Application of Artificial Intelligence in Diagnosis of Craniopharyngioma

Automated segmentation of craniopharyngioma on MR images using U-Net-based deep convolutional neural network

OBJECTIVES

To develop a U-Net-based deep learning model for automated segmentation of craniopharyngioma.

METHODS

A total number of 264 patients diagnosed with craniopharyngiomas were included in this research. Pre-treatment MRIs were collected, annotated, and used as ground truth to learn and evaluate the deep learning model. Thirty-eight patients from another institution were used for independently external testing. The proposed segmentation model was constructed based on a U-Net architecture. Dice similarity coefficients (DSCs), Hausdorff distance of 95% percentile (95HD), Jaccard value, true positive rate (TPR), and false positive rate (FPR) of each case were calculated. One-way ANOVA analysis was used to investigate if the model performance was associated with the radiological characteristics of tumors.

RESULTS

The proposed model showed a good performance in segmentation with average DSCs of 0.840, Jaccard of 0.734, TPR of 0.820, FPR of 0.000, and 95HD of 3.669 mm. It performed feasibly in the independent external test set, with average DSCs of 0.816, Jaccard of 0.704, TPR of 0.765, FPR of 0.000, and 95HD of 4.201 mm. Also, one-way ANOVA suggested the performance was not statistically associated with radiological characteristics, including predominantly composition (p = 0.370), lobulated shape (p = 0.353), compressed or enclosed ICA (p = 0.809), and cavernous sinus invasion (p = 0.283).

CONCLUSIONS

The proposed deep learning model shows promising results for the automated segmentation of craniopharyngioma.

KEY POINTS

• The segmentation model based on U-Net showed good performance in segmentation of craniopharyngioma. • The proposed model showed good performance regardless of the radiological characteristics of craniopharyngioma. • The model achieved feasibility in the independent external dataset obtained from another center.

The Challenging Management of Craniopharyngiomas in Adults: Time for a Reappraisal?

Simple Summary

Craniopharyngiomas (CPs) currently represent one of the most challenging diseases to deal with in the group of skull base tumors. Due to their location near, within, or surrounding the pituitary gland and stalk, CPs can be revealed by pituitary tumor syndrome and/or symptoms of hormonal deficiencies. Furthermore, surgery, which represents the first-line therapy, almost always results in hypopituitarism, diabetes insipidus and, in the case of hypothalamic involvement by the tumor, the occurrence of hypothalamic syndrome. The latter is characterized by intractable weight gain associated with severe morbid obesity, memory impairment, attention deficit, reduced impulse control and, eventually, increased risk of cardiovascular and metabolic disorders. Recent progress made in the understanding of the molecular pathways involved in CPs tumorigenesis paves the way for promising alternative therapeutic approaches and diagnostic procedures. Taken together, they lay the groundwork for new paradigms in the management of CPs in adults.

Abstract

Craniopharyngiomas (CPs) are rare tumors of the skull base, developing near the pituitary gland and hypothalamus and responsible for severe hormonal deficiencies and an overall increase in mortality rate. While surgery and radiotherapy represent the recommended first-line therapies for CPs, a new paradigm for treatment is currently emerging, as a consequence of accumulated knowledge concerning the molecular mechanisms involved in tumor growth, paving the way for anticipated use of targeted therapies. Significant clinical and basic research conducted in the field of CPs will undoubtedly constitute a real step forward for a better understanding of the behavior of these tumors and prevent associated complications. In this review, our aim is to summarize the multiple steps in the management of CPs in adults and emphasize the most recent studies that will contribute to advancing the diagnostic and therapeutic algorithms.