Objective: To evaluate the utility of radiomics to differentiate essential tremor (ET) from healthy controls using radiomics features of the posterior fossa on routine T1 weighted images.

Background: The cerebellum and cerebellar peduncles are suggested to play a significant role in the pathogenesis of ET. However, these abnormalities are seldom evident in basic structural imaging.  It is plausible that advanced techniques such as radiomics which extract subtle intensity, statistical and textural features could provide a biological signature for ET.

Method: T1 weighted images were acquired for 40 subjects with ET and 38 HC.  Seventeen regions in the cerebellar lobules were defined bilaterally from the SUIT atlas. These included lobules I-IV, V, VI, Crus I, Crus II, Vermis Crus I, Vermis Crus II, VIIb, VIIIa, VIIIb, IX, Vermis IX, X, Vermis X, dentate, interposed and fastigial nuclei. The inferior, superior and middle cerebellar peduncles defined bilaterally on a probabilistic atlas of cerebellar white matter at a 50% probability threshold.  Pyradiomics was used to extract features, and 60 most discriminant features of were used as inputs to the final classifier. The sklearn machine learning library was used for all implementations.

Results: The radiomics based RF classifier performed with an accuracy of 83.33% (average CV accuracy 85.14%, Area under- ROC =0.92).  The 60 most important features that were retained after two-stage feature selection included 22 GLCM features, 16 GLRLM, 9 first order, 6 GLDM and 7 GLSZM features. The ROIs that these belonged to were mostly right hemispheric, and consisted of the Right Dentate, Right V, Right VI, Right I-1V, bilaterally the Crus I and SCP, Vermis VI, and the Left MCP  of which Right Dentate and Right V alone contributed to the top ten of the most discriminative features.

Conclusion: The top 10 most discriminative features identified in patients with ET included the dentate nucleus and the lobule V of the cerebellum.  The dentate nucleus has been frequently suggested to be involved in the pathogenesis of ET, additionally, lobule V is part of the motor cerebellum which is involved in the cerebello-thalamo-cortical network, a part of the tremor network.  These results suggest that radiomics was able to accurately identify regions suggested to be involved in the pathogenesis of ET.  Future studies including other tremor disorders will be key to establish the utility of radiomics in diagnosing ET.

« Back to 2022 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/t1-posterior-fossa-radiomics-based-diagnosis-of-essential-tremor/