Yanghua Fan1, Yichao Li2, Xinjie Bao1, Huijuan Zhu3, Lin Lu3, Yong Yao1, Yansheng Li2, Mingliang Su2, Feng Feng4, Shanshan Feng1, Ming Feng1, Renzhi Wang1. 1. Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. 2. DHC Software Co. Ltd, Beijing, China. 3. Department of Endocrinology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. 4. Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
Abstract
CONTEXT: Postoperative hypercortisolemia mandates further therapy in patients with Cushing's disease (CD). Delayed remission (DR) is defined as not achieving postoperative immediate remission (IR), but having spontaneous remission during long-term follow-up. OBJECTIVE: We aimed to develop and validate machine learning (ML) models for predicting DR in non-IR patients with CD. METHODS: We enrolled 201 CD patients, and randomly divided them into training and test datasets. We then used the recursive feature elimination (RFE) algorithm to select features and applied 5 ML algorithms to construct DR prediction models. We used permutation importance and local interpretable model-agnostic explanation (LIME) algorithms to determine the importance of the selected features and interpret the ML models. RESULTS: Eighty-eight (43.8%) of the 201 CD patients met the criteria for DR. Overall, patients who were younger, had a low body mass index, a Knosp grade of III-IV, and a tumor not found by pathological examination tended to achieve a lower rate of DR. After RFE feature selection, the Adaboost model, which comprised 18 features, had the greatest discriminatory ability, and its predictive ability was significantly better than using Knosp grading and postoperative immediate morning serum cortisol (PoC). The results obtained from permutation importance and LIME algorithms showed that preoperative 24-hour urine free cortisol, PoC, and age were the most important features, and showed the reliability and clinical practicability of the Adaboost model in DC prediction. CONCLUSIONS: Machine learning-based models could serve as an effective noninvasive approach to predicting DR, and could aid in determining individual treatment and follow-up strategies for CD patients.
CONTEXT: Postoperative hypercortisolemia mandates further therapy in patients with Cushing's disease (CD). Delayed remission (DR) is defined as not achieving postoperative immediate remission (IR), but having spontaneous remission during long-term follow-up. OBJECTIVE: We aimed to develop and validate machine learning (ML) models for predicting DR in non-IR patients with CD. METHODS: We enrolled 201 CD patients, and randomly divided them into training and test datasets. We then used the recursive feature elimination (RFE) algorithm to select features and applied 5 ML algorithms to construct DR prediction models. We used permutation importance and local interpretable model-agnostic explanation (LIME) algorithms to determine the importance of the selected features and interpret the ML models. RESULTS: Eighty-eight (43.8%) of the 201 CD patients met the criteria for DR. Overall, patients who were younger, had a low body mass index, a Knosp grade of III-IV, and a tumor not found by pathological examination tended to achieve a lower rate of DR. After RFE feature selection, the Adaboost model, which comprised 18 features, had the greatest discriminatory ability, and its predictive ability was significantly better than using Knosp grading and postoperative immediate morning serum cortisol (PoC). The results obtained from permutation importance and LIME algorithms showed that preoperative 24-hour urine free cortisol, PoC, and age were the most important features, and showed the reliability and clinical practicability of the Adaboost model in DC prediction. CONCLUSIONS: Machine learning-based models could serve as an effective noninvasive approach to predicting DR, and could aid in determining individual treatment and follow-up strategies for CD patients.
Authors: Maria Fleseriu; Richard Auchus; Irina Bancos; Anat Ben-Shlomo; Jerome Bertherat; Nienke R Biermasz; Cesar L Boguszewski; Marcello D Bronstein; Michael Buchfelder; John D Carmichael; Felipe F Casanueva; Frederic Castinetti; Philippe Chanson; James Findling; Mônica Gadelha; Eliza B Geer; Andrea Giustina; Ashley Grossman; Mark Gurnell; Ken Ho; Adriana G Ioachimescu; Ursula B Kaiser; Niki Karavitaki; Laurence Katznelson; Daniel F Kelly; André Lacroix; Ann McCormack; Shlomo Melmed; Mark Molitch; Pietro Mortini; John Newell-Price; Lynnette Nieman; Alberto M Pereira; Stephan Petersenn; Rosario Pivonello; Hershel Raff; Martin Reincke; Roberto Salvatori; Carla Scaroni; Ilan Shimon; Constantine A Stratakis; Brooke Swearingen; Antoine Tabarin; Yutaka Takahashi; Marily Theodoropoulou; Stylianos Tsagarakis; Elena Valassi; Elena V Varlamov; Greisa Vila; John Wass; Susan M Webb; Maria C Zatelli; Beverly M K Biller Journal: Lancet Diabetes Endocrinol Date: 2021-10-20 Impact factor: 32.069