih.gov/34393718/
- https://pubmed.ncbi.nlm.nih.gov/33841095/
Fuzzy Rule Based segmentation
Fuzzy rule based system is also widely used in segmentation of different structure in MR images, using different features which can be extracted from Images. These features can be intensity, shape, texture, histogram, pixels’s neighbourhood properties etc. Here Fuzzy rules, which may have structure like “If then else” are framed based on expert in the related field called handcrafted fuzzy rules[].
Review of MR image segmentation techniques using pattern recognition