Matteo Zoli1,2, Victor E Staartjes3,4, Federica Guaraldi1,2, Filippo Friso1, Arianna Rustici5,6, Sofia Asioli1,2,7, Giacomo Sollini1,8, Ernesto Pasquini1,8, Luca Regli3, Carlo Serra3, Diego Mazzatenta1,2. 1. 1Pituitary Unit, Center for the Diagnosis and Treatment of Hypothalamic-Pituitary Diseases, IRCCS Institute of Neurological Sciences of Bologna. 2. 2Department of Biomedical and Motor Sciences (DIBINEM), University of Bologna, Italy. 3. 3Department of Neurosurgery, Clinical Neuroscience Center, University Hospital of Zurich, University of Zurich, Switzerland. 4. 4Neurosurgery, Amsterdam Movement Sciences, Amsterdam UMC, Vrije Universiteit Amsterdam, The Netherlands. 5. 5Department of Neuroradiology, IRCCS Istitute of Neurological Sciences of Bologna. 6. 6Department of Experimental, Diagnostic and Specialty Medicine (DIMES), University of Bologna. 7. 7Section of Anatomic Pathology 'M. Malpighi' at Bellaria Hospital, Bologna; and. 8. 8ENT Department, Bellaria Hospital, Bologna, Italy.
Abstract
OBJECTIVE: Machine learning (ML) is an innovative method to analyze large and complex data sets. The aim of this study was to evaluate the use of ML to identify predictors of early postsurgical and long-term outcomes in patients treated for Cushing disease (CD). METHODS: All consecutive patients in our center who underwent surgery for CD through the endoscopic endonasal approach were retrospectively reviewed. Study endpoints were gross-tumor removal (GTR), postsurgical remission, and long-term control of disease. Several demographic, radiological, and histological factors were assessed as potential predictors. For ML-based modeling, data were randomly divided into 2 sets with an 80% to 20% ratio for bootstrapped training and testing, respectively. Several algorithms were tested and tuned for the area under the curve (AUC). RESULTS: The study included 151 patients. GTR was achieved in 137 patients (91%), and postsurgical hypersecretion remission was achieved in 133 patients (88%). At last follow-up, 116 patients (77%) were still in remission after surgery and in 21 patients (14%), CD was controlled with complementary treatment (overall, of 131 cases, 87% were under control at follow-up). At internal validation, the endpoints were predicted with AUCs of 0.81-1.00, accuracy of 81%-100%, and Brier scores of 0.035-0.151. Tumor size and invasiveness and histological confirmation of adrenocorticotropic hormone (ACTH)-secreting cells were the main predictors for the 3 endpoints of interest. CONCLUSIONS: ML algorithms were used to train and internally validate robust models for all the endpoints, giving accurate outcome predictions in CD cases. This analytical method seems promising for potentially improving future patient care and counseling; however, careful clinical interpretation of the results remains necessary before any clinical adoption of ML. Moreover, further studies and increased sample sizes are definitely required before the widespread adoption of ML to the study of CD.
OBJECTIVE: Machine learning (ML) is an innovative method to analyze large and complex data sets. The aim of this study was to evaluate the use of ML to identify predictors of early postsurgical and long-term outcomes in patients treated for Cushing disease (CD). METHODS: All consecutive patients in our center who underwent surgery for CD through the endoscopic endonasal approach were retrospectively reviewed. Study endpoints were gross-tumor removal (GTR), postsurgical remission, and long-term control of disease. Several demographic, radiological, and histological factors were assessed as potential predictors. For ML-based modeling, data were randomly divided into 2 sets with an 80% to 20% ratio for bootstrapped training and testing, respectively. Several algorithms were tested and tuned for the area under the curve (AUC). RESULTS: The study included 151 patients. GTR was achieved in 137 patients (91%), and postsurgical hypersecretion remission was achieved in 133 patients (88%). At last follow-up, 116 patients (77%) were still in remission after surgery and in 21 patients (14%), CD was controlled with complementary treatment (overall, of 131 cases, 87% were under control at follow-up). At internal validation, the endpoints were predicted with AUCs of 0.81-1.00, accuracy of 81%-100%, and Brier scores of 0.035-0.151. Tumor size and invasiveness and histological confirmation of adrenocorticotropic hormone (ACTH)-secreting cells were the main predictors for the 3 endpoints of interest. CONCLUSIONS: ML algorithms were used to train and internally validate robust models for all the endpoints, giving accurate outcome predictions in CD cases. This analytical method seems promising for potentially improving future patient care and counseling; however, careful clinical interpretation of the results remains necessary before any clinical adoption of ML. Moreover, further studies and increased sample sizes are definitely required before the widespread adoption of ML to the study of CD.
Entities:
Keywords:
ACTH = adrenocorticotropic hormone; ACTH-secreting tumor; AUC = area under the curve; CD = Cushing disease; CS = cavernous sinus; Cushing disease; DI = diabetes insipidus; EEA = endoscopic endonasal approach; GBM = gradient boosting machine; GLM = generalized linear model; GTR = gross-tumor removal; IPSS = inferior petrosal sinus sampling; KNN = k-nearest neighbor; ML = machine learning; NPV = negative predictive value; PAS = periodic acid Schiff; PPV = positive predictive value; RF = random forest; ROC = receiver operating characteristic; SF-1 = steroidogenic factor–1; SVM = support vector machine; endoscopic endonasal surgery; machine learning; outcome; predictors
Authors: Quinlan D Buchlak; Nazanin Esmaili; Christine Bennett; Yi Yuen Wang; James King; Tony Goldschlager Journal: PLoS One Date: 2022-07-27 Impact factor: 3.752