MRI Pattern Recognition in Multiple Sclerosis Normal-Appearing Brain Areas
2011

MRI Pattern Recognition in Multiple Sclerosis

Sample size: 67 publication Evidence: high

Author Information

Author(s): Weygandt Martin, Hackmack Kerstin, Pfüller Caspar, Bellmann–Strobl Judith, Paul Friedemann, Zipp Frauke, Haynes John­–Dylan

Primary Institution: Charité - University Medicine Berlin

Hypothesis

Can multiple sclerosis patients be distinguished from healthy controls using pattern classification of MRI data?

Conclusion

The study found that specific brain regions can indicate multiple sclerosis in both lesioned and normal-appearing brain areas.

Supporting Evidence

  • Maximal accuracy of 96% was achieved in a posterior parietal WM area for separating MS patients from controls.
  • Cerebellar regions showed 84% accuracy in identifying MS in normal-appearing grey matter.
  • 91% accuracy was found in a posterior region of normal-appearing white matter.

Takeaway

Doctors can use special computer programs to look at brain scans and tell if someone has multiple sclerosis, even in areas that look normal.

Methodology

Pattern classification was applied to MRI data from MS patients and healthy controls to identify diagnostic information in brain tissue.

Potential Biases

Potential biases in lesion mapping and classification accuracy due to the small sample size.

Limitations

The study may not generalize to all MS patients as it focused on a specific type of MS and a limited sample.

Participant Demographics

41 MS patients (mean age 35.7 years, 21 female) and 26 healthy controls (mean age 38.7 years, 14 female).

Statistical Information

P-Value

p<10−13

Statistical Significance

p<10−13

Digital Object Identifier (DOI)

10.1371/journal.pone.0021138

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication