Differentiating intradiploic orbital dermoid and epidermoid cysts utilizing clinical features and machine learning

Correlation Between Radiologic and Histopathologic Features of Orbital Epidermoid and Dermoid Cysts

PURPOSE

Epidermoid cysts (EC) and dermoid cysts (DC) typically appear as well-circumscribed lesions on CT. This study aimed to clarify the radiologic and histopathologic characteristics of orbital EC and DC and to determine the correlations between them.

METHODS

The medical records of 69 patients who underwent surgery for orbital DC or EC at Samsung Medical Center between January 2001 and August 2016 were retrospectively reviewed. The size and location of the cysts, rim enhancement, homogeneity of contents, presence of hemorrhagic or calcific components, radiodensity of contents, and extent of bony remodeling were evaluated using CT. Additionally, the cyst lining and contents were examined histopathologically.

RESULTS

Among patients with orbital cysts, EC and DC were diagnosed in 10 (14.5%) and 59 (85.5%) patients, respectively. Further, 50.0% of EC and 79.7% of DC were located in the superotemporal quadrant of the orbit. On orbital CT, the average radiodensity of EC and DC was 18.9 ± 56.2 and -67.9 ± 63.3 HU, respectively. The cystic contents were more frequently homogeneous than heterogeneous in both EC and DC; however, the radiodensity of cysts differed significantly, which may be attributed to sebaceous gland activity. Focal bony notching, bone remodeling under pressure, and bony changes from dumbbell-shaped cysts were observed more frequently in DC than in EC.

CONCLUSIONS

Radiological and histopathological features are correlated in orbital EC and DC. Therefore, orbital EC and DC can be preoperatively differentiated using CT, based on the average radiodensity and bony remodeling.

Toward Precision Diagnosis: Machine Learning in Identifying Malignant Orbital Tumors With Multiparametric 3 T MRI

BACKGROUND

Orbital tumors present a diagnostic challenge due to their varied locations and histopathological differences. Although recent advancements in imaging have improved diagnosis, classification remains a challenge. The integration of artificial intelligence in radiology and ophthalmology has demonstrated promising outcomes.

PURPOSE

This study aimed to evaluate the performance of machine learning models in accurately distinguishing malignant orbital tumors from benign ones using multiparametric 3 T magnetic resonance imaging (MRI) data.

MATERIALS AND METHODS

In this single-center prospective study, patients with orbital masses underwent presurgery 3 T MRI scans between December 2015 and May 2021. The MRI protocol comprised multiparametric imaging including dynamic contrast-enhanced (DCE), diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), as well as morphological imaging acquisitions. A repeated nested cross-validation strategy using random forest classifiers was used for model training and evaluation, considering 8 combinations of explanatory features. Shapley additive explanations (SHAP) values were used to assess feature contributions, and the model performance was evaluated using multiple metrics.

RESULTS

One hundred thirteen patients were analyzed (57/113 [50.4%] were women; average age was 51.5 ± 17.5 years, range: 19-88 years). Among the 8 combinations of explanatory features assessed, the performance on predicting malignancy when using the most comprehensive model, which is the most exhaustive one incorporating all 46 explanatory features-including morphology, DWI, DCE, and IVIM, achieved an area under the curve of 0.9 [0.73-0.99]. When using the streamlined "10-feature signature" model, performance reached an area under the curve of 0.88 [0.71-0.99]. Random forest feature importance graphs measured by the mean of SHAP values pinpointed the 10 most impactful features, which comprised 3 quantitative IVIM features, 4 quantitative DCE features, 1 quantitative DWI feature, 1 qualitative DWI feature, and age.

CONCLUSIONS

Our findings demonstrate that a machine learning approach, integrating multiparametric MRI data such as DCE, DWI, IVIM, and morphological imaging, offers high-performing models for differentiating malignant from benign orbital tumors. The streamlined 10-feature signature, with a performance close to the comprehensive model, may be more suitable for clinical application.

Endoscope-Assisted Removal of Post-Traumatic Orbital Epidermoid Inclusion Cyst: A Useful Adjunct

Orbital epidermoid cysts are uncommon lesions within the bony orbit with varied symptomatology related to both the eye and the sino-nasal system. They are often slow-growing cystic masses which may cause facial asymmetry and visual loss due to pressure symptoms. Cross-sectional imaging such as computed tomography and magnetic resonance imaging are contributory and useful for assessment of the size and actual extent and should be mandatory before planning any surgical intervention. Open approaches and needle aspiration have been traditionally described; however, the use of the rigid nasal endoscope in the intraorbital compartment is a useful adjunct for exploration of the extent of the lesion and for complete surgical clearance. In this report, a 69 years old female with an old post- traumatic orbital epidermoid cyst which was removed completely using an endoscope via transorbital route was presented with the review of literature.