POS0928 DEEP LEARNING BASED ASSESSMENT OF SALIVARY GLAND ULTRASONOGRAPHY IMAGES FOR SUPPORTING THE DIAGNOSIS OF PRIMARY SJOGREN’S SYNDROME

Background

Salivary gland ultrasound (SGUS) has proven to be a promising tool for diagnosing primary Sjögren's syndrome (pSS). However, the widespread use of it as standardized diagnostic tools is limited by inter/intra operator variability.

Objectives

The aim of this study was to evaluate the utility of deep learning-based SGUS image assessment in the diagnosis of pSS.

Methods

Between Sep 2021 and Oct 2022, 1133 SGUS images of 137 patients from one center were included in this retrospective study. Among them, 61 patients with 480 images were diagnosed as pSS and 76 patients with 653 images were non-SS. All the SGUS image data were randomly divided into training dataset (50%), validation dataset (20%) and testing dataset (30%). The SGUS automatic classification model was developed by using a deep residual convolutional network architecture (RESNET). The predictive performance was validated by sensitivity, specifcity and area under reciver operating characteristic curve (ROC).

Results

When applying deep learning-based SGUS image assessment, it showed better performance than operator based SGUS score system by improving the specifcity (86.9% vs. 80.1%), while similar sensitivity (59.4% vs. 61.4%). The area under the ROC were comparable between them (0.800 vs 0.775).

Conclusion

Deep learning-based SGUS image assessment maybe an objective and promising tool compared to expert-based scoring of SGUS in the diagnosis of pSS. This may support SGUS as an effective and prospective diagnostic tool to supplement current diagnostic methods.

References

[1]Vukicevic AM, Milic V, Zabotti A, Hocevar A, De Lucia O, Filippou G, Frangi AF, Tzioufas A, De Vita S, Filipovic N. Radiomics-Based Assessment of Primary Sjögren's Syndrome From Salivary Gland Ultrasonography Images. IEEE J Biomed Health Inform. 2020 Mar;24(3):835-843. [2]Vukicevic AM, Radovic M, Zabotti A, Milic V, Hocevar A, Callegher SZ, De Lucia O, De Vita S, Filipovic N. Deep learning segmentation of Primary Sjögren's syndrome affected salivary glands from ultrasonography images. Comput Biol Med. 2021 Feb;129:104154.

Acknowledgements:

NIL.

Disclosure of Interests

None Declared.