Predicting visual recovery in pituitary adenoma patients post-endoscopic endonasal transsphenoidal surgery: Harnessing delta-radiomics of the optic chiasm from MRI

Abstract

Objectives: To investigate whether morphological changes after surgery and delta-radiomics of the optic chiasm obtained from routine MRI could help predict postoperative visual recovery of pituitary adenoma patients.

Methods: A total of 130 pituitary adenoma patients were retrospectively enrolled and divided into the recovery group (n = 87) and non-recovery group (n = 43) according to visual outcome 1 year after endoscopic endonasal transsphenoidal surgery. Morphological parameters of the optic chiasm were measured preoperatively and postoperatively, including chiasmal thickness, deformed angle, and suprasellar extension. Delta-radiomics of the optic chiasm were calculated based on features extracted from preoperative and postoperative coronal T2-weighted images, followed by machine learning modeling using least absolute shrinkage and selection operator wrapped with support vector machine through fivefold cross-validation in the development set. The delta-radiomic model was independently evaluated in the test set, and compared with the combined model that incorporated delta-radiomics, significant clinical and morphological parameters.

Results: Postoperative morphological changes of the optic chiasm could not significantly be used as predictors for the visual outcome. In contrast, the delta-radiomics model represented good performances in predicting visual recovery, with an AUC of 0.821 in the development set and 0.811 in the independent test set. Moreover, the combined model that incorporated age and delta-radiomics features of the optic chiasm achieved the highest AUC of 0.841 and 0.840 in the development set and independent test set, respectively.

Conclusions: Our proposed machine learning models based on delta-radiomics of the optic chiasm can be used to predict postoperative visual recovery of pituitary adenoma patients.

Clinical relevance statement: Our delta-radiomics-based models from MRI enable accurate visual recovery predictions in pituitary adenoma patients who underwent endoscopic endonasal transsphenoidal surgery, facilitating better clinical decision-making and ultimately improving patient outcomes.

Key points: • Prediction of the postoperative visual outcome for pituitary adenoma patients is important but challenging. • Delta-radiomics of the optic chiasm after surgical decompression represented better prognostic performances compared with its morphological changes. • The proposed machine learning models can serve as novel approaches to predict visual recovery for pituitary adenoma patients in clinical practice.

Keywords: Machine learning; Magnetic resonance imaging; Optic chiasm; Pituitary adenoma.

Fig. 1

Flow chart of patient enrollment in this study

Fig. 2

The example of measuring morphological parameters of the optic chiasm on coronal T2-weighted MR images, including chiasmal thickness, chiasmal deformed angle, and chiasmal suprasellar extension

Fig. 3

The workflow of developing machine learning models based on delta-radiomics of the optic chiasm to predict the postoperative visual outcomes of pituitary adenoma patients

Fig. 4

Preoperative and postoperative morphological parameters of the optic chiasm in the recovery group and non-recovery group, including chiasmal thickness (A), chiasmal deformed angle (B), and chiasmal suprasellar extension (C). All three morphological parameters of the optic chiasm showed significant changes after surgical decompression, while neither preoperative nor postoperative morphological parameters showed significant differences between the recovery and non-recovery group. ***: p < 0.001; **: p < 0.01; ns: not significant

Fig. 5

Receiver operating characteristic curves of different models in predicting the postoperative visual outcome of pituitary adenoma patients in the development set (A) and independent test set (B). The delta-radiomics model and combined model represented higher AUC than the age in both the development and independent test set

Fig. 6

Calibration plots and decision curves of different models in the development set and independent test set. A, B The calibration plots illustrated that the delta-radiomics model and combined model showed good calibration with a closer fit to the diagonal dashed line that represents an ideal evaluation by a perfect model; C, D The decision curves illustrated that our models were clinically available. The black solid line represents all patients with visual recovery, while the black dashed line represents all patients without visual recovery